maryjane
/
syncevolution
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
syncevolution/src/syncevo/Cmdline.cpp

3280 lines
140 KiB
C++
Raw Normal View History

next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
/*
copyright updated update-copyright.sh can be used to add copyright remarks for the current year. It finds the authors who made a change in each file and adds/updates their copyright remark. Intel employees are grouped under "Intel Corporation".
2009-03-25 15:21:04 +01:00
* Copyright (C) 2008-2009 Patrick Ohly <patrick.ohly@gmx.de>
* Copyright (C) 2009 Intel Corporation
license: changed to LGPL v2.1 + v3 As with the previous change from GPL to LGPL v2.1, this is covered by copyright ownership and/or the Funambol contributor agreement. The motivation for adding v3 is greater flexibility regarding reusing the code in other places and relicensing it. Also, now the license and disclaimer are mentioned at the start of the source files. That is not strictly necessary, but more explicit.
2009-04-30 18:14:03 +02:00
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) version 3.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
* 02110-1301 USA
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
*/
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
#include <syncevo/Cmdline.h>
#include <syncevo/FilterConfigNode.h>
#include <syncevo/VolatileConfigNode.h>
#include <syncevo/SyncSource.h>
#include <syncevo/SyncContext.h>
#include <syncevo/util.h>
introduced "namespace SyncEvo" consistently Added syncevo/declarations.h, which has This is now used for all SyncEvolution source files, except for the GTK UI, which is written in plain C. In the library it helps to avoid name clashes. The reason for using defines instead of spelling out "namespace SyncEvo" is twofold: 1. if that should ever become necessary, it is easier to rename the namespace via configure options by changing the define 2. editors don't indent the whole file content
2009-10-02 17:23:53 +02:00
#include "test.h"
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
#include <unistd.h>
#include <errno.h>
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
#include <fstream>
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
#include <iostream>
log handling: save start and end time in log dir, retrieve as part of SyncReport EvolutionSyncClient::readSessionInfo() returns this information. All other fields in such a sync report are not yet restored. The SyncReport available after a sync does not yet include start and end times; this still needs to be added.
2009-04-15 21:03:26 +02:00
#include <iomanip>
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
#include <sstream>
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
#include <memory>
#include <set>
command line: check for invalid peer properties (MB #8048) When using --print-config or --configure with a context as config name (like @default), then per-peer properties do not have any effect. This might be confusing for users, so the command line should better report this as error and refuse to execute the operation. As usual, "type" is a bit special because it is both shared and per-peer.
2010-03-03 16:20:53 +01:00
#include <list>
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
#include <algorithm>
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
using namespace std;
#include <boost/shared_ptr.hpp>
Debian Etch: fixes some compilation issues With some tweaks the autotools on Debian Etch are good enough to compile the code. However, distributions of SyncEvolution should better be built with recent versions of the tools and then get compiled on Debian Etch. We need BOOST >= 1.34 because of BOOST_FOREACH(). The configure script now checks for this. Debian Etch can get 1.35 from www.backports.org. That version of BOOST shows that in some files <boost/algorithm/string/join.hpp> was not explicitly included.
2009-06-10 17:28:45 +02:00
#include <boost/algorithm/string/join.hpp>
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
#include <boost/algorithm/string/split.hpp>
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
#include <boost/algorithm/string.hpp>
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
#include <boost/foreach.hpp>
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
#include <fstream>
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
#include <syncevo/declarations.h>
introduced "namespace SyncEvo" consistently Added syncevo/declarations.h, which has This is now used for all SyncEvolution source files, except for the GTK UI, which is written in plain C. In the library it helps to avoid name clashes. The reason for using defines instead of spelling out "namespace SyncEvo" is twofold: 1. if that should ever become necessary, it is easier to rename the namespace via configure options by changing the define 2. editors don't indent the whole file content
2009-10-02 17:23:53 +02:00
SE_BEGIN_CXX
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
Cmdline::Cmdline(int argc, const char * const * argv, ostream &out, ostream &err) :
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_argc(argc),
m_argv(argv),
m_out(out),
m_err(err),
more classes renamed EvolutionSyncConfig => SyncConfig SyncEvolutionException => Exception EvolutionUnref* => Unref*
2009-10-06 17:22:47 +02:00
m_validSyncProps(SyncConfig::getRegistry()),
redesigned SyncSource base class + API The main motivation for this change is that it allows the implementor of a backend to choose the implementations for the different aspects of a datasource (change tracking, item import/export, logging, ...) independently of each other. For example, change tracking via revision strings can now be combined with exchanging data with the Synthesis engine via a single string (the traditional method in SyncEvolution) and with direct access to the Synthesis field list (now possible for the first time). The new backend API is based on the concept of providing implementations for certain functionality via function objects instead of implementing certain virtual methods. The advantage is that implementors can define their own, custom interfaces and mix and match implementations of the different groups of functionality. Logging (see SyncSourceLogging in a later commit) can be done by wrapping some arbitrary other item import/export function objects (decorator design pattern). The class hierarchy is now this: - SyncSourceBase: interface for common utility code, all other classes are derived from it and thus can use that code - SyncSource: base class which implements SyncSourceBase and hooks a datasource into the SyncEvolution core; its "struct Operations" holds the function objects which can be implemented in different ways - TestingSyncSource: combines some of the following classes into an interface that is expected by the client-test program; backends only have to derive from (and implement this) if they want to use the automated testing - TrackingSyncSource: provides the same functionality as before (change tracking via revision strings, item import/export as string) in a single interface; the description of the pure virtual methods are duplicated so that developers can go through this class and find everything they need to know to implement it The following classes contain the code that was previously found in the EvolutionSyncSource base class. Implementors can derive from them and call the init() methods to inherit and activate the functionality: - SyncSourceSession: binds Synthesis session callbacks to virtual methods beginSync(), endSync() - SyncSourceChanges: implements Synthesis item tracking callbacks with set of LUIDs that the user of the class has to fill - SyncSourceDelete: binds Synthesis delete callback to virtual method - SyncSourceRaw: read and write items in the backends format, used for testing and backup/restore - SyncSourceSerialize: exchanges items with Synthesis engine using a string representation of the data; this is how EvolutionSyncSource has traditionally worked, so much of the same virtual methods are now in this class - SyncSourceRevisions: utility class which does change tracking via some kind of "revision" string which changes each time an item is modified; this code was previously in the TrackingSyncSource
2009-08-25 09:27:46 +02:00
m_validSourceProps(SyncSourceConfig::getRegistry())
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
{}
Cmdline: move cmdline to dbus server (MB#5043) Implement cmdline with support of dbus server. To enable cmdline with dbus server, use the option '--use-daemon yes/no' in case that you enable dbus service when configuration. In a typical scenario, a new session is created for the purpose of execution of arguments. It is scheduled with other sessions but with a highest priority. Once it becomes active, command line call 'Session.Execute', a newly added method to execute command line arguments. The config name of a session should be known for dbus clients like command line. A new property 'configName' is added in the properties when calling 'Session.GetConfig'. CTRL-C handling are processed once executing a real sync to dbus server. It is mapped to invoke 'Session.Suspend' and 'Session.Abort'. The meaning of '--enable-dbus-service' is expanded accordingly. '--status' without server means printing all running session in the dbus server. '--monitor' could accept an optional config name. If one is given, only attach to a session of that config, otherwise print an error. If none is given, pick the first.
2010-03-15 02:33:43 +01:00
Cmdline::Cmdline(const vector<string> &args, ostream &out, ostream &err) :
m_args(args),
m_out(out),
m_err(err),
m_validSyncProps(SyncConfig::getRegistry()),
m_validSourceProps(SyncSourceConfig::getRegistry())
{
m_argc = args.size();
m_argvArray.reset(new const char *[args.size()]);
for(int i = 0; i < m_argc; i++) {
m_argvArray[i] = m_args[i].c_str();
}
m_argv = m_argvArray.get();
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
bool Cmdline::parse()
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
{
command line + D-Bus: pass absolute paths to dbus server (MB#10461) Convert relative paths into absolute paths before passing arguments to dbus server because command line and dbus may have different working directory. Leave this kind of task in Command line for it knows the meaning of parameters and convert them if necessary. Add a new function 'relToAbs' to convert a relative path to a canonicalized absolute path.
2010-04-06 10:38:24 +02:00
vector<string> parsed;
return parse(parsed);
}
bool Cmdline::parse(vector<string> &parsed)
{
parsed.clear();
command line + D-Bus: fix for "pass absolute paths to dbus server" (MB #10461) argv[0] is skipped by Cmdline::parse(), therefore wasn't added to the updated argument list for the D-Bus, which then basically ignored the first command line parameter.
2010-04-07 18:21:54 +02:00
if (m_argc) {
parsed.push_back(m_argv[0]);
}
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
m_delimiter = "\n\n";
command line + D-Bus: fix for "pass absolute paths to dbus server" (MB #10461) argv[0] is skipped by Cmdline::parse(), therefore wasn't added to the updated argument list for the D-Bus, which then basically ignored the first command line parameter.
2010-04-07 18:21:54 +02:00
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
// All command line options which ask for a specific operation,
// like --restore, --print-config, ... Used to detect conflicting
// operations.
vector<string> operations;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
int opt = 1;
command line + daemon: usability improvements (MB #5043) This patch changes what is shown to the user and how the user interacts with the command line. Details below. "--use-daemon [yes/no]" implies that yes/no is optional (square brackets!), with "yes" being the option that could be expected for a plain "--use-daemon" parameter. But the implementation always expects a "yes" or "no" parameter. The original format suggested was "--use-daemon[=yes/no]" This patch switches to that format, changes --use-daemon into --daemon (to be consistent with --keyring) and enables the same format for --keyring. Although not documented, 0/f/false and 1/t/true are also accepted. Because the value becomes part of the parameter, m_argc checks had to be adapted. The documentation for "--use-daemon" was inserted in the middle of the "--keyring documentation". "Parameter not set" has to be available to the Cmdline caller in the command line too, in addition to true/false. This was done originally with a string which is empty, "yes" or "no". Using a tri-state Cmdline::Bool class makes this a bit more explicit and removes the ambiguity around what useDaemon() returns (values weren't documented). When running without --daemon and no daemon available, the first lines of output were: ERROR: org.<cryptic error> [INFO] itodo20: inactive .... => The command line should fall back *silently* to running in-process. => Error messages should be formatted and logged as such, using SE_LOG_ERROR(). Old code might not have done that and we need to preserve that for compatibility with frontends, but new code should use [ERROR] as output. => D-Bus error messages (as in the case above) must have some user (and developer!) comprehensible explanation what went wrong. The D-Bus error class itself is no enough. Although I haven't tested it, I suspect that the code also would have re-run the operation in-process after the D-Bus server already executed it and failed. I rewrote this so that a check for "daemon available" without error messages is done first before committing to using the daemon. Once that decision is made, the command line will not fall back to in-process execution. Rewrote several error messages. Telling a user of a distro's binary to "re-configure" is misleading (he didn't configure himself). "can't" => "cannot", punctuation changes. Not sure whether is always an improvement, comments welcome. Comment on coding style: I've used "if ()" instead of "if()" because that is the GNU recommendation.
2010-03-26 18:47:41 +01:00
bool ok;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
while (opt < m_argc) {
command line + D-Bus: pass absolute paths to dbus server (MB#10461) Convert relative paths into absolute paths before passing arguments to dbus server because command line and dbus may have different working directory. Leave this kind of task in Command line for it knows the meaning of parameters and convert them if necessary. Add a new function 'relToAbs' to convert a relative path to a canonicalized absolute path.
2010-04-06 10:38:24 +02:00
parsed.push_back(m_argv[opt]);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
if (m_argv[opt][0] != '-') {
break;
}
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
if (boost::iequals(m_argv[opt], "--sync") ||
boost::iequals(m_argv[opt], "-s")) {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
opt++;
string param;
string cmdopt(m_argv[opt - 1]);
minor fixes for command line handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@525 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 13:11:39 +01:00
if (!parseProp(m_validSourceProps, m_sourceProps,
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_argv[opt - 1], opt == m_argc ? NULL : m_argv[opt],
redesigned SyncSource base class + API The main motivation for this change is that it allows the implementor of a backend to choose the implementations for the different aspects of a datasource (change tracking, item import/export, logging, ...) independently of each other. For example, change tracking via revision strings can now be combined with exchanging data with the Synthesis engine via a single string (the traditional method in SyncEvolution) and with direct access to the Synthesis field list (now possible for the first time). The new backend API is based on the concept of providing implementations for certain functionality via function objects instead of implementing certain virtual methods. The advantage is that implementors can define their own, custom interfaces and mix and match implementations of the different groups of functionality. Logging (see SyncSourceLogging in a later commit) can be done by wrapping some arbitrary other item import/export function objects (decorator design pattern). The class hierarchy is now this: - SyncSourceBase: interface for common utility code, all other classes are derived from it and thus can use that code - SyncSource: base class which implements SyncSourceBase and hooks a datasource into the SyncEvolution core; its "struct Operations" holds the function objects which can be implemented in different ways - TestingSyncSource: combines some of the following classes into an interface that is expected by the client-test program; backends only have to derive from (and implement this) if they want to use the automated testing - TrackingSyncSource: provides the same functionality as before (change tracking via revision strings, item import/export as string) in a single interface; the description of the pure virtual methods are duplicated so that developers can go through this class and find everything they need to know to implement it The following classes contain the code that was previously found in the EvolutionSyncSource base class. Implementors can derive from them and call the init() methods to inherit and activate the functionality: - SyncSourceSession: binds Synthesis session callbacks to virtual methods beginSync(), endSync() - SyncSourceChanges: implements Synthesis item tracking callbacks with set of LUIDs that the user of the class has to fill - SyncSourceDelete: binds Synthesis delete callback to virtual method - SyncSourceRaw: read and write items in the backends format, used for testing and backup/restore - SyncSourceSerialize: exchanges items with Synthesis engine using a string representation of the data; this is how EvolutionSyncSource has traditionally worked, so much of the same virtual methods are now in this class - SyncSourceRevisions: utility class which does change tracking via some kind of "revision" string which changes each time an item is modified; this code was previously in the TrackingSyncSource
2009-08-25 09:27:46 +02:00
SyncSourceConfig::m_sourcePropSync.getName().c_str())) {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
return false;
}
command line: fix a regression checked by nightly There is a regression caused by a string exception "basic_string::_S_construct NULL not valid". This is because I push back a non-existing string into vector when collecting converted command line arguments.
2010-04-09 05:24:03 +02:00
parsed.push_back(m_argv[opt]);
--sync implies --run, to be compatible with old command lines git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@706 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-08-03 22:48:45 +02:00
// disable requirement to add --run explicitly in order to
// be compatible with traditional command lines
m_run = true;
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
} else if(boost::iequals(m_argv[opt], "--sync-property") ||
boost::iequals(m_argv[opt], "-y")) {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
opt++;
if (!parseProp(m_validSyncProps, m_syncProps,
m_argv[opt - 1], opt == m_argc ? NULL : m_argv[opt])) {
return false;
}
command line: fix a regression checked by nightly There is a regression caused by a string exception "basic_string::_S_construct NULL not valid". This is because I push back a non-existing string into vector when collecting converted command line arguments.
2010-04-09 05:24:03 +02:00
parsed.push_back(m_argv[opt]);
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
} else if(boost::iequals(m_argv[opt], "--source-property") ||
boost::iequals(m_argv[opt], "-z")) {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
opt++;
if (!parseProp(m_validSourceProps, m_sourceProps,
m_argv[opt - 1], opt == m_argc ? NULL : m_argv[opt])) {
return false;
}
command line: fix a regression checked by nightly There is a regression caused by a string exception "basic_string::_S_construct NULL not valid". This is because I push back a non-existing string into vector when collecting converted command line arguments.
2010-04-09 05:24:03 +02:00
parsed.push_back(m_argv[opt]);
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
}else if(boost::iequals(m_argv[opt], "--template") ||
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
boost::iequals(m_argv[opt], "-l")) {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
opt++;
if (opt >= m_argc) {
usage(true, string("missing parameter for ") + cmdOpt(m_argv[opt - 1]));
return false;
}
command line: fix a regression checked by nightly There is a regression caused by a string exception "basic_string::_S_construct NULL not valid". This is because I push back a non-existing string into vector when collecting converted command line arguments.
2010-04-09 05:24:03 +02:00
parsed.push_back(m_argv[opt]);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_template = m_argv[opt];
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
m_configure = true;
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
string temp = boost::trim_copy (m_template);
if (temp.find ("?") == 0){
m_printTemplates = true;
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
m_dontrun = true;
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
m_template = temp.substr (1);
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
}
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
} else if(boost::iequals(m_argv[opt], "--print-servers") ||
boost::iequals(m_argv[opt], "--print-peers") ||
boost::iequals(m_argv[opt], "--print-configs")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_printServers = true;
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
} else if(boost::iequals(m_argv[opt], "--print-config") ||
boost::iequals(m_argv[opt], "-p")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_printConfig = true;
log handling: added access to previous sync sessions EvolutionSyncClient::getSessions() returns the full pathnames of the log dirs of the previous sync sessions. --print-sessions shows those on the command line. Right now this feature is only useful to remove these directories or (for the user) to find backup files. Further information has to written to and extracted from them to make --print-sessions and its graphical pendant more useful.
2009-04-15 15:58:05 +02:00
} else if(boost::iequals(m_argv[opt], "--print-sessions")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
log handling: added access to previous sync sessions EvolutionSyncClient::getSessions() returns the full pathnames of the log dirs of the previous sync sessions. --print-sessions shows those on the command line. Right now this feature is only useful to remove these directories or (for the user) to find backup files. Further information has to written to and extracted from them to make --print-sessions and its graphical pendant more useful.
2009-04-15 15:58:05 +02:00
m_printSessions = true;
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
} else if(boost::iequals(m_argv[opt], "--configure") ||
boost::iequals(m_argv[opt], "-c")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_configure = true;
config handling: added removing of a configuration Only files which are known to belong to SyncEvolution are deleted. This list is hard-coded in FileConfigTree.cpp and includes the hidden .synthesis meta files in addition to the normal config files. The --remove command line option exposes this new feature. It removes the configuration *without* asking for confirmation. Log dirs are not removed yet. This needs to be done via a different API which will be added as part of access to old session information.
2009-04-15 14:02:09 +02:00
} else if(boost::iequals(m_argv[opt], "--remove")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
config handling: added removing of a configuration Only files which are known to belong to SyncEvolution are deleted. This list is hard-coded in FileConfigTree.cpp and includes the hidden .synthesis meta files in addition to the normal config files. The --remove command line option exposes this new feature. It removes the configuration *without* asking for confirmation. Log dirs are not removed yet. This needs to be done via a different API which will be added as part of access to old session information.
2009-04-15 14:02:09 +02:00
m_remove = true;
improved command line and its documented More explicit description of the different operations. Added --run|-r switch: if properties are on the command line, then an operation must be specified explicitly, with --run standing for a sync run. Otherwise a sync might accidentally be done when a configuration change was intented, but the --configure switch was forgotten. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@696 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-08-03 15:00:51 +02:00
} else if(boost::iequals(m_argv[opt], "--run") ||
boost::iequals(m_argv[opt], "-r")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
improved command line and its documented More explicit description of the different operations. Added --run|-r switch: if properties are on the command line, then an operation must be specified explicitly, with --run standing for a sync run. Otherwise a sync might accidentally be done when a configuration change was intented, but the --configure switch was forgotten. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@696 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-08-03 15:00:51 +02:00
m_run = true;
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
} else if(boost::iequals(m_argv[opt], "--restore")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
opt++;
if (opt >= m_argc) {
usage(true, string("missing parameter for ") + cmdOpt(m_argv[opt - 1]));
return false;
}
m_restore = m_argv[opt];
if (m_restore.empty()) {
usage(true, string("missing parameter for ") + cmdOpt(m_argv[opt - 1]));
return false;
}
command line + D-Bus: pass absolute paths to dbus server (MB#10461) Convert relative paths into absolute paths before passing arguments to dbus server because command line and dbus may have different working directory. Leave this kind of task in Command line for it knows the meaning of parameters and convert them if necessary. Add a new function 'relToAbs' to convert a relative path to a canonicalized absolute path.
2010-04-06 10:38:24 +02:00
//if can't convert it successfully, it's an invalid path
if (!relToAbs(m_restore)) {
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
usage(true, string("parameter '") + m_restore + "' for " + cmdOpt(m_argv[opt - 1]) + " must be log directory");
return false;
}
command line + D-Bus: pass absolute paths to dbus server (MB#10461) Convert relative paths into absolute paths before passing arguments to dbus server because command line and dbus may have different working directory. Leave this kind of task in Command line for it knows the meaning of parameters and convert them if necessary. Add a new function 'relToAbs' to convert a relative path to a canonicalized absolute path.
2010-04-06 10:38:24 +02:00
parsed.push_back(m_restore);
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
} else if(boost::iequals(m_argv[opt], "--before")) {
m_before = true;
} else if(boost::iequals(m_argv[opt], "--after")) {
m_after = true;
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
} else if (boost::iequals(m_argv[opt], "--print-items")) {
operations.push_back(m_argv[opt]);
m_printItems = m_accessItems = true;
} else if ((boost::iequals(m_argv[opt], "--export") && (m_export = true)) ||
(boost::iequals(m_argv[opt], "--import") && (m_import = true)) ||
(boost::iequals(m_argv[opt], "--update") && (m_update = true))) {
operations.push_back(m_argv[opt]);
m_accessItems = true;
opt++;
if (opt >= m_argc || !m_argv[opt][0]) {
usage(true, string("missing parameter for ") + cmdOpt(m_argv[opt - 1]));
return false;
}
m_itemPath = m_argv[opt];
if (m_itemPath != "-") {
string dir, file;
splitPath(m_itemPath, dir, file);
if (dir.empty()) {
dir = ".";
}
if (!relToAbs(dir)) {
SyncContext::throwError(dir, errno);
}
m_itemPath = dir + "/" + file;
}
parsed.push_back(m_itemPath);
} else if (boost::iequals(m_argv[opt], "--delimiter")) {
opt++;
if (opt >= m_argc) {
usage(true, string("missing parameter for ") + cmdOpt(m_argv[opt - 1]));
return false;
}
m_delimiter = m_argv[opt];
parsed.push_back(m_delimiter);
command line: added --delete-items This complements --import/update/export/list-items. --remove was already taken, for deleting a configuration. Because the removal is done via the normal m_deleteItem operation, we inherit the normal SyncML behavior (logging, no error return code for non-existent items). It would have been nice to let the user choose how to handle that situation (like --force in rm), but that can't be done without SyncSource API changes.
2010-06-11 18:43:01 +02:00
} else if (boost::iequals(m_argv[opt], "--delete-items")) {
operations.push_back(m_argv[opt]);
m_deleteItems = m_accessItems = true;
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
} else if(boost::iequals(m_argv[opt], "--dry-run")) {
m_dryrun = true;
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
} else if(boost::iequals(m_argv[opt], "--migrate")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_migrate = true;
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
} else if(boost::iequals(m_argv[opt], "--status") ||
boost::iequals(m_argv[opt], "-t")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_status = true;
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
} else if(boost::iequals(m_argv[opt], "--quiet") ||
boost::iequals(m_argv[opt], "-q")) {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_quiet = true;
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
} else if(boost::iequals(m_argv[opt], "--help") ||
boost::iequals(m_argv[opt], "-h")) {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_usage = true;
using boost::algorithm::string is okay from now on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@544 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 16:41:11 +01:00
} else if(boost::iequals(m_argv[opt], "--version")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_version = true;
command line + daemon: usability improvements (MB #5043) This patch changes what is shown to the user and how the user interacts with the command line. Details below. "--use-daemon [yes/no]" implies that yes/no is optional (square brackets!), with "yes" being the option that could be expected for a plain "--use-daemon" parameter. But the implementation always expects a "yes" or "no" parameter. The original format suggested was "--use-daemon[=yes/no]" This patch switches to that format, changes --use-daemon into --daemon (to be consistent with --keyring) and enables the same format for --keyring. Although not documented, 0/f/false and 1/t/true are also accepted. Because the value becomes part of the parameter, m_argc checks had to be adapted. The documentation for "--use-daemon" was inserted in the middle of the "--keyring documentation". "Parameter not set" has to be available to the Cmdline caller in the command line too, in addition to true/false. This was done originally with a string which is empty, "yes" or "no". Using a tri-state Cmdline::Bool class makes this a bit more explicit and removes the ambiguity around what useDaemon() returns (values weren't documented). When running without --daemon and no daemon available, the first lines of output were: ERROR: org.<cryptic error> [INFO] itodo20: inactive .... => The command line should fall back *silently* to running in-process. => Error messages should be formatted and logged as such, using SE_LOG_ERROR(). Old code might not have done that and we need to preserve that for compatibility with frontends, but new code should use [ERROR] as output. => D-Bus error messages (as in the case above) must have some user (and developer!) comprehensible explanation what went wrong. The D-Bus error class itself is no enough. Although I haven't tested it, I suspect that the code also would have re-run the operation in-process after the D-Bus server already executed it and failed. I rewrote this so that a check for "daemon available" without error messages is done first before committing to using the daemon. Once that decision is made, the command line will not fall back to in-process execution. Rewrote several error messages. Telling a user of a distro's binary to "re-configure" is misleading (he didn't configure himself). "can't" => "cannot", punctuation changes. Not sure whether is always an improvement, comments welcome. Comment on coding style: I've used "if ()" instead of "if()" because that is the GNU recommendation.
2010-03-26 18:47:41 +01:00
} else if (parseBool(opt, "--keyring", "-k", true, m_keyring, ok)) {
if (!ok) {
Cmdline: move cmdline to dbus server (MB#5043) Implement cmdline with support of dbus server. To enable cmdline with dbus server, use the option '--use-daemon yes/no' in case that you enable dbus service when configuration. In a typical scenario, a new session is created for the purpose of execution of arguments. It is scheduled with other sessions but with a highest priority. Once it becomes active, command line call 'Session.Execute', a newly added method to execute command line arguments. The config name of a session should be known for dbus clients like command line. A new property 'configName' is added in the properties when calling 'Session.GetConfig'. CTRL-C handling are processed once executing a real sync to dbus server. It is mapped to invoke 'Session.Suspend' and 'Session.Abort'. The meaning of '--enable-dbus-service' is expanded accordingly. '--status' without server means printing all running session in the dbus server. '--monitor' could accept an optional config name. If one is given, only attach to a session of that config, otherwise print an error. If none is given, pick the first.
2010-03-15 02:33:43 +01:00
return false;
}
command line + daemon: usability improvements (MB #5043) This patch changes what is shown to the user and how the user interacts with the command line. Details below. "--use-daemon [yes/no]" implies that yes/no is optional (square brackets!), with "yes" being the option that could be expected for a plain "--use-daemon" parameter. But the implementation always expects a "yes" or "no" parameter. The original format suggested was "--use-daemon[=yes/no]" This patch switches to that format, changes --use-daemon into --daemon (to be consistent with --keyring) and enables the same format for --keyring. Although not documented, 0/f/false and 1/t/true are also accepted. Because the value becomes part of the parameter, m_argc checks had to be adapted. The documentation for "--use-daemon" was inserted in the middle of the "--keyring documentation". "Parameter not set" has to be available to the Cmdline caller in the command line too, in addition to true/false. This was done originally with a string which is empty, "yes" or "no". Using a tri-state Cmdline::Bool class makes this a bit more explicit and removes the ambiguity around what useDaemon() returns (values weren't documented). When running without --daemon and no daemon available, the first lines of output were: ERROR: org.<cryptic error> [INFO] itodo20: inactive .... => The command line should fall back *silently* to running in-process. => Error messages should be formatted and logged as such, using SE_LOG_ERROR(). Old code might not have done that and we need to preserve that for compatibility with frontends, but new code should use [ERROR] as output. => D-Bus error messages (as in the case above) must have some user (and developer!) comprehensible explanation what went wrong. The D-Bus error class itself is no enough. Although I haven't tested it, I suspect that the code also would have re-run the operation in-process after the D-Bus server already executed it and failed. I rewrote this so that a check for "daemon available" without error messages is done first before committing to using the daemon. Once that decision is made, the command line will not fall back to in-process execution. Rewrote several error messages. Telling a user of a distro's binary to "re-configure" is misleading (he didn't configure himself). "can't" => "cannot", punctuation changes. Not sure whether is always an improvement, comments welcome. Comment on coding style: I've used "if ()" instead of "if()" because that is the GNU recommendation.
2010-03-26 18:47:41 +01:00
} else if (parseBool(opt, "--daemon", NULL, true, m_useDaemon, ok)) {
if (!ok) {
return false;
Cmdline: move cmdline to dbus server (MB#5043) Implement cmdline with support of dbus server. To enable cmdline with dbus server, use the option '--use-daemon yes/no' in case that you enable dbus service when configuration. In a typical scenario, a new session is created for the purpose of execution of arguments. It is scheduled with other sessions but with a highest priority. Once it becomes active, command line call 'Session.Execute', a newly added method to execute command line arguments. The config name of a session should be known for dbus clients like command line. A new property 'configName' is added in the properties when calling 'Session.GetConfig'. CTRL-C handling are processed once executing a real sync to dbus server. It is mapped to invoke 'Session.Suspend' and 'Session.Abort'. The meaning of '--enable-dbus-service' is expanded accordingly. '--status' without server means printing all running session in the dbus server. '--monitor' could accept an optional config name. If one is given, only attach to a session of that config, otherwise print an error. If none is given, pick the first.
2010-03-15 02:33:43 +01:00
}
} else if(boost::iequals(m_argv[opt], "--monitor")||
boost::iequals(m_argv[opt], "-m")) {
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
operations.push_back(m_argv[opt]);
Cmdline: move cmdline to dbus server (MB#5043) Implement cmdline with support of dbus server. To enable cmdline with dbus server, use the option '--use-daemon yes/no' in case that you enable dbus service when configuration. In a typical scenario, a new session is created for the purpose of execution of arguments. It is scheduled with other sessions but with a highest priority. Once it becomes active, command line call 'Session.Execute', a newly added method to execute command line arguments. The config name of a session should be known for dbus clients like command line. A new property 'configName' is added in the properties when calling 'Session.GetConfig'. CTRL-C handling are processed once executing a real sync to dbus server. It is mapped to invoke 'Session.Suspend' and 'Session.Abort'. The meaning of '--enable-dbus-service' is expanded accordingly. '--status' without server means printing all running session in the dbus server. '--monitor' could accept an optional config name. If one is given, only attach to a session of that config, otherwise print an error. If none is given, pick the first.
2010-03-15 02:33:43 +01:00
m_monitor = true;
minor fixes for command line handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@525 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 13:11:39 +01:00
} else {
usage(false, string(m_argv[opt]) + ": unknown parameter");
return false;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
opt++;
}
if (opt < m_argc) {
m_server = m_argv[opt++];
while (opt < m_argc) {
command line + D-Bus: pass absolute paths to dbus server (MB#10461) Convert relative paths into absolute paths before passing arguments to dbus server because command line and dbus may have different working directory. Leave this kind of task in Command line for it knows the meaning of parameters and convert them if necessary. Add a new function 'relToAbs' to convert a relative path to a canonicalized absolute path.
2010-04-06 10:38:24 +02:00
parsed.push_back(m_argv[opt]);
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
if (m_sources.empty() ||
!m_accessItems) {
m_sources.insert(m_argv[opt++]);
} else {
// first additional parameter was source, rest are luids
m_luids.push_back(m_argv[opt++]);
}
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
}
command line parsing: check for conflicting operations Previously, it was possible to list several different operations on the command line ("--version --configure --run ..."). SyncEvolution then picked one of them in (for users) unpredictable ways. Now all requested operations are gathered while parsing the command line and the tool aborts with an error message listing all of them if more than one is given.
2010-06-11 12:52:35 +02:00
// check whether we have conflicting operations requested by user
if (operations.size() > 1) {
usage(false, boost::join(operations, " ") + ": mutually exclusive operations");
return false;
}
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
// common sanity checking for item listing/import/export/update
if (m_accessItems) {
if (m_server.empty()) {
usage(false, operations[0] + ": needs configuration name");
return false;
}
if (m_sources.size() == 0) {
usage(false, operations[0] + ": needs source name");
return false;
}
if ((m_import || m_update) && m_dryrun) {
usage(false, operations[0] + ": --dry-run not supported");
return false;
}
}
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
return true;
}
command line + daemon: usability improvements (MB #5043) This patch changes what is shown to the user and how the user interacts with the command line. Details below. "--use-daemon [yes/no]" implies that yes/no is optional (square brackets!), with "yes" being the option that could be expected for a plain "--use-daemon" parameter. But the implementation always expects a "yes" or "no" parameter. The original format suggested was "--use-daemon[=yes/no]" This patch switches to that format, changes --use-daemon into --daemon (to be consistent with --keyring) and enables the same format for --keyring. Although not documented, 0/f/false and 1/t/true are also accepted. Because the value becomes part of the parameter, m_argc checks had to be adapted. The documentation for "--use-daemon" was inserted in the middle of the "--keyring documentation". "Parameter not set" has to be available to the Cmdline caller in the command line too, in addition to true/false. This was done originally with a string which is empty, "yes" or "no". Using a tri-state Cmdline::Bool class makes this a bit more explicit and removes the ambiguity around what useDaemon() returns (values weren't documented). When running without --daemon and no daemon available, the first lines of output were: ERROR: org.<cryptic error> [INFO] itodo20: inactive .... => The command line should fall back *silently* to running in-process. => Error messages should be formatted and logged as such, using SE_LOG_ERROR(). Old code might not have done that and we need to preserve that for compatibility with frontends, but new code should use [ERROR] as output. => D-Bus error messages (as in the case above) must have some user (and developer!) comprehensible explanation what went wrong. The D-Bus error class itself is no enough. Although I haven't tested it, I suspect that the code also would have re-run the operation in-process after the D-Bus server already executed it and failed. I rewrote this so that a check for "daemon available" without error messages is done first before committing to using the daemon. Once that decision is made, the command line will not fall back to in-process execution. Rewrote several error messages. Telling a user of a distro's binary to "re-configure" is misleading (he didn't configure himself). "can't" => "cannot", punctuation changes. Not sure whether is always an improvement, comments welcome. Comment on coding style: I've used "if ()" instead of "if()" because that is the GNU recommendation.
2010-03-26 18:47:41 +01:00
bool Cmdline::parseBool(int opt, const char *longName, const char *shortName,
bool def, Bool &value,
bool &ok)
{
string option = m_argv[opt];
string param;
size_t pos = option.find('=');
if (pos != option.npos) {
param = option.substr(pos + 1);
option.resize(pos);
}
if ((longName && boost::iequals(option, longName)) ||
(shortName && boost::iequals(option, shortName))) {
ok = true;
if (param.empty()) {
value = def;
} else if (boost::iequals(param, "t") ||
boost::iequals(param, "1") ||
boost::iequals(param, "true") ||
boost::iequals(param, "yes")) {
value = true;
} else if (boost::iequals(param, "f") ||
boost::iequals(param, "0") ||
boost::iequals(param, "false") ||
boost::iequals(param, "no")) {
value = false;
} else {
usage(true, string("parameter in '") + m_argv[opt] + "' must be 1/t/true/yes or 0/f/false/no");
ok = false;
}
// was our option
return true;
} else {
// keep searching for match
return false;
}
}
Cmdline: move cmdline to dbus server (MB#5043) Implement cmdline with support of dbus server. To enable cmdline with dbus server, use the option '--use-daemon yes/no' in case that you enable dbus service when configuration. In a typical scenario, a new session is created for the purpose of execution of arguments. It is scheduled with other sessions but with a highest priority. Once it becomes active, command line call 'Session.Execute', a newly added method to execute command line arguments. The config name of a session should be known for dbus clients like command line. A new property 'configName' is added in the properties when calling 'Session.GetConfig'. CTRL-C handling are processed once executing a real sync to dbus server. It is mapped to invoke 'Session.Suspend' and 'Session.Abort'. The meaning of '--enable-dbus-service' is expanded accordingly. '--status' without server means printing all running session in the dbus server. '--monitor' could accept an optional config name. If one is given, only attach to a session of that config, otherwise print an error. If none is given, pick the first.
2010-03-15 02:33:43 +01:00
bool Cmdline::isSync()
{
Cmdline::isSync(): cosmetic change Instead of repeating "return false" multiple times, use one if/else check.
2010-06-11 13:00:05 +02:00
// make sure command line arguments really try to run sync
if (m_usage || m_version ||
m_printServers || boost::trim_copy(m_server) == "?" ||
m_printTemplates || m_dontrun ||
m_argc == 1 || (m_useDaemon.wasSet() && m_argc == 2) ||
m_printConfig || m_remove ||
(m_server == "" && m_argc > 1) ||
m_configure || m_migrate ||
m_status || m_printSessions ||
!m_restore.empty() ||
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
m_accessItems ||
Cmdline::isSync(): cosmetic change Instead of repeating "return false" multiple times, use one if/else check.
2010-06-11 13:00:05 +02:00
m_dryrun ||
(!m_run && (m_syncProps.size() || m_sourceProps.size()))) {
Cmdline: move cmdline to dbus server (MB#5043) Implement cmdline with support of dbus server. To enable cmdline with dbus server, use the option '--use-daemon yes/no' in case that you enable dbus service when configuration. In a typical scenario, a new session is created for the purpose of execution of arguments. It is scheduled with other sessions but with a highest priority. Once it becomes active, command line call 'Session.Execute', a newly added method to execute command line arguments. The config name of a session should be known for dbus clients like command line. A new property 'configName' is added in the properties when calling 'Session.GetConfig'. CTRL-C handling are processed once executing a real sync to dbus server. It is mapped to invoke 'Session.Suspend' and 'Session.Abort'. The meaning of '--enable-dbus-service' is expanded accordingly. '--status' without server means printing all running session in the dbus server. '--monitor' could accept an optional config name. If one is given, only attach to a session of that config, otherwise print an error. If none is given, pick the first.
2010-03-15 02:33:43 +01:00
return false;
Cmdline::isSync(): cosmetic change Instead of repeating "return false" multiple times, use one if/else check.
2010-06-11 13:00:05 +02:00
} else {
return true;
Cmdline: move cmdline to dbus server (MB#5043) Implement cmdline with support of dbus server. To enable cmdline with dbus server, use the option '--use-daemon yes/no' in case that you enable dbus service when configuration. In a typical scenario, a new session is created for the purpose of execution of arguments. It is scheduled with other sessions but with a highest priority. Once it becomes active, command line call 'Session.Execute', a newly added method to execute command line arguments. The config name of a session should be known for dbus clients like command line. A new property 'configName' is added in the properties when calling 'Session.GetConfig'. CTRL-C handling are processed once executing a real sync to dbus server. It is mapped to invoke 'Session.Suspend' and 'Session.Abort'. The meaning of '--enable-dbus-service' is expanded accordingly. '--status' without server means printing all running session in the dbus server. '--monitor' could accept an optional config name. If one is given, only attach to a session of that config, otherwise print an error. If none is given, pick the first.
2010-03-15 02:33:43 +01:00
}
}
command line + D-Bus: fix for "some parameters are already handled by parse()" (MB #5013) The criteria for not invoking Cmdline::run() was a bit too broad. For "--template ?", "--help", "--version", "?", "--print-servers" no output was produced in --daemon mode. Perhaps the whole m_dontrun logic should be updated. In order to minimize changes, this patch only duplicates the Cmdline::run() logic.
2010-04-07 18:23:33 +02:00
bool Cmdline::dontRun() const
{
// this mimics the if() checks in run()
if (m_usage || m_version ||
m_printServers || boost::trim_copy(m_server) == "?" ||
m_printTemplates) {
return false;
} else {
return m_dontrun;
}
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
bool Cmdline::run() {
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
// --dry-run is only supported by some operations.
// Be very strict about it and make sure it is off in all
// potentially harmful operations, otherwise users might
// expect it to have an effect when it doesn't.
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
if (m_usage) {
usage(true);
} else if (m_version) {
printf("SyncEvolution %s\n", VERSION);
--enable-evolution-compatibility: run with reduced functionality when functions are missing Availability is logged in the "syncevolution --version" output and at the start of the log. Also log device ID. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@756 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-10-10 23:57:55 +02:00
printf("%s", EDSAbiWrapperInfo());
revised backend API: fixed potential crash after calling SyncSourceBackendDebug/Info() std::iostream.str() created a temporary std::string which was destructed when these functions returned, therefore the c_str() pointer returned by these functions was invalid. Changed the return type so that a dynamically created std::string can be returned. Moved the functions into SyncSource, because conceptually they are related to SyncSource::createSource().
2009-09-27 22:18:24 +02:00
printf("%s", SyncSource::backendsInfo().c_str());
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
} else if (m_printServers || boost::trim_copy(m_server) == "?") {
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
dumpConfigs("Configured servers:",
SyncConfig::getConfigs());
} else if (m_printTemplates) {
SyncConfig::DeviceList devices;
if (m_template.empty()){
dumpConfigTemplates("Available configuration templates:",
SyncConfig::getPeerTemplates(devices), false);
} else {
//limiting at templates for syncml clients only.
DBus server: refine implementation of bluetooth devices 1) Add description for how to implement this feature and give comments of each class for bluetooth devices detection 2) Move sync device list from BluezAdapter to DBusServer. Thus operations on sync device list are also moved to DBusServer. So 'TemplatesChanged' signal could track device list in a more reasonable way 3) Once default adapter is changed, remove devices in the sync device list that belongs to this adapter 4) Other minor changes
2010-02-23 05:06:05 +01:00
devices.push_back (SyncConfig::DeviceDescription("", m_template, SyncConfig::MATCH_FOR_SERVER_MODE));
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
dumpConfigTemplates("Available configuration templates:",
SyncConfig::matchPeerTemplates(devices), true);
}
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
} else if (m_dontrun) {
// user asked for information
command line + daemon: usability improvements (MB #5043) This patch changes what is shown to the user and how the user interacts with the command line. Details below. "--use-daemon [yes/no]" implies that yes/no is optional (square brackets!), with "yes" being the option that could be expected for a plain "--use-daemon" parameter. But the implementation always expects a "yes" or "no" parameter. The original format suggested was "--use-daemon[=yes/no]" This patch switches to that format, changes --use-daemon into --daemon (to be consistent with --keyring) and enables the same format for --keyring. Although not documented, 0/f/false and 1/t/true are also accepted. Because the value becomes part of the parameter, m_argc checks had to be adapted. The documentation for "--use-daemon" was inserted in the middle of the "--keyring documentation". "Parameter not set" has to be available to the Cmdline caller in the command line too, in addition to true/false. This was done originally with a string which is empty, "yes" or "no". Using a tri-state Cmdline::Bool class makes this a bit more explicit and removes the ambiguity around what useDaemon() returns (values weren't documented). When running without --daemon and no daemon available, the first lines of output were: ERROR: org.<cryptic error> [INFO] itodo20: inactive .... => The command line should fall back *silently* to running in-process. => Error messages should be formatted and logged as such, using SE_LOG_ERROR(). Old code might not have done that and we need to preserve that for compatibility with frontends, but new code should use [ERROR] as output. => D-Bus error messages (as in the case above) must have some user (and developer!) comprehensible explanation what went wrong. The D-Bus error class itself is no enough. Although I haven't tested it, I suspect that the code also would have re-run the operation in-process after the D-Bus server already executed it and failed. I rewrote this so that a check for "daemon available" without error messages is done first before committing to using the daemon. Once that decision is made, the command line will not fall back to in-process execution. Rewrote several error messages. Telling a user of a distro's binary to "re-configure" is misleading (he didn't configure himself). "can't" => "cannot", punctuation changes. Not sure whether is always an improvement, comments welcome. Comment on coding style: I've used "if ()" instead of "if()" because that is the GNU recommendation.
2010-03-26 18:47:41 +01:00
} else if (m_argc == 1 || (m_useDaemon.wasSet() && m_argc == 2)) {
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
// no parameters: list databases and short usage
redesigned SyncSource base class + API The main motivation for this change is that it allows the implementor of a backend to choose the implementations for the different aspects of a datasource (change tracking, item import/export, logging, ...) independently of each other. For example, change tracking via revision strings can now be combined with exchanging data with the Synthesis engine via a single string (the traditional method in SyncEvolution) and with direct access to the Synthesis field list (now possible for the first time). The new backend API is based on the concept of providing implementations for certain functionality via function objects instead of implementing certain virtual methods. The advantage is that implementors can define their own, custom interfaces and mix and match implementations of the different groups of functionality. Logging (see SyncSourceLogging in a later commit) can be done by wrapping some arbitrary other item import/export function objects (decorator design pattern). The class hierarchy is now this: - SyncSourceBase: interface for common utility code, all other classes are derived from it and thus can use that code - SyncSource: base class which implements SyncSourceBase and hooks a datasource into the SyncEvolution core; its "struct Operations" holds the function objects which can be implemented in different ways - TestingSyncSource: combines some of the following classes into an interface that is expected by the client-test program; backends only have to derive from (and implement this) if they want to use the automated testing - TrackingSyncSource: provides the same functionality as before (change tracking via revision strings, item import/export as string) in a single interface; the description of the pure virtual methods are duplicated so that developers can go through this class and find everything they need to know to implement it The following classes contain the code that was previously found in the EvolutionSyncSource base class. Implementors can derive from them and call the init() methods to inherit and activate the functionality: - SyncSourceSession: binds Synthesis session callbacks to virtual methods beginSync(), endSync() - SyncSourceChanges: implements Synthesis item tracking callbacks with set of LUIDs that the user of the class has to fill - SyncSourceDelete: binds Synthesis delete callback to virtual method - SyncSourceRaw: read and write items in the backends format, used for testing and backup/restore - SyncSourceSerialize: exchanges items with Synthesis engine using a string representation of the data; this is how EvolutionSyncSource has traditionally worked, so much of the same virtual methods are now in this class - SyncSourceRevisions: utility class which does change tracking via some kind of "revision" string which changes each time an item is modified; this code was previously in the TrackingSyncSource
2009-08-25 09:27:46 +02:00
const SourceRegistry &registry(SyncSource::getSourceRegistry());
config: reorganized for shared config layout (MB#7707) This patch introduces code changes for the new layout without actually using it yet. Therefore all existing tests for the older layout still pass. The new meaning of the former "server name" is introduced: - A plain string now refers to a peer configuration (can be client or server). - The @ sign allows selecting a specific context. Different contexts have independent sets of local sources and peer definitions. - An empty peer name selects a view on the configuration which contains no peer-specific properties. Not fully implemented yet. The FileConfigTree is instantiated with a root which is one level high up compare to before this patch (for example, "~/.config/syncevolution" instead of "./config/syncevolution/scheduleworld") and then config files include that dropped level in their relative path name ("scheduleworld/config.ini" instead of "config.ini"). This allows accessing the global properties in "~/.config/syncevolution/config.ini" and will be used to move peers further down the hierarchy ("~/.config/syncevolution/peers/scheduleworld/config.ini"). To keep the output of "--print-servers" consistent, the FileConfigTree gets another parameter which identifies the subset of the larger tree that is referenced by this FileConfigTree instance. One side effect of this change is that FileConfigTree instances are no longer completely separate. Something to keep in mind when instantiating SyncContext multiple times (MB#8006). Code may no longer make assumptions in which config node a property is stored. This is determined by the new getNode() calls based on the property attributes (hidden, sharing). The new layout is represented as a set of config nodes. Older layouts use the same set of nodes with identical instances assigned to them, if they don't really have separate files for each of them. SyncSourceNodes no longer grants direct access to the nodes, to catch code which incorrectly access a specific node directly. For the same reason the name of the nodes changed. Code which needs access to all hidden or all visible properties now does this via a config node returned by getProperties(). Currently this is identical to the underlying nodes. Once the new layout is active, this node will act as a multiplexer which gathers properties from all underlying nodes when reading and picks the right one when writing. The "change ID" parameter for sources has been obsolete for a while and was removed. Reorganized the property registration so that it is a bit easier to see which properties are hidden and which use non-default sharing. The default sharing is "no sharing". Some other code was improved while touching it: - removed useless visibility[] array in favor of a i != 0 check in SyncConfig::copy() - added default parameters to save/checkPassword() methods - some constness definition changes - Property::getProperty() is a virtual call which could only be overloaded in one case because the constness was wrong; now getProperty() always returns the string value and getPropertyValue() some other kind of representation of it, depending on the class. - ConstSyncSourceNodes is based on SyncSourceNodes instead of duplicating it, which simplifies the implementation. The simplified SyncSourceAdmin API changed slightly: instead of passing a pointer to the source's SyncSourceNodes, the default nodes are now found via the SyncSource pointer. For callers this is a bit less work and it is more general.
2009-11-13 14:52:00 +01:00
boost::shared_ptr<FilterConfigNode> sharedNode(new VolatileConfigNode());
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
boost::shared_ptr<FilterConfigNode> configNode(new VolatileConfigNode());
boost::shared_ptr<FilterConfigNode> hiddenNode(new VolatileConfigNode());
boost::shared_ptr<FilterConfigNode> trackingNode(new VolatileConfigNode());
SyncML server: handle admin data inside SyncEvolution, use <simpleauthuser/pw> Previously, sync failed because the datastore configuration specified no means of storing the admin data (single chunk of text and local ID/remote ID mapping). This could have been added by configuring the SDK_textdb as <plugin_admin_module>. Instead this patch implements the functionality inside SyncEvolution, using a new ".server.ini" config node for the mapping and an internal "adminData" property for the admin data text chunk. This is more natural because it keeps the data under our control. Authentication is configured via <simpleauthuser/pw>. In contrast to using the SDK_textdb, this allows choosing the username and password. Ultimately this should also be done inside SyncEvolution, to avoid writing the username/password into the XML (unsolved encoding issue) and into log files (privacy issue).
2009-09-29 22:41:06 +02:00
boost::shared_ptr<FilterConfigNode> serverNode(new VolatileConfigNode());
enable suspend and saving blobs (MB #2425) Implement blob support. Blobs are stored in the per-peer source directory in a ".cache" sub-directory. We could have relied on the peer name and device name to make file names unique when sharing a common directory. Perhaps that is indeed the better solution: ~/.cache/syncevolution/blobs with logdir specifying the root? Blob support is only needed when running as server. When running as client, the binfile layer provides the necessary implementation. Support for suspend was off by default because <resumesupport> was not set. Adding it, because we should have whatever is needed. <resumeitemsupport> depends on the ability to store blobs. Not sure whether this is relevant on the client side, where these services are provided by the binfile layer.
2010-03-04 17:50:01 +01:00
SyncSourceNodes nodes(true, sharedNode, configNode, hiddenNode, trackingNode, serverNode, "");
config: reorganized for shared config layout (MB#7707) This patch introduces code changes for the new layout without actually using it yet. Therefore all existing tests for the older layout still pass. The new meaning of the former "server name" is introduced: - A plain string now refers to a peer configuration (can be client or server). - The @ sign allows selecting a specific context. Different contexts have independent sets of local sources and peer definitions. - An empty peer name selects a view on the configuration which contains no peer-specific properties. Not fully implemented yet. The FileConfigTree is instantiated with a root which is one level high up compare to before this patch (for example, "~/.config/syncevolution" instead of "./config/syncevolution/scheduleworld") and then config files include that dropped level in their relative path name ("scheduleworld/config.ini" instead of "config.ini"). This allows accessing the global properties in "~/.config/syncevolution/config.ini" and will be used to move peers further down the hierarchy ("~/.config/syncevolution/peers/scheduleworld/config.ini"). To keep the output of "--print-servers" consistent, the FileConfigTree gets another parameter which identifies the subset of the larger tree that is referenced by this FileConfigTree instance. One side effect of this change is that FileConfigTree instances are no longer completely separate. Something to keep in mind when instantiating SyncContext multiple times (MB#8006). Code may no longer make assumptions in which config node a property is stored. This is determined by the new getNode() calls based on the property attributes (hidden, sharing). The new layout is represented as a set of config nodes. Older layouts use the same set of nodes with identical instances assigned to them, if they don't really have separate files for each of them. SyncSourceNodes no longer grants direct access to the nodes, to catch code which incorrectly access a specific node directly. For the same reason the name of the nodes changed. Code which needs access to all hidden or all visible properties now does this via a config node returned by getProperties(). Currently this is identical to the underlying nodes. Once the new layout is active, this node will act as a multiplexer which gathers properties from all underlying nodes when reading and picks the right one when writing. The "change ID" parameter for sources has been obsolete for a while and was removed. Reorganized the property registration so that it is a bit easier to see which properties are hidden and which use non-default sharing. The default sharing is "no sharing". Some other code was improved while touching it: - removed useless visibility[] array in favor of a i != 0 check in SyncConfig::copy() - added default parameters to save/checkPassword() methods - some constness definition changes - Property::getProperty() is a virtual call which could only be overloaded in one case because the constness was wrong; now getProperty() always returns the string value and getPropertyValue() some other kind of representation of it, depending on the class. - ConstSyncSourceNodes is based on SyncSourceNodes instead of duplicating it, which simplifies the implementation. The simplified SyncSourceAdmin API changed slightly: instead of passing a pointer to the source's SyncSourceNodes, the default nodes are now found via the SyncSource pointer. For callers this is a bit less work and it is more general.
2009-11-13 14:52:00 +01:00
SyncSourceParams params("list", nodes);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
BOOST_FOREACH(const RegisterSyncSource *source, registry) {
BOOST_FOREACH(const Values::value_type &alias, source->m_typeValues) {
if (!alias.empty() && source->m_enabled) {
configNode->setProperty("type", *alias.begin());
redesigned SyncSource base class + API The main motivation for this change is that it allows the implementor of a backend to choose the implementations for the different aspects of a datasource (change tracking, item import/export, logging, ...) independently of each other. For example, change tracking via revision strings can now be combined with exchanging data with the Synthesis engine via a single string (the traditional method in SyncEvolution) and with direct access to the Synthesis field list (now possible for the first time). The new backend API is based on the concept of providing implementations for certain functionality via function objects instead of implementing certain virtual methods. The advantage is that implementors can define their own, custom interfaces and mix and match implementations of the different groups of functionality. Logging (see SyncSourceLogging in a later commit) can be done by wrapping some arbitrary other item import/export function objects (decorator design pattern). The class hierarchy is now this: - SyncSourceBase: interface for common utility code, all other classes are derived from it and thus can use that code - SyncSource: base class which implements SyncSourceBase and hooks a datasource into the SyncEvolution core; its "struct Operations" holds the function objects which can be implemented in different ways - TestingSyncSource: combines some of the following classes into an interface that is expected by the client-test program; backends only have to derive from (and implement this) if they want to use the automated testing - TrackingSyncSource: provides the same functionality as before (change tracking via revision strings, item import/export as string) in a single interface; the description of the pure virtual methods are duplicated so that developers can go through this class and find everything they need to know to implement it The following classes contain the code that was previously found in the EvolutionSyncSource base class. Implementors can derive from them and call the init() methods to inherit and activate the functionality: - SyncSourceSession: binds Synthesis session callbacks to virtual methods beginSync(), endSync() - SyncSourceChanges: implements Synthesis item tracking callbacks with set of LUIDs that the user of the class has to fill - SyncSourceDelete: binds Synthesis delete callback to virtual method - SyncSourceRaw: read and write items in the backends format, used for testing and backup/restore - SyncSourceSerialize: exchanges items with Synthesis engine using a string representation of the data; this is how EvolutionSyncSource has traditionally worked, so much of the same virtual methods are now in this class - SyncSourceRevisions: utility class which does change tracking via some kind of "revision" string which changes each time an item is modified; this code was previously in the TrackingSyncSource
2009-08-25 09:27:46 +02:00
auto_ptr<SyncSource> source(SyncSource::createSource(params, false));
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
if (source.get() != NULL) {
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
listSources(*source, boost::join(alias, " = "));
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
m_out << "\n";
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
}
}
}
usage(false);
} else if (m_printConfig) {
more classes renamed EvolutionSyncConfig => SyncConfig SyncEvolutionException => Exception EvolutionUnref* => Unref*
2009-10-06 17:22:47 +02:00
boost::shared_ptr<SyncConfig> config;
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
ConfigProps syncFilter;
SourceFilters_t sourceFilters;
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
if (m_template.empty()) {
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
if (m_server.empty()) {
m_err << "ERROR: --print-config requires either a --template or a server name." << endl;
return false;
}
more classes renamed EvolutionSyncConfig => SyncConfig SyncEvolutionException => Exception EvolutionUnref* => Unref*
2009-10-06 17:22:47 +02:00
config.reset(new SyncConfig(m_server));
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
if (!config->exists()) {
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
m_err << "ERROR: server '" << m_server << "' has not been configured yet." << endl;
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
return false;
}
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
syncFilter = m_syncProps;
sourceFilters[""] = m_sourceProps;
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
} else {
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
string peer, context;
SyncConfig::splitConfigString(SyncConfig::normalizeConfigString(m_template), peer, context);
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
config = SyncConfig::createPeerTemplate(peer);
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
if (!config.get()) {
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
m_err << "ERROR: no configuration template for '" << m_template << "' available." << endl;
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
return false;
}
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
getFilters(context, syncFilter, sourceFilters);
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
}
command line: --print-config of context should ignore per-peer properties (MB #8048) When printing the context config, per-peer properties where shown with their default values, which was a bit confusing. Now they are skipped.
2010-03-02 16:27:22 +01:00
// determine whether we dump a peer or a context
int flags = DUMP_PROPS_NORMAL;
string peer, context;
SyncConfig::splitConfigString(config->getConfigName(), peer, context);
if (peer.empty()) {
flags |= HIDE_PER_PEER;
command line: check for invalid peer properties (MB #8048) When using --print-config or --configure with a context as config name (like @default), then per-peer properties do not have any effect. This might be confusing for users, so the command line should better report this as error and refuse to execute the operation. As usual, "type" is a bit special because it is both shared and per-peer.
2010-03-03 16:20:53 +01:00
checkForPeerProps();
command line: --print-config of context should ignore per-peer properties (MB #8048) When printing the context config, per-peer properties where shown with their default values, which was a bit confusing. Now they are skipped.
2010-03-02 16:27:22 +01:00
}
improved command line and its documented More explicit description of the different operations. Added --run|-r switch: if properties are on the command line, then an operation must be specified explicitly, with --run standing for a sync run. Otherwise a sync might accidentally be done when a configuration change was intented, but the --configure switch was forgotten. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@696 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-08-03 15:00:51 +02:00
if (m_sources.empty() ||
m_sources.find("main") != m_sources.end()) {
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
boost::shared_ptr<FilterConfigNode> syncProps(config->getProperties());
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
syncProps->setFilter(syncFilter);
command line: --print-config of context should ignore per-peer properties (MB #8048) When printing the context config, per-peer properties where shown with their default values, which was a bit confusing. Now they are skipped.
2010-03-02 16:27:22 +01:00
dumpProperties(*syncProps, config->getRegistry(), flags);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
code cleanup (boost::join, list sort) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@659 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-06 22:49:19 +02:00
list<string> sources = config->getSyncSources();
sources.sort();
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
BOOST_FOREACH(const string &name, sources) {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
if (m_sources.empty() ||
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
m_sources.find(name) != m_sources.end()) {
m_out << endl << "[" << name << "]" << endl;
config: reorganized for shared config layout (MB#7707) This patch introduces code changes for the new layout without actually using it yet. Therefore all existing tests for the older layout still pass. The new meaning of the former "server name" is introduced: - A plain string now refers to a peer configuration (can be client or server). - The @ sign allows selecting a specific context. Different contexts have independent sets of local sources and peer definitions. - An empty peer name selects a view on the configuration which contains no peer-specific properties. Not fully implemented yet. The FileConfigTree is instantiated with a root which is one level high up compare to before this patch (for example, "~/.config/syncevolution" instead of "./config/syncevolution/scheduleworld") and then config files include that dropped level in their relative path name ("scheduleworld/config.ini" instead of "config.ini"). This allows accessing the global properties in "~/.config/syncevolution/config.ini" and will be used to move peers further down the hierarchy ("~/.config/syncevolution/peers/scheduleworld/config.ini"). To keep the output of "--print-servers" consistent, the FileConfigTree gets another parameter which identifies the subset of the larger tree that is referenced by this FileConfigTree instance. One side effect of this change is that FileConfigTree instances are no longer completely separate. Something to keep in mind when instantiating SyncContext multiple times (MB#8006). Code may no longer make assumptions in which config node a property is stored. This is determined by the new getNode() calls based on the property attributes (hidden, sharing). The new layout is represented as a set of config nodes. Older layouts use the same set of nodes with identical instances assigned to them, if they don't really have separate files for each of them. SyncSourceNodes no longer grants direct access to the nodes, to catch code which incorrectly access a specific node directly. For the same reason the name of the nodes changed. Code which needs access to all hidden or all visible properties now does this via a config node returned by getProperties(). Currently this is identical to the underlying nodes. Once the new layout is active, this node will act as a multiplexer which gathers properties from all underlying nodes when reading and picks the right one when writing. The "change ID" parameter for sources has been obsolete for a while and was removed. Reorganized the property registration so that it is a bit easier to see which properties are hidden and which use non-default sharing. The default sharing is "no sharing". Some other code was improved while touching it: - removed useless visibility[] array in favor of a i != 0 check in SyncConfig::copy() - added default parameters to save/checkPassword() methods - some constness definition changes - Property::getProperty() is a virtual call which could only be overloaded in one case because the constness was wrong; now getProperty() always returns the string value and getPropertyValue() some other kind of representation of it, depending on the class. - ConstSyncSourceNodes is based on SyncSourceNodes instead of duplicating it, which simplifies the implementation. The simplified SyncSourceAdmin API changed slightly: instead of passing a pointer to the source's SyncSourceNodes, the default nodes are now found via the SyncSource pointer. For callers this is a bit less work and it is more general.
2009-11-13 14:52:00 +01:00
SyncSourceNodes nodes = config->getSyncSourceNodes(name);
boost::shared_ptr<FilterConfigNode> sourceProps = nodes.getProperties();
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
SourceFilters_t::const_iterator it = sourceFilters.find(name);
if (it != sourceFilters.end()) {
sourceProps->setFilter(it->second);
} else {
sourceProps->setFilter(sourceFilters[""]);
}
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
dumpProperties(*sourceProps, SyncSourceConfig::getRegistry(),
command line: --print-config of context should ignore per-peer properties (MB #8048) When printing the context config, per-peer properties where shown with their default values, which was a bit confusing. Now they are skipped.
2010-03-02 16:27:22 +01:00
flags | ((name != *(--sources.end())) ? HIDE_LEGEND : DUMP_PROPS_NORMAL));
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
}
implemented server config templates (scheduleworld, funambol, synthesis) and printing of them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@526 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:35:22 +01:00
} else if (m_server == "" && m_argc > 1) {
// Options given, but no server - not sure what the user wanted?!
usage(true, "server name missing");
return false;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
} else if (m_configure || m_migrate) {
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
if (m_dryrun) {
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
SyncContext::throwError("--dry-run not supported for configuration changes");
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
}
command line: add keyring support (MB#3604) --keyring|-k Save or retrieve passwords from the GNOME keyring when modifying the configuration or running a synchronization. Note that using this option applies to *all* passwords in a configuration, so setting a single password as follows moves the other passwords into the keyring, if they were not stored there already: --keyring --configure --sync-property proxyPassword=foo When passwords were stored in the keyring, their value is set to "-" in the configuration. This means that when running a synchronization without the --keyring argument, the password has to be entered interactively. The implementation introduces new virtual methods for password handling to properties and iterates over all properties to activate that special behavior. Another change is no longer to use cached strings in the SyncEvolutionConfig to store retrieved passwords. Instead, they are saved as the filter in the config node. This could help their users to get them but not flush to files. The main purpose of this change is to make it flexible to meet the requirements of dynamically added passwords in the backend.
2009-09-21 10:07:05 +02:00
if (m_keyring) {
#ifndef USE_GNOME_KEYRING
m_err << "Error: this syncevolution binary was compiled without support for storing "
"passwords in a keyring. Either store passwords in your configuration "
"files or enter them interactively on each program run." << endl;
return false;
#endif
}
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
bool fromScratch = false;
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
string peer, context;
SyncConfig::splitConfigString(SyncConfig::normalizeConfigString(m_server), peer, context);
command line: check for invalid peer properties (MB #8048) When using --print-config or --configure with a context as config name (like @default), then per-peer properties do not have any effect. This might be confusing for users, so the command line should better report this as error and refuse to execute the operation. As usual, "type" is a bit special because it is both shared and per-peer.
2010-03-03 16:20:53 +01:00
if (peer.empty()) {
checkForPeerProps();
}
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
// Both config changes and migration are implemented as copying from
// another config (template resp. old one). Migration also moves
// the old config.
more classes renamed EvolutionSyncConfig => SyncConfig SyncEvolutionException => Exception EvolutionUnref* => Unref*
2009-10-06 17:22:47 +02:00
boost::shared_ptr<SyncConfig> from;
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
if (m_migrate) {
syncevolution --migrate: support migration into a certain context This depends on peer/context awareness in the code which looks for the old config and renames it. If the old config is not in the same context as the new one, it is searched by peer name alone. This works for all old-style configs as well as rewriting a peer inside the same context.
2009-11-25 16:10:44 +01:00
string oldContext = context;
more classes renamed EvolutionSyncConfig => SyncConfig SyncEvolutionException => Exception EvolutionUnref* => Unref*
2009-10-06 17:22:47 +02:00
from.reset(new SyncConfig(m_server));
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
if (!from->exists()) {
syncevolution --migrate: support migration into a certain context This depends on peer/context awareness in the code which looks for the old config and renames it. If the old config is not in the same context as the new one, it is searched by peer name alone. This works for all old-style configs as well as rewriting a peer inside the same context.
2009-11-25 16:10:44 +01:00
// for migration into a different context, search for config without context
oldContext = "";
from.reset(new SyncConfig(peer));
if (!from->exists()) {
m_err << "ERROR: server '" << m_server << "' has not been configured yet." << endl;
return false;
}
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
}
int counter = 0;
string oldRoot = from->getRootPath();
string suffix;
while (true) {
string newname;
ostringstream newsuffix;
newsuffix << ".old";
if (counter) {
newsuffix << "." << counter;
}
suffix = newsuffix.str();
newname = oldRoot + suffix;
if (!rename(oldRoot.c_str(),
newname.c_str())) {
break;
} else if (errno != EEXIST && errno != ENOTEMPTY) {
m_err << "ERROR: renaming " << oldRoot << " to " <<
newname << ": " << strerror(errno) << endl;
return false;
}
counter++;
}
syncevolution --migrate: support migration into a certain context This depends on peer/context awareness in the code which looks for the old config and renames it. If the old config is not in the same context as the new one, it is searched by peer name alone. This works for all old-style configs as well as rewriting a peer inside the same context.
2009-11-25 16:10:44 +01:00
from.reset(new SyncConfig(peer + suffix +
(oldContext.empty() ? "" : "@") +
oldContext));
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
} else {
more classes renamed EvolutionSyncConfig => SyncConfig SyncEvolutionException => Exception EvolutionUnref* => Unref*
2009-10-06 17:22:47 +02:00
from.reset(new SyncConfig(m_server));
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
if (!from->exists()) {
// creating from scratch, look for template
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
fromScratch = true;
command line --template new config in non-standard context: wrong properties written (MB #10007) First, "syncevolution --configure --template scheduleworld@context ..." should trigger an error. The context, if any, must be specified as part of the config name. Second, "syncevolution --configure template scheduleworld foo@bar", when foo@bar did not exist yet, copied the shared properties from the default context into foo@bar and the context @bar. Determining the context had to distinguish between "--template scheduleworld" and "scheduleworld@bar", which implies the "scheduleworld" template.
2010-03-02 21:28:28 +01:00
string configTemplate;
if (m_template.empty()) {
// template is the peer name
configTemplate = m_server;
SyncConfig::splitConfigString(SyncConfig::normalizeConfigString(configTemplate), peer, context);
} else {
// Template is specified explicitly. It must not contain a context,
// because the context comes from the config name.
configTemplate = m_template;
if (SyncConfig::splitConfigString(SyncConfig::normalizeConfigString(configTemplate), peer, context)) {
m_err << "ERROR: template " << configTemplate << " must not specify a context." << endl;
return false;
}
string tmp;
SyncConfig::splitConfigString(SyncConfig::normalizeConfigString(m_server), tmp, context);
}
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
from = SyncConfig::createPeerTemplate(peer);
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
if (!from.get()) {
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
m_err << "ERROR: no configuration template for '" << configTemplate << "' available." << endl;
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
dumpConfigTemplates("Available configuration templates:",
SyncConfig::getPeerTemplates(SyncConfig::DeviceList()));
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
return false;
}
username/password: empty in templates We used to have the strings "your SyncML server account name" and "your SyncML server password" hard-coded in the code which generated templates. As Jussi pointed out, this is a problem for GUIs which have to hide these strings from their users. The purpose of these strings was to have them in configs in cases where the user forgot to set them, as a reminder. This path moves the strings into the command line config setup, thus achieving both of these goals.
2009-11-19 16:30:46 +01:00
username/password must be empty for clients (MB #9437) The "your SyncML server account name/password" strings inserted into configs by the command line only make sense when really talking to a server. In configs on the server side they are harmful, because clients like phones are then expected to authorize themselves accordingly. This patch removes the strings from configs where peerIsClient is true.
2010-03-02 21:30:13 +01:00
if (!from->getPeerIsClient()) {
// Templates no longer contain these strings, because
// GUIs would have to localize them. For configs created
// via the command line, the extra hint that these
// properties need to be set is useful, so set these
// strings here. They'll get copied into the new
// config only if no other value was given on the
// command line.
if (!from->getUsername()[0]) {
from->setUsername("your SyncML server account name");
}
if (!from->getPassword()[0]) {
from->setPassword("your SyncML server password");
}
Configuration templates: uncomment SyncURL (MB#10015) Set 'SyncURL = input your peer address' to avoid SyncURL been commented, which will cause the configuration not shown by UI.
2010-03-04 03:54:54 +01:00
} else {
// uncomment SyncURL, so that it can be shown by
// sync-ui
if (from->getSyncURL().size() == 0) {
from->setSyncURL ("input your peer address here");
}
username/password: empty in templates We used to have the strings "your SyncML server account name" and "your SyncML server password" hard-coded in the code which generated templates. As Jussi pointed out, this is a problem for GUIs which have to hide these strings from their users. The purpose of these strings was to have them in configs in cases where the user forgot to set them, as a reminder. This path moves the strings into the command line config setup, thus achieving both of these goals.
2009-11-19 16:30:46 +01:00
}
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
}
}
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
// Apply config changes on-the-fly. Regardless what we do
// (changing an existing config, migrating, creating from
// a template), existing shared properties in the desired
// context must be preserved unless explicitly overwritten.
// Therefore read those, update with command line properties,
// then set as filter.
ConfigProps syncFilter;
SourceFilters_t sourceFilters;
getFilters(context, syncFilter, sourceFilters);
from->setConfigFilter(true, "", syncFilter);
BOOST_FOREACH(const SourceFilters_t::value_type &entry, sourceFilters) {
from->setConfigFilter(false, entry.first, entry.second);
}
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
// write into the requested configuration, creating it if necessary
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
boost::shared_ptr<SyncContext> to(createSyncClient());
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
to->copy(*from, !fromScratch && !m_sources.empty() ? &m_sources : NULL);
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
SF #1991286: when creating config for the first time, only enable sync sources which can be synchronized The check for that was completely missing. Now SyncEvolution checks that the backend is available and there is at least one database (the first one will be synchronized unless explicitly changed). If these checks fail and the sync source was explicitly requested by the user by listing it after the server name, then an error is printed and no configuration is written. If the user wants the default setup, then the source is silently disabled. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@650 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-02 21:36:02 +02:00
// Sources are active now according to the server default.
// Disable all sources not selected by user (if any selected)
// and those which have no database.
if (fromScratch) {
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
list<string> configuredSources = to->getSyncSources();
SF #1991286: when creating config for the first time, only enable sync sources which can be synchronized The check for that was completely missing. Now SyncEvolution checks that the backend is available and there is at least one database (the first one will be synchronized unless explicitly changed). If these checks fail and the sync source was explicitly requested by the user by listing it after the server name, then an error is printed and no configuration is written. If the user wants the default setup, then the source is silently disabled. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@650 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-02 21:36:02 +02:00
set<string> sources = m_sources;
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
BOOST_FOREACH(const string &source, configuredSources) {
redesigned SyncSource base class + API The main motivation for this change is that it allows the implementor of a backend to choose the implementations for the different aspects of a datasource (change tracking, item import/export, logging, ...) independently of each other. For example, change tracking via revision strings can now be combined with exchanging data with the Synthesis engine via a single string (the traditional method in SyncEvolution) and with direct access to the Synthesis field list (now possible for the first time). The new backend API is based on the concept of providing implementations for certain functionality via function objects instead of implementing certain virtual methods. The advantage is that implementors can define their own, custom interfaces and mix and match implementations of the different groups of functionality. Logging (see SyncSourceLogging in a later commit) can be done by wrapping some arbitrary other item import/export function objects (decorator design pattern). The class hierarchy is now this: - SyncSourceBase: interface for common utility code, all other classes are derived from it and thus can use that code - SyncSource: base class which implements SyncSourceBase and hooks a datasource into the SyncEvolution core; its "struct Operations" holds the function objects which can be implemented in different ways - TestingSyncSource: combines some of the following classes into an interface that is expected by the client-test program; backends only have to derive from (and implement this) if they want to use the automated testing - TrackingSyncSource: provides the same functionality as before (change tracking via revision strings, item import/export as string) in a single interface; the description of the pure virtual methods are duplicated so that developers can go through this class and find everything they need to know to implement it The following classes contain the code that was previously found in the EvolutionSyncSource base class. Implementors can derive from them and call the init() methods to inherit and activate the functionality: - SyncSourceSession: binds Synthesis session callbacks to virtual methods beginSync(), endSync() - SyncSourceChanges: implements Synthesis item tracking callbacks with set of LUIDs that the user of the class has to fill - SyncSourceDelete: binds Synthesis delete callback to virtual method - SyncSourceRaw: read and write items in the backends format, used for testing and backup/restore - SyncSourceSerialize: exchanges items with Synthesis engine using a string representation of the data; this is how EvolutionSyncSource has traditionally worked, so much of the same virtual methods are now in this class - SyncSourceRevisions: utility class which does change tracking via some kind of "revision" string which changes each time an item is modified; this code was previously in the TrackingSyncSource
2009-08-25 09:27:46 +02:00
boost::shared_ptr<PersistentSyncSourceConfig> sourceConfig(to->getSyncSourceConfig(source));
SF #1991286: when creating config for the first time, only enable sync sources which can be synchronized The check for that was completely missing. Now SyncEvolution checks that the backend is available and there is at least one database (the first one will be synchronized unless explicitly changed). If these checks fail and the sync source was explicitly requested by the user by listing it after the server name, then an error is printed and no configuration is written. If the user wants the default setup, then the source is silently disabled. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@650 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-02 21:36:02 +02:00
string disable = "";
set<string>::iterator entry = sources.find(source);
bool selected = entry != sources.end();
if (!m_sources.empty() &&
!selected) {
disable = "not selected";
} else {
if (entry != sources.end()) {
// The command line parameter matched a valid source.
// All entries left afterwards must have been typos.
sources.erase(entry);
}
// check whether the sync source works
config: reorganized for shared config layout (MB#7707) This patch introduces code changes for the new layout without actually using it yet. Therefore all existing tests for the older layout still pass. The new meaning of the former "server name" is introduced: - A plain string now refers to a peer configuration (can be client or server). - The @ sign allows selecting a specific context. Different contexts have independent sets of local sources and peer definitions. - An empty peer name selects a view on the configuration which contains no peer-specific properties. Not fully implemented yet. The FileConfigTree is instantiated with a root which is one level high up compare to before this patch (for example, "~/.config/syncevolution" instead of "./config/syncevolution/scheduleworld") and then config files include that dropped level in their relative path name ("scheduleworld/config.ini" instead of "config.ini"). This allows accessing the global properties in "~/.config/syncevolution/config.ini" and will be used to move peers further down the hierarchy ("~/.config/syncevolution/peers/scheduleworld/config.ini"). To keep the output of "--print-servers" consistent, the FileConfigTree gets another parameter which identifies the subset of the larger tree that is referenced by this FileConfigTree instance. One side effect of this change is that FileConfigTree instances are no longer completely separate. Something to keep in mind when instantiating SyncContext multiple times (MB#8006). Code may no longer make assumptions in which config node a property is stored. This is determined by the new getNode() calls based on the property attributes (hidden, sharing). The new layout is represented as a set of config nodes. Older layouts use the same set of nodes with identical instances assigned to them, if they don't really have separate files for each of them. SyncSourceNodes no longer grants direct access to the nodes, to catch code which incorrectly access a specific node directly. For the same reason the name of the nodes changed. Code which needs access to all hidden or all visible properties now does this via a config node returned by getProperties(). Currently this is identical to the underlying nodes. Once the new layout is active, this node will act as a multiplexer which gathers properties from all underlying nodes when reading and picks the right one when writing. The "change ID" parameter for sources has been obsolete for a while and was removed. Reorganized the property registration so that it is a bit easier to see which properties are hidden and which use non-default sharing. The default sharing is "no sharing". Some other code was improved while touching it: - removed useless visibility[] array in favor of a i != 0 check in SyncConfig::copy() - added default parameters to save/checkPassword() methods - some constness definition changes - Property::getProperty() is a virtual call which could only be overloaded in one case because the constness was wrong; now getProperty() always returns the string value and getPropertyValue() some other kind of representation of it, depending on the class. - ConstSyncSourceNodes is based on SyncSourceNodes instead of duplicating it, which simplifies the implementation. The simplified SyncSourceAdmin API changed slightly: instead of passing a pointer to the source's SyncSourceNodes, the default nodes are now found via the SyncSource pointer. For callers this is a bit less work and it is more general.
2009-11-13 14:52:00 +01:00
SyncSourceParams params("list", to->getSyncSourceNodes(source));
VirtualSyncSource: implement getDatabases() This is required when creating the configuration with virtual sources, otherwise a virtual SyncSource will always be marked as 'disabled' because it thought there is "no database to synchronize".
2010-03-17 10:03:59 +01:00
auto_ptr<SyncSource> syncSource(SyncSource::createSource(params, false, to.get()));
SF #1991286: when creating config for the first time, only enable sync sources which can be synchronized The check for that was completely missing. Now SyncEvolution checks that the backend is available and there is at least one database (the first one will be synchronized unless explicitly changed). If these checks fail and the sync source was explicitly requested by the user by listing it after the server name, then an error is printed and no configuration is written. If the user wants the default setup, then the source is silently disabled. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@650 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-02 21:36:02 +02:00
if (syncSource.get() == NULL) {
disable = "no backend available";
} else {
try {
redesigned SyncSource base class + API The main motivation for this change is that it allows the implementor of a backend to choose the implementations for the different aspects of a datasource (change tracking, item import/export, logging, ...) independently of each other. For example, change tracking via revision strings can now be combined with exchanging data with the Synthesis engine via a single string (the traditional method in SyncEvolution) and with direct access to the Synthesis field list (now possible for the first time). The new backend API is based on the concept of providing implementations for certain functionality via function objects instead of implementing certain virtual methods. The advantage is that implementors can define their own, custom interfaces and mix and match implementations of the different groups of functionality. Logging (see SyncSourceLogging in a later commit) can be done by wrapping some arbitrary other item import/export function objects (decorator design pattern). The class hierarchy is now this: - SyncSourceBase: interface for common utility code, all other classes are derived from it and thus can use that code - SyncSource: base class which implements SyncSourceBase and hooks a datasource into the SyncEvolution core; its "struct Operations" holds the function objects which can be implemented in different ways - TestingSyncSource: combines some of the following classes into an interface that is expected by the client-test program; backends only have to derive from (and implement this) if they want to use the automated testing - TrackingSyncSource: provides the same functionality as before (change tracking via revision strings, item import/export as string) in a single interface; the description of the pure virtual methods are duplicated so that developers can go through this class and find everything they need to know to implement it The following classes contain the code that was previously found in the EvolutionSyncSource base class. Implementors can derive from them and call the init() methods to inherit and activate the functionality: - SyncSourceSession: binds Synthesis session callbacks to virtual methods beginSync(), endSync() - SyncSourceChanges: implements Synthesis item tracking callbacks with set of LUIDs that the user of the class has to fill - SyncSourceDelete: binds Synthesis delete callback to virtual method - SyncSourceRaw: read and write items in the backends format, used for testing and backup/restore - SyncSourceSerialize: exchanges items with Synthesis engine using a string representation of the data; this is how EvolutionSyncSource has traditionally worked, so much of the same virtual methods are now in this class - SyncSourceRevisions: utility class which does change tracking via some kind of "revision" string which changes each time an item is modified; this code was previously in the TrackingSyncSource
2009-08-25 09:27:46 +02:00
SyncSource::Databases databases = syncSource->getDatabases();
cleaned up EvolutionSyncSource interface+implementation git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@664 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-10 21:17:42 +02:00
if (databases.empty()) {
SF #1991286: when creating config for the first time, only enable sync sources which can be synchronized The check for that was completely missing. Now SyncEvolution checks that the backend is available and there is at least one database (the first one will be synchronized unless explicitly changed). If these checks fail and the sync source was explicitly requested by the user by listing it after the server name, then an error is printed and no configuration is written. If the user wants the default setup, then the source is silently disabled. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@650 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-02 21:36:02 +02:00
disable = "no database to synchronize";
}
} catch (...) {
disable = "backend failed";
}
}
}
if (!disable.empty()) {
// abort if the user explicitly asked for the sync source
// and it cannot be enabled, otherwise disable it silently
if (selected) {
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
SyncContext::throwError(source + ": " + disable);
SF #1991286: when creating config for the first time, only enable sync sources which can be synchronized The check for that was completely missing. Now SyncEvolution checks that the backend is available and there is at least one database (the first one will be synchronized unless explicitly changed). If these checks fail and the sync source was explicitly requested by the user by listing it after the server name, then an error is printed and no configuration is written. If the user wants the default setup, then the source is silently disabled. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@650 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-02 21:36:02 +02:00
}
sourceConfig->setSync("disabled");
} else if (selected) {
// user absolutely wants it: enable even if off by default
FilterConfigNode::ConfigFilter::const_iterator sync =
redesigned SyncSource base class + API The main motivation for this change is that it allows the implementor of a backend to choose the implementations for the different aspects of a datasource (change tracking, item import/export, logging, ...) independently of each other. For example, change tracking via revision strings can now be combined with exchanging data with the Synthesis engine via a single string (the traditional method in SyncEvolution) and with direct access to the Synthesis field list (now possible for the first time). The new backend API is based on the concept of providing implementations for certain functionality via function objects instead of implementing certain virtual methods. The advantage is that implementors can define their own, custom interfaces and mix and match implementations of the different groups of functionality. Logging (see SyncSourceLogging in a later commit) can be done by wrapping some arbitrary other item import/export function objects (decorator design pattern). The class hierarchy is now this: - SyncSourceBase: interface for common utility code, all other classes are derived from it and thus can use that code - SyncSource: base class which implements SyncSourceBase and hooks a datasource into the SyncEvolution core; its "struct Operations" holds the function objects which can be implemented in different ways - TestingSyncSource: combines some of the following classes into an interface that is expected by the client-test program; backends only have to derive from (and implement this) if they want to use the automated testing - TrackingSyncSource: provides the same functionality as before (change tracking via revision strings, item import/export as string) in a single interface; the description of the pure virtual methods are duplicated so that developers can go through this class and find everything they need to know to implement it The following classes contain the code that was previously found in the EvolutionSyncSource base class. Implementors can derive from them and call the init() methods to inherit and activate the functionality: - SyncSourceSession: binds Synthesis session callbacks to virtual methods beginSync(), endSync() - SyncSourceChanges: implements Synthesis item tracking callbacks with set of LUIDs that the user of the class has to fill - SyncSourceDelete: binds Synthesis delete callback to virtual method - SyncSourceRaw: read and write items in the backends format, used for testing and backup/restore - SyncSourceSerialize: exchanges items with Synthesis engine using a string representation of the data; this is how EvolutionSyncSource has traditionally worked, so much of the same virtual methods are now in this class - SyncSourceRevisions: utility class which does change tracking via some kind of "revision" string which changes each time an item is modified; this code was previously in the TrackingSyncSource
2009-08-25 09:27:46 +02:00
m_sourceProps.find(SyncSourceConfig::m_sourcePropSync.getName());
SF #1991286: when creating config for the first time, only enable sync sources which can be synchronized The check for that was completely missing. Now SyncEvolution checks that the backend is available and there is at least one database (the first one will be synchronized unless explicitly changed). If these checks fail and the sync source was explicitly requested by the user by listing it after the server name, then an error is printed and no configuration is written. If the user wants the default setup, then the source is silently disabled. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@650 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-02 21:36:02 +02:00
sourceConfig->setSync(sync == m_sourceProps.end() ? "two-way" : sync->second);
}
}
if (!sources.empty()) {
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
SyncContext::throwError(string("no such source(s): ") + boost::join(sources, " "));
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
}
}
command line: add keyring support (MB#3604) --keyring|-k Save or retrieve passwords from the GNOME keyring when modifying the configuration or running a synchronization. Note that using this option applies to *all* passwords in a configuration, so setting a single password as follows moves the other passwords into the keyring, if they were not stored there already: --keyring --configure --sync-property proxyPassword=foo When passwords were stored in the keyring, their value is set to "-" in the configuration. This means that when running a synchronization without the --keyring argument, the password has to be entered interactively. The implementation introduces new virtual methods for password handling to properties and iterates over all properties to activate that special behavior. Another change is no longer to use cached strings in the SyncEvolutionConfig to store retrieved passwords. Instead, they are saved as the filter in the config node. This could help their users to get them but not flush to files. The main purpose of this change is to make it flexible to meet the requirements of dynamically added passwords in the backend.
2009-09-21 10:07:05 +02:00
// give a change to do something before flushing configs to files
to->preFlush(*to);
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
// done, now write it
to->flush();
syncevolution --migrate: copy .synthesis directory (MB #8048) Without copying this directory, using the migrated configuration will lead to a slow sync. SyncEvo::CmdlineTest::testMigrate was extended to cover this problem. Manually testing shows that the slow sync is indeed avoided with this change.
2009-11-25 15:30:07 +01:00
// also copy .synthesis dir?
if (m_migrate) {
string fromDir, toDir;
fromDir = from->getRootPath() + "/.synthesis";
toDir = to->getRootPath() + "/.synthesis";
if (isDir(fromDir)) {
cp_r(fromDir, toDir);
}
}
config handling: added removing of a configuration Only files which are known to belong to SyncEvolution are deleted. This list is hard-coded in FileConfigTree.cpp and includes the hidden .synthesis meta files in addition to the normal config files. The --remove command line option exposes this new feature. It removes the configuration *without* asking for confirmation. Log dirs are not removed yet. This needs to be done via a different API which will be added as part of access to old session information.
2009-04-15 14:02:09 +02:00
} else if (m_remove) {
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
if (m_dryrun) {
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
SyncContext::throwError("--dry-run not supported for removing configurations");
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
}
config handling: added removing of a configuration Only files which are known to belong to SyncEvolution are deleted. This list is hard-coded in FileConfigTree.cpp and includes the hidden .synthesis meta files in addition to the normal config files. The --remove command line option exposes this new feature. It removes the configuration *without* asking for confirmation. Log dirs are not removed yet. This needs to be done via a different API which will be added as part of access to old session information.
2009-04-15 14:02:09 +02:00
// extra sanity check
if (!m_sources.empty() ||
!m_syncProps.empty() ||
!m_sourceProps.empty()) {
usage(true, "too many parameters for --remove");
return false;
} else {
more classes renamed EvolutionSyncConfig => SyncConfig SyncEvolutionException => Exception EvolutionUnref* => Unref*
2009-10-06 17:22:47 +02:00
boost::shared_ptr<SyncConfig> config;
config.reset(new SyncConfig(m_server));
command line: removing non-existant configuration prints error (MB #6673) $ ./syncevolution --remove scheduleworld_foo_bar [ERROR] no such configuration: scheduleworld_foo_bar This helps the user to detect typos.
2009-10-13 22:08:33 +02:00
if (!config->exists()) {
SyncContext::throwError(string("no such configuration: ") + m_server);
}
config handling: added removing of a configuration Only files which are known to belong to SyncEvolution are deleted. This list is hard-coded in FileConfigTree.cpp and includes the hidden .synthesis meta files in addition to the normal config files. The --remove command line option exposes this new feature. It removes the configuration *without* asking for confirmation. Log dirs are not removed yet. This needs to be done via a different API which will be added as part of access to old session information.
2009-04-15 14:02:09 +02:00
config->remove();
return true;
}
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
} else if (m_accessItems) {
// need access to specific source
boost::shared_ptr<SyncContext> context;
context.reset(createSyncClient());
context->setOutput(&m_out);
string sourceName = *m_sources.begin();
command line import/export: failed for non-peer configs Besides the issue with reading the "type" (MBC #3157), operations like "--print-items @some-context addressbook" also failed because the TrackingSyncSource would try to write updated id/rev pairs into a DevNullConfigNode, set because there is no tracking node in @some-context. DevNullConfigNode does not allow writing. This patch uses a set of config nodes where the tracking node is volatile and allows writing. The additional benefit also for peer configs is that the SyncSource can no longer overwrite the change tracking for that peer accidentally because it never gets access to that node.
2010-06-16 13:51:50 +02:00
SyncSourceNodes sourceNodes = context->getSyncSourceNodesNoTracking(sourceName);
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
SyncSourceParams params(sourceName, sourceNodes);
cxxptr<SyncSource> source(SyncSource::createSource(params, true));
sysync::TSyError err;
#define CHECK_ERROR(_op) if (err) { SE_THROW_EXCEPTION_STATUS(StatusException, string(source->getName()) + ": " + (_op), SyncMLStatus(err)); }
source->open();
const SyncSource::Operations &ops = source->getOperations();
if (m_printItems) {
SyncSourceLogging *logging = dynamic_cast<SyncSourceLogging *>(source.get());
if (!ops.m_startDataRead ||
!ops.m_readNextItem) {
source->throwError("reading items not supported");
}
err = ops.m_startDataRead("", "");
CHECK_ERROR("reading items");
list<string> luids;
readLUIDs(source, luids);
BOOST_FOREACH(string &luid, luids) {
string description;
command line --print-items: added short description SyncSourceLogging has a functional getDescription() method in many backends that works with only the local data; use that to extract a description.
2010-06-11 18:50:03 +02:00
if (logging) {
description = logging->getDescription(luid);
if (!description.empty()) {
description.insert(0, ": ");
}
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
}
m_out << luid << description << std::endl;
}
command line: added --delete-items This complements --import/update/export/list-items. --remove was already taken, for deleting a configuration. Because the removal is done via the normal m_deleteItem operation, we inherit the normal SyncML behavior (logging, no error return code for non-existent items). It would have been nice to let the user choose how to handle that situation (like --force in rm), but that can't be done without SyncSource API changes.
2010-06-11 18:43:01 +02:00
} else if (m_deleteItems) {
if (!ops.m_deleteItem) {
source->throwError("deleting items not supported");
}
err = ops.m_startDataRead("", "");
CHECK_ERROR("reading items");
if (ops.m_endDataRead) {
err = ops.m_endDataRead();
CHECK_ERROR("stop reading items");
}
if (ops.m_startDataWrite) {
err = ops.m_startDataWrite();
CHECK_ERROR("writing items");
}
BOOST_FOREACH(const string &luid, m_luids) {
sysync::ItemIDType id;
command line: fix for new update/delete operations Local IDs were not decoded before invoking the underlying SyncSource operation. As a result, it failed with "item not found" errors for those items which really had special characters in their IDs.
2010-06-21 17:14:25 +02:00
string tmp = SafeConfigNode::unescape(luid);
id.item = (char *)tmp.c_str();
command line: added --delete-items This complements --import/update/export/list-items. --remove was already taken, for deleting a configuration. Because the removal is done via the normal m_deleteItem operation, we inherit the normal SyncML behavior (logging, no error return code for non-existent items). It would have been nice to let the user choose how to handle that situation (like --force in rm), but that can't be done without SyncSource API changes.
2010-06-11 18:43:01 +02:00
err = ops.m_deleteItem(&id);
CHECK_ERROR("deleting item");
}
// NO err = ops.m_endDataWrite(), see import/update below.
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
} else {
SyncSourceRaw *raw = dynamic_cast<SyncSourceRaw *>(source.get());
if (!raw) {
source->throwError("reading/writing items directly not supported");
}
if (m_import || m_update) {
err = ops.m_startDataRead("", "");
CHECK_ERROR("reading items");
if (ops.m_endDataRead) {
err = ops.m_endDataRead();
CHECK_ERROR("stop reading items");
}
if (ops.m_startDataWrite) {
err = ops.m_startDataWrite();
CHECK_ERROR("writing items");
}
cxxptr<ifstream> inFile;
if (m_itemPath =="-" ||
!isDir(m_itemPath)) {
string content;
string luid;
if (m_itemPath == "-") {
context->readStdin(content);
} else if (!ReadFile(m_itemPath, content)) {
SyncContext::throwError(m_itemPath, errno);
}
if (m_delimiter == "none") {
if (m_update) {
if (m_luids.size() != 1) {
SyncContext::throwError("need exactly one LUID parameter");
} else {
luid = *m_luids.begin();
}
}
m_out << "#0: "
<< insertItem(raw, luid, content)
<< endl;
} else {
typedef boost::split_iterator<string::iterator> string_split_iterator;
int count = 0;
// when updating, check number of luids in advance
if (m_update) {
unsigned long total = 0;
for (string_split_iterator it =
boost::make_split_iterator(content,
boost::first_finder(m_delimiter, boost::is_iequal()));
it != string_split_iterator();
++it) {
total++;
}
if (total != m_luids.size()) {
SyncContext::throwError(StringPrintf("%lu items != %lu luids, must match => aborting",
total, (unsigned long)m_luids.size()));
}
}
list<string>::const_iterator luidit = m_luids.begin();
for (string_split_iterator it =
boost::make_split_iterator(content,
boost::first_finder(m_delimiter, boost::is_iequal()));
it != string_split_iterator();
++it) {
m_out << "#" << count << ": ";
string luid;
if (m_update) {
if (luidit == m_luids.end()) {
// was checked above
SyncContext::throwError("internal error, not enough luids");
}
luid = *luidit;
++luidit;
}
m_out << insertItem(raw,
luid,
string(it->begin(), it->end()))
<< endl;
count++;
}
}
} else {
ReadDir dir(m_itemPath);
int count = 0;
BOOST_FOREACH(const string &entry, dir) {
string content;
string path = m_itemPath + "/" + entry;
m_out << count << ": " << entry << ": ";
if (!ReadFile(path, content)) {
SyncContext::throwError(path, errno);
}
m_out << insertItem(raw, "", content) << endl;
}
}
// NO err = ops.m_endDataWrite()! That's
// intentional. It ensures that for most (all?!)
// backends the change tracking isn't updated and thus
// future syncs see the imports/updates as changes
// made by the user.
} else if (m_export) {
err = ops.m_startDataRead("", "");
CHECK_ERROR("reading items");
ostream *out = NULL;
cxxptr<ofstream> outFile;
if (m_itemPath == "-") {
out = &m_out;
} else if(!isDir(m_itemPath)) {
outFile.set(new ofstream(m_itemPath.c_str()));
out = outFile;
}
if (m_luids.empty()) {
readLUIDs(source, m_luids);
}
bool haveItem = false; // have written one item
bool haveNewline = false; // that item had a newline at the end
try {
BOOST_FOREACH(const string &luid, m_luids) {
string item;
raw->readItemRaw(SafeConfigNode::unescape(luid), item);
if (!out) {
// write into directory
string fullPath = m_itemPath + "/" + luid;
ofstream file((m_itemPath + "/" + luid).c_str());
file << item;
file.close();
if (file.bad()) {
SyncContext::throwError(fullPath, errno);
}
} else {
if (haveItem) {
if (m_delimiter.size() > 1 &&
haveNewline &&
m_delimiter[0] == '\n') {
// already wrote initial newline, skip it
*out << m_delimiter.substr(1);
} else {
*out << m_delimiter;
}
}
*out << item;
haveNewline = !item.empty() && item[item.size() - 1] == '\n';
haveItem = true;
}
}
} catch (...) {
// ensure that we start following output on new line
if (m_itemPath == "-" && haveItem && !haveNewline) {
m_out << endl;
}
throw;
}
if (outFile) {
outFile->close();
if (outFile->bad()) {
SyncContext::throwError(m_itemPath, errno);
}
}
}
}
source->close();
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
} else {
sync source handling: implemented per-source property filtering, Cmdline uses it Selecting active sources during a sync or status check was done with a combination of setting a sync mode via a source config filter and setting a list of active sources. Now the SyncConfig supports a source filter which is applied to all sources and source filters for each source. The latter override the former. This is powerful enough to start syncs with full control over which sources are active in which mode, as described in the new D-Bus API. As part of this patch, the command line semantic is implemented entirely using a combination of different source filters.
2009-09-14 21:54:43 +02:00
std::set<std::string> unmatchedSources;
boost::shared_ptr<SyncContext> context;
context.reset(createSyncClient());
context->setQuiet(m_quiet);
context->setDryRun(m_dryrun);
context->setConfigFilter(true, "", m_syncProps);
Cmdline::run(): context should use same output as command line (MB #5041) For cases where the Cmdline instance uses a non-standard output stream, the SyncContext should also use that one. Needs to be tested as part of MB #5042/5043.
2010-03-26 10:17:32 +01:00
context->setOutput(&m_out);
sync source handling: implemented per-source property filtering, Cmdline uses it Selecting active sources during a sync or status check was done with a combination of setting a sync mode via a source config filter and setting a list of active sources. Now the SyncConfig supports a source filter which is applied to all sources and source filters for each source. The latter override the former. This is powerful enough to start syncs with full control over which sources are active in which mode, as described in the new D-Bus API. As part of this patch, the command line semantic is implemented entirely using a combination of different source filters.
2009-09-14 21:54:43 +02:00
if (m_sources.empty()) {
if (m_sourceProps.empty()) {
// empty source list, empty source filter => run with
// existing configuration without filtering it
} else {
// Special semantic of 'no source selected': apply
// filter only to sources which are
// *active*. Configuration of inactive sources is left
// unchanged. This way we don't activate sync sources
// accidentally when the sync mode is modified
// temporarily.
BOOST_FOREACH(const std::string &source,
context->getSyncSources()) {
boost::shared_ptr<PersistentSyncSourceConfig> source_config =
context->getSyncSourceConfig(source);
if (strcmp(source_config->getSync(), "disabled")) {
context->setConfigFilter(false, source, m_sourceProps);
}
}
}
} else {
// apply (possibly empty) source filter to selected sources
BOOST_FOREACH(const std::string &source,
m_sources) {
boost::shared_ptr<PersistentSyncSourceConfig> source_config =
context->getSyncSourceConfig(source);
if (!source_config || !source_config->exists()) {
// invalid source name in m_sources, remember and
// report this below
unmatchedSources.insert(source);
} else if (m_sourceProps.find(SyncSourceConfig::m_sourcePropSync.getName()) ==
m_sourceProps.end()) {
// Sync mode is not set, must override the
// "sync=disabled" set below with the original
// sync mode for the source or (if that is also
// "disabled") with "two-way". The latter is part
// of the command line semantic that listing a
// source activates it.
FilterConfigNode::ConfigFilter filter = m_sourceProps;
string sync = source_config->getSync();
filter[SyncSourceConfig::m_sourcePropSync.getName()] =
sync == "disabled" ? "two-way" : sync;
context->setConfigFilter(false, source, filter);
} else {
// sync mode is set, can use m_sourceProps
// directly to apply it
context->setConfigFilter(false, source, m_sourceProps);
}
}
// temporarily disable the rest
FilterConfigNode::ConfigFilter disabled;
disabled[SyncSourceConfig::m_sourcePropSync.getName()] = "disabled";
context->setConfigFilter(false, "", disabled);
}
// check whether there were any sources specified which do not exist
if (unmatchedSources.size()) {
context->throwError(string("no such source(s): ") + boost::join(unmatchedSources, " "));
}
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
if (m_status) {
sync source handling: implemented per-source property filtering, Cmdline uses it Selecting active sources during a sync or status check was done with a combination of setting a sync mode via a source config filter and setting a list of active sources. Now the SyncConfig supports a source filter which is applied to all sources and source filters for each source. The latter override the former. This is powerful enough to start syncs with full control over which sources are active in which mode, as described in the new D-Bus API. As part of this patch, the command line semantic is implemented entirely using a combination of different source filters.
2009-09-14 21:54:43 +02:00
context->status();
log handling: added access to previous sync sessions EvolutionSyncClient::getSessions() returns the full pathnames of the log dirs of the previous sync sessions. --print-sessions shows those on the command line. Right now this feature is only useful to remove these directories or (for the user) to find backup files. Further information has to written to and extracted from them to make --print-sessions and its graphical pendant more useful.
2009-04-15 15:58:05 +02:00
} else if (m_printSessions) {
vector<string> dirs;
sync source handling: implemented per-source property filtering, Cmdline uses it Selecting active sources during a sync or status check was done with a combination of setting a sync mode via a source config filter and setting a list of active sources. Now the SyncConfig supports a source filter which is applied to all sources and source filters for each source. The latter override the former. This is powerful enough to start syncs with full control over which sources are active in which mode, as described in the new D-Bus API. As part of this patch, the command line semantic is implemented entirely using a combination of different source filters.
2009-09-14 21:54:43 +02:00
context->getSessions(dirs);
log handling: SyncReport is stored in log dir and can be retrieved Source specific values are stored with a key that is composed of the prefix "source-" followed by the source name (with - escaped) and other components representing the various properties. Empty values are not stored, which is fairly efficient for the statistics array. When reading back, only sources which have any value are recreated. Because the sync mode is always set, this ensures that the set of active sources is restored. Inactive sources are not part of that set! --print-sessions now prints the retrieved sync report just like --sync does. Combined with --quiet only the directory name is printed.
2009-04-16 12:07:49 +02:00
bool first = true;
log handling: added access to previous sync sessions EvolutionSyncClient::getSessions() returns the full pathnames of the log dirs of the previous sync sessions. --print-sessions shows those on the command line. Right now this feature is only useful to remove these directories or (for the user) to find backup files. Further information has to written to and extracted from them to make --print-sessions and its graphical pendant more useful.
2009-04-15 15:58:05 +02:00
BOOST_FOREACH(const string &dir, dirs) {
log handling: SyncReport is stored in log dir and can be retrieved Source specific values are stored with a key that is composed of the prefix "source-" followed by the source name (with - escaped) and other components representing the various properties. Empty values are not stored, which is fairly efficient for the statistics array. When reading back, only sources which have any value are recreated. Because the sync mode is always set, this ensures that the set of active sources is restored. Inactive sources are not part of that set! --print-sessions now prints the retrieved sync report just like --sync does. Combined with --quiet only the directory name is printed.
2009-04-16 12:07:49 +02:00
if (first) {
first = false;
} else if(!m_quiet) {
stdout: never write to std::cout directly (MB #5041) Cmdline and SyncContext standard output should always go through a variable that can be pointed towards the actual output channel. Cmdline already had m_out for that, but it wasn't used everywhere. Fixed. SyncContext now has setOutput()/getOutput() and that is used in SyncContext.cpp. It is not set anywhere yet. To catch incorrect use of cout or cerr inside SyncEvolution, the SyncEvo namespace defines its own cout and cerr which cannot be used like std::cout/cerr, thus triggering compiler errors. Use "std::cout/cerr" when necessary.
2010-03-26 09:53:27 +01:00
m_out << endl;
log handling: SyncReport is stored in log dir and can be retrieved Source specific values are stored with a key that is composed of the prefix "source-" followed by the source name (with - escaped) and other components representing the various properties. Empty values are not stored, which is fairly efficient for the statistics array. When reading back, only sources which have any value are recreated. Because the sync mode is always set, this ensures that the set of active sources is restored. Inactive sources are not part of that set! --print-sessions now prints the retrieved sync report just like --sync does. Combined with --quiet only the directory name is printed.
2009-04-16 12:07:49 +02:00
}
stdout: never write to std::cout directly (MB #5041) Cmdline and SyncContext standard output should always go through a variable that can be pointed towards the actual output channel. Cmdline already had m_out for that, but it wasn't used everywhere. Fixed. SyncContext now has setOutput()/getOutput() and that is used in SyncContext.cpp. It is not set anywhere yet. To catch incorrect use of cout or cerr inside SyncEvolution, the SyncEvo namespace defines its own cout and cerr which cannot be used like std::cout/cerr, thus triggering compiler errors. Use "std::cout/cerr" when necessary.
2010-03-26 09:53:27 +01:00
m_out << dir << endl;
log handling: SyncReport is stored in log dir and can be retrieved Source specific values are stored with a key that is composed of the prefix "source-" followed by the source name (with - escaped) and other components representing the various properties. Empty values are not stored, which is fairly efficient for the statistics array. When reading back, only sources which have any value are recreated. Because the sync mode is always set, this ensures that the set of active sources is restored. Inactive sources are not part of that set! --print-sessions now prints the retrieved sync report just like --sync does. Combined with --quiet only the directory name is printed.
2009-04-16 12:07:49 +02:00
if (!m_quiet) {
SyncReport report;
sync source handling: implemented per-source property filtering, Cmdline uses it Selecting active sources during a sync or status check was done with a combination of setting a sync mode via a source config filter and setting a list of active sources. Now the SyncConfig supports a source filter which is applied to all sources and source filters for each source. The latter override the former. This is powerful enough to start syncs with full control over which sources are active in which mode, as described in the new D-Bus API. As part of this patch, the command line semantic is implemented entirely using a combination of different source filters.
2009-09-14 21:54:43 +02:00
context->readSessionInfo(dir, report);
stdout: never write to std::cout directly (MB #5041) Cmdline and SyncContext standard output should always go through a variable that can be pointed towards the actual output channel. Cmdline already had m_out for that, but it wasn't used everywhere. Fixed. SyncContext now has setOutput()/getOutput() and that is used in SyncContext.cpp. It is not set anywhere yet. To catch incorrect use of cout or cerr inside SyncEvolution, the SyncEvo namespace defines its own cout and cerr which cannot be used like std::cout/cerr, thus triggering compiler errors. Use "std::cout/cerr" when necessary.
2010-03-26 09:53:27 +01:00
m_out << report;
log handling: SyncReport is stored in log dir and can be retrieved Source specific values are stored with a key that is composed of the prefix "source-" followed by the source name (with - escaped) and other components representing the various properties. Empty values are not stored, which is fairly efficient for the statistics array. When reading back, only sources which have any value are recreated. Because the sync mode is always set, this ensures that the set of active sources is restored. Inactive sources are not part of that set! --print-sessions now prints the retrieved sync report just like --sync does. Combined with --quiet only the directory name is printed.
2009-04-16 12:07:49 +02:00
}
log handling: added access to previous sync sessions EvolutionSyncClient::getSessions() returns the full pathnames of the log dirs of the previous sync sessions. --print-sessions shows those on the command line. Right now this feature is only useful to remove these directories or (for the user) to find backup files. Further information has to written to and extracted from them to make --print-sessions and its graphical pendant more useful.
2009-04-15 15:58:05 +02:00
}
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
} else if (!m_restore.empty()) {
// sanity checks: either --after or --before must be given, sources must be selected
if ((!m_after && !m_before) ||
(m_after && m_before)) {
usage(false, "--restore <log dir> must be used with either --after (restore database as it was after that sync) or --before (restore data from before sync)");
return false;
}
if (m_sources.empty()) {
usage(false, "Sources must be selected explicitly for --restore to prevent accidental restore.");
return false;
}
sync source handling: implemented per-source property filtering, Cmdline uses it Selecting active sources during a sync or status check was done with a combination of setting a sync mode via a source config filter and setting a list of active sources. Now the SyncConfig supports a source filter which is applied to all sources and source filters for each source. The latter override the former. This is powerful enough to start syncs with full control over which sources are active in which mode, as described in the new D-Bus API. As part of this patch, the command line semantic is implemented entirely using a combination of different source filters.
2009-09-14 21:54:43 +02:00
context->restore(m_restore,
m_after ?
SyncContext::DATABASE_AFTER_SYNC :
SyncContext::DATABASE_BEFORE_SYNC);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
} else {
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
if (m_dryrun) {
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
SyncContext::throwError("--dry-run not supported for running a synchronization");
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
}
improved command line and its documented More explicit description of the different operations. Added --run|-r switch: if properties are on the command line, then an operation must be specified explicitly, with --run standing for a sync run. Otherwise a sync might accidentally be done when a configuration change was intented, but the --configure switch was forgotten. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@696 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-08-03 15:00:51 +02:00
// safety catch: if props are given, then --run
// is required
if (!m_run &&
(m_syncProps.size() || m_sourceProps.size())) {
usage(false, "Properties specified, but neither '--configure' nor '--run' - what did you want?");
return false;
}
sync source handling: implemented per-source property filtering, Cmdline uses it Selecting active sources during a sync or status check was done with a combination of setting a sync mode via a source config filter and setting a list of active sources. Now the SyncConfig supports a source filter which is applied to all sources and source filters for each source. The latter override the former. This is powerful enough to start syncs with full control over which sources are active in which mode, as described in the new D-Bus API. As part of this patch, the command line semantic is implemented entirely using a combination of different source filters.
2009-09-14 21:54:43 +02:00
return (context->sync() == STATUS_OK);
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
}
return true;
}
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
void Cmdline::readLUIDs(SyncSource *source, list<string> &luids)
{
const SyncSource::Operations &ops = source->getOperations();
sysync::ItemIDType id;
sysync::sInt32 status;
sysync::TSyError err = ops.m_readNextItem(&id, &status, true);
CHECK_ERROR("next item");
while (status != sysync::ReadNextItem_EOF) {
luids.push_back(SafeConfigNode::escape(id.item, true, true));
err = ops.m_readNextItem(&id, &status, false);
CHECK_ERROR("next item");
}
}
string Cmdline::insertItem(SyncSourceRaw *source, const string &luid, const string &data)
{
command line: fix for new update/delete operations Local IDs were not decoded before invoking the underlying SyncSource operation. As a result, it failed with "item not found" errors for those items which really had special characters in their IDs.
2010-06-21 17:14:25 +02:00
SyncSourceRaw::InsertItemResult res = source->insertItemRaw(SafeConfigNode::unescape(luid), data);
command line: added --import/export/update/print-items These operations provide a common interface for manipulating databases supported by SyncEvolution backends. --delete is an obvious gap, will be added soon. The implementation supports all backends offering the SyncSourceRaw interface (Evolution, Akonadi, XMLRPC, file, but not the sqlite demo), which is determined via a dynamic cast. SyncSourceRaw is necessary because it allows direct access to the native data format without going through the Synthesis engine. Because the Synthesis engine is not activated, printing the short description (SyncSourceLogging interface) mentioned in the README.rst is not possible yet. Having such an engine would be useful also for importing/exporting data which is not in the native format of the backend; additional command line parameters will be needed to define that format. All of the new operations have in common that they work on one source and multiple LUIDs. This is a slight deviation from other command line synopsises where all arguments after the server config refer to sources. The new m_accessItems flag is set for these special operations; they also share much of the setup code. The implementation itself tries to use the generic SyncSource interface as much as possible. It only falls back to more specialized implementations where necessary (SyncSourceRaw). The builtin synopsis and usage was intentionally not updated. The expection is that before this patch lands in the "master" branch, the builtin text will come directly from README.rst (which was updated).
2010-06-11 15:44:57 +02:00
return res.m_luid;
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
string Cmdline::cmdOpt(const char *opt, const char *param)
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
{
string res = "'";
res += opt;
if (param) {
res += " ";
res += param;
}
res += "'";
return res;
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
bool Cmdline::parseProp(const ConfigPropertyRegistry &validProps,
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
FilterConfigNode::ConfigFilter &props,
const char *opt,
const char *param,
const char *propname)
{
if (!param) {
usage(true, string("missing parameter for ") + cmdOpt(opt, param));
return false;
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
} else if (boost::trim_copy(string(param)) == "?") {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_dontrun = true;
if (propname) {
return listPropValues(validProps, propname, opt);
} else {
return listProperties(validProps, opt);
}
} else {
string propstr;
string paramstr;
if (propname) {
propstr = propname;
paramstr = param;
} else {
const char *equal = strchr(param, '=');
if (!equal) {
usage(true, string("the '=<value>' part is missing in: ") + cmdOpt(opt, param));
return false;
}
propstr.assign(param, equal - param);
paramstr.assign(equal + 1);
}
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
boost::trim(propstr);
boost::trim_left(paramstr);
if (boost::trim_copy(paramstr) == "?") {
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
m_dontrun = true;
return listPropValues(validProps, propstr, cmdOpt(opt, param));
} else {
const ConfigProperty *prop = validProps.find(propstr);
if (!prop) {
minor fixes for command line handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@525 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 13:11:39 +01:00
m_err << "ERROR: " << cmdOpt(opt, param) << ": no such property" << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
return false;
} else {
string error;
if (!prop->checkValue(paramstr, error)) {
m_err << "ERROR: " << cmdOpt(opt, param) << ": " << error << endl;
return false;
} else {
simplified ConfigFilter: now uses keys as they are, but with case-insensitive compare git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@656 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-06 22:26:23 +02:00
props[propstr] = paramstr;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
return true;
}
}
}
}
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
bool Cmdline::listPropValues(const ConfigPropertyRegistry &validProps,
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
const string &propName,
const string &opt)
{
const ConfigProperty *prop = validProps.find(propName);
if (!prop) {
minor fixes for command line handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@525 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 13:11:39 +01:00
m_err << "ERROR: "<< opt << ": no such property" << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
return false;
} else {
m_out << opt << endl;
string comment = prop->getComment();
if (comment != "") {
list<string> commentLines;
ConfigProperty::splitComment(comment, commentLines);
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
BOOST_FOREACH(const string &line, commentLines) {
m_out << " " << line << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
} else {
added missing newline in error message git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@561 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-01 21:44:12 +02:00
m_out << " no documentation available" << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
return true;
}
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
bool Cmdline::listProperties(const ConfigPropertyRegistry &validProps,
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
const string &opt)
{
// The first of several related properties has a comment.
// Remember that comment and print it as late as possible,
// that way related properties preceed their comment.
string comment;
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
BOOST_FOREACH(const ConfigProperty *prop, validProps) {
if (!prop->isHidden()) {
string newComment = prop->getComment();
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
if (newComment != "") {
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
if (!comment.empty()) {
dumpComment(m_out, " ", comment);
m_out << endl;
}
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
comment = newComment;
}
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
m_out << prop->getName() << ":" << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
}
dumpComment(m_out, " ", comment);
return true;
}
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
void Cmdline::getFilters(const string &context,
ConfigProps &syncFilter,
map<string, ConfigProps> &sourceFilters)
{
shared config + templates: must share global properties The command line did not read the global defaultPeer property unless it found an existing context. This is an unnecessary optimization, reading from an non-existant context yields no properties except perhaps for defaultPeer, so we can do that and get the correct result with and without a context. Jussi mentioned that he didn't get defaultPeer when reading a template. Added TestMultipleConfigs.testSharedTemplate to verify that, but it passes without issues.
2009-12-01 18:03:39 +01:00
// Read from context. If it does not exist, we simply set no properties
// as filter. Previously there was a check for existance, but that was
// flawed because it ignored the global property "defaultPeer".
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
boost::shared_ptr<SyncConfig> shared(new SyncConfig(string("@") + context));
shared config + templates: must share global properties The command line did not read the global defaultPeer property unless it found an existing context. This is an unnecessary optimization, reading from an non-existant context yields no properties except perhaps for defaultPeer, so we can do that and get the correct result with and without a context. Jussi mentioned that he didn't get defaultPeer when reading a template. Added TestMultipleConfigs.testSharedTemplate to verify that, but it passes without issues.
2009-12-01 18:03:39 +01:00
shared->getProperties()->readProperties(syncFilter);
BOOST_FOREACH(StringPair entry, m_syncProps) {
syncFilter[entry.first] = entry.second;
}
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
shared config + templates: must share global properties The command line did not read the global defaultPeer property unless it found an existing context. This is an unnecessary optimization, reading from an non-existant context yields no properties except perhaps for defaultPeer, so we can do that and get the correct result with and without a context. Jussi mentioned that he didn't get defaultPeer when reading a template. Added TestMultipleConfigs.testSharedTemplate to verify that, but it passes without issues.
2009-12-01 18:03:39 +01:00
BOOST_FOREACH(std::string source, shared->getSyncSources()) {
SyncSourceNodes nodes = shared->getSyncSourceNodes(source, "");
ConfigProps &props = sourceFilters[source];
nodes.getProperties()->readProperties(props);
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
shared config + templates: must share global properties The command line did not read the global defaultPeer property unless it found an existing context. This is an unnecessary optimization, reading from an non-existant context yields no properties except perhaps for defaultPeer, so we can do that and get the correct result with and without a context. Jussi mentioned that he didn't get defaultPeer when reading a template. Added TestMultipleConfigs.testSharedTemplate to verify that, but it passes without issues.
2009-12-01 18:03:39 +01:00
// Special case "type" property: the value in the context
// is not preserved. Every new peer must ensure that
// its own value is compatible (= same backend) with
// the other peers.
props.erase("type");
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
shared config + templates: must share global properties The command line did not read the global defaultPeer property unless it found an existing context. This is an unnecessary optimization, reading from an non-existant context yields no properties except perhaps for defaultPeer, so we can do that and get the correct result with and without a context. Jussi mentioned that he didn't get defaultPeer when reading a template. Added TestMultipleConfigs.testSharedTemplate to verify that, but it passes without issues.
2009-12-01 18:03:39 +01:00
BOOST_FOREACH(StringPair entry, m_sourceProps) {
props[entry.first] = entry.second;
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
}
}
shared config + templates: must share global properties The command line did not read the global defaultPeer property unless it found an existing context. This is an unnecessary optimization, reading from an non-existant context yields no properties except perhaps for defaultPeer, so we can do that and get the correct result with and without a context. Jussi mentioned that he didn't get defaultPeer when reading a template. Added TestMultipleConfigs.testSharedTemplate to verify that, but it passes without issues.
2009-12-01 18:03:39 +01:00
sourceFilters[""] = m_sourceProps;
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
}
command line: check for invalid peer properties (MB #8048) When using --print-config or --configure with a context as config name (like @default), then per-peer properties do not have any effect. This might be confusing for users, so the command line should better report this as error and refuse to execute the operation. As usual, "type" is a bit special because it is both shared and per-peer.
2010-03-03 16:20:53 +01:00
static void findPeerProps(FilterConfigNode::ConfigFilter &filter,
ConfigPropertyRegistry &registry,
list<string> &peerProps)
{
BOOST_FOREACH(StringPair entry, filter) {
const ConfigProperty *prop = registry.find(entry.first);
if (prop &&
prop->getSharing() == ConfigProperty::NO_SHARING &&
!(prop->getFlags() & ConfigProperty::SHARED_AND_UNSHARED)) {
peerProps.push_back(entry.first);
}
}
}
void Cmdline::checkForPeerProps()
{
list<string> peerProps;
findPeerProps(m_syncProps, SyncConfig::getRegistry(), peerProps);
findPeerProps(m_sourceProps, SyncSourceConfig::getRegistry(), peerProps);
if (!peerProps.empty()) {
SyncContext::throwError(string("per-peer (unshared) properties not allowed: ") +
boost::join(peerProps, ", "));
}
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
void Cmdline::listSources(SyncSource &syncSource, const string &header)
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
{
m_out << header << ":\n";
redesigned SyncSource base class + API The main motivation for this change is that it allows the implementor of a backend to choose the implementations for the different aspects of a datasource (change tracking, item import/export, logging, ...) independently of each other. For example, change tracking via revision strings can now be combined with exchanging data with the Synthesis engine via a single string (the traditional method in SyncEvolution) and with direct access to the Synthesis field list (now possible for the first time). The new backend API is based on the concept of providing implementations for certain functionality via function objects instead of implementing certain virtual methods. The advantage is that implementors can define their own, custom interfaces and mix and match implementations of the different groups of functionality. Logging (see SyncSourceLogging in a later commit) can be done by wrapping some arbitrary other item import/export function objects (decorator design pattern). The class hierarchy is now this: - SyncSourceBase: interface for common utility code, all other classes are derived from it and thus can use that code - SyncSource: base class which implements SyncSourceBase and hooks a datasource into the SyncEvolution core; its "struct Operations" holds the function objects which can be implemented in different ways - TestingSyncSource: combines some of the following classes into an interface that is expected by the client-test program; backends only have to derive from (and implement this) if they want to use the automated testing - TrackingSyncSource: provides the same functionality as before (change tracking via revision strings, item import/export as string) in a single interface; the description of the pure virtual methods are duplicated so that developers can go through this class and find everything they need to know to implement it The following classes contain the code that was previously found in the EvolutionSyncSource base class. Implementors can derive from them and call the init() methods to inherit and activate the functionality: - SyncSourceSession: binds Synthesis session callbacks to virtual methods beginSync(), endSync() - SyncSourceChanges: implements Synthesis item tracking callbacks with set of LUIDs that the user of the class has to fill - SyncSourceDelete: binds Synthesis delete callback to virtual method - SyncSourceRaw: read and write items in the backends format, used for testing and backup/restore - SyncSourceSerialize: exchanges items with Synthesis engine using a string representation of the data; this is how EvolutionSyncSource has traditionally worked, so much of the same virtual methods are now in this class - SyncSourceRevisions: utility class which does change tracking via some kind of "revision" string which changes each time an item is modified; this code was previously in the TrackingSyncSource
2009-08-25 09:27:46 +02:00
SyncSource::Databases databases = syncSource.getDatabases();
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
redesigned SyncSource base class + API The main motivation for this change is that it allows the implementor of a backend to choose the implementations for the different aspects of a datasource (change tracking, item import/export, logging, ...) independently of each other. For example, change tracking via revision strings can now be combined with exchanging data with the Synthesis engine via a single string (the traditional method in SyncEvolution) and with direct access to the Synthesis field list (now possible for the first time). The new backend API is based on the concept of providing implementations for certain functionality via function objects instead of implementing certain virtual methods. The advantage is that implementors can define their own, custom interfaces and mix and match implementations of the different groups of functionality. Logging (see SyncSourceLogging in a later commit) can be done by wrapping some arbitrary other item import/export function objects (decorator design pattern). The class hierarchy is now this: - SyncSourceBase: interface for common utility code, all other classes are derived from it and thus can use that code - SyncSource: base class which implements SyncSourceBase and hooks a datasource into the SyncEvolution core; its "struct Operations" holds the function objects which can be implemented in different ways - TestingSyncSource: combines some of the following classes into an interface that is expected by the client-test program; backends only have to derive from (and implement this) if they want to use the automated testing - TrackingSyncSource: provides the same functionality as before (change tracking via revision strings, item import/export as string) in a single interface; the description of the pure virtual methods are duplicated so that developers can go through this class and find everything they need to know to implement it The following classes contain the code that was previously found in the EvolutionSyncSource base class. Implementors can derive from them and call the init() methods to inherit and activate the functionality: - SyncSourceSession: binds Synthesis session callbacks to virtual methods beginSync(), endSync() - SyncSourceChanges: implements Synthesis item tracking callbacks with set of LUIDs that the user of the class has to fill - SyncSourceDelete: binds Synthesis delete callback to virtual method - SyncSourceRaw: read and write items in the backends format, used for testing and backup/restore - SyncSourceSerialize: exchanges items with Synthesis engine using a string representation of the data; this is how EvolutionSyncSource has traditionally worked, so much of the same virtual methods are now in this class - SyncSourceRevisions: utility class which does change tracking via some kind of "revision" string which changes each time an item is modified; this code was previously in the TrackingSyncSource
2009-08-25 09:27:46 +02:00
BOOST_FOREACH(const SyncSource::Database &database, databases) {
cleaned up EvolutionSyncSource interface+implementation git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@664 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-10 21:17:42 +02:00
m_out << " " << database.m_name << " (" << database.m_uri << ")";
if (database.m_isDefault) {
simplified configuration - deviceId is at the end of the config now because it hardly ever has to be set - evolutionsource can be left unset in most cases: the first Evolution database is the default - mark default data source when listing them git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@558 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 21:08:19 +02:00
m_out << " <default>";
}
m_out << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
}
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
void Cmdline::dumpConfigs(const string &preamble,
const SyncConfig::ConfigList &servers)
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
{
m_out << preamble << endl;
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
BOOST_FOREACH(const SyncConfig::ConfigList::value_type &server,servers) {
m_out << " " << server.first << " = " << server.second <<endl;
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
}
if (!servers.size()) {
m_out << " none" << endl;
}
}
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
void Cmdline::dumpConfigTemplates(const string &preamble,
const SyncConfig::TemplateList &templates,
bool printRank)
{
m_out << preamble << endl;
Cmdline Test: testMatchTemplate fix (MB#10004) The syncevolutionclient template was changed to only match with 'syncevolutionclient' exactly to avoid collision with the server side template 'syncevolution'. The cmdline test has to adapt accordingly, added a header line to describe the meaning of each output column.
2010-03-03 09:24:58 +01:00
m_out << " " << "template name" << " = " << "template description";
if (printRank) {
MatchTemplate better output (MB#10004) Instead of a 1-5 valued rank, convert to percentage. ./syncevolution -l \?nokia_7210c Available configuration templates: template name = template description matching score in percent (100% = exact match) Nokia_7210c = Template for Nokia 7210c phone and as default template for all Nokia phones 100% SyncEvolutionClient = SyncEvolution server side template 40% ServerDefault = server side default template 20%
2010-03-04 03:26:03 +01:00
m_out << " " << "matching score in percent (100% = exact match)";
Cmdline Test: testMatchTemplate fix (MB#10004) The syncevolutionclient template was changed to only match with 'syncevolutionclient' exactly to avoid collision with the server side template 'syncevolution'. The cmdline test has to adapt accordingly, added a header line to describe the meaning of each output column.
2010-03-03 09:24:58 +01:00
}
m_out << endl;
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
BOOST_FOREACH(const SyncConfig::TemplateList::value_type server,templates) {
Config Templates: Add 'templateName' optional property (MB10406) 'templateName' property is introduced to present the 'device class' for a matching device. The property is optional. There might be templates which only match one specific model or where the class is so obscure that showing it is useless. There are 'template id' and 'template name' concepts inside the core, 'template id' is the handle of the template, user can later use it to uniquely retrieve the corresponding template, 'template name' on the other hand is an explict property 'templateName' inside the template metadata, describing the class of the devices for UI. 'template id' is only used by Cmdline and it use the first entry of the fingerprint.
2010-03-25 04:26:13 +01:00
m_out << " " << server->m_templateId << " = " << server->m_description;
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
if (printRank){
MatchTemplate better output (MB#10004) Instead of a 1-5 valued rank, convert to percentage. ./syncevolution -l \?nokia_7210c Available configuration templates: template name = template description matching score in percent (100% = exact match) Nokia_7210c = Template for Nokia 7210c phone and as default template for all Nokia phones 100% SyncEvolutionClient = SyncEvolution server side template 40% ServerDefault = server side default template 20%
2010-03-04 03:26:03 +01:00
m_out << " " << server->m_rank *20 << "%";
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
}
m_out << endl;
}
if (!templates.size()) {
m_out << " none" << endl;
}
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
void Cmdline::dumpProperties(const ConfigNode &configuredProps,
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
const ConfigPropertyRegistry &allProps,
command line: --print-config of context should ignore per-peer properties (MB #8048) When printing the context config, per-peer properties where shown with their default values, which was a bit confusing. Now they are skipped.
2010-03-02 16:27:22 +01:00
int flags)
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
{
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
list<string> perPeer, perContext, global;
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
BOOST_FOREACH(const ConfigProperty *prop, allProps) {
command line: --print-config of context should ignore per-peer properties (MB #8048) When printing the context config, per-peer properties where shown with their default values, which was a bit confusing. Now they are skipped.
2010-03-02 16:27:22 +01:00
if (prop->isHidden() ||
((flags & HIDE_PER_PEER) &&
prop->getSharing() == ConfigProperty::NO_SHARING &&
!(prop->getFlags() & ConfigProperty::SHARED_AND_UNSHARED))) {
implemented --migrate and --configure git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@528 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 17:43:31 +01:00
continue;
}
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
if (!m_quiet) {
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
string comment = prop->getComment();
avoid printing empty comments in --print-config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@527 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 15:38:19 +01:00
if (!comment.empty()) {
m_out << endl;
dumpComment(m_out, "# ", comment);
}
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
bool isDefault;
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
prop->getProperty(configuredProps, &isDefault);
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
if (isDefault) {
m_out << "# ";
}
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
m_out << prop->getName() << " = " << prop->getProperty(configuredProps) << endl;
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
list<string> *type = NULL;
switch (prop->getSharing()) {
case ConfigProperty::GLOBAL_SHARING:
type = &global;
break;
case ConfigProperty::SOURCE_SET_SHARING:
type = &perContext;
break;
case ConfigProperty::NO_SHARING:
type = &perPeer;
break;
}
if (type) {
type->push_back(prop->getName());
}
}
command line: --print-config of context should ignore per-peer properties (MB #8048) When printing the context config, per-peer properties where shown with their default values, which was a bit confusing. Now they are skipped.
2010-03-02 16:27:22 +01:00
if (!m_quiet && !(flags & HIDE_LEGEND)) {
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
if (!perPeer.empty() ||
!perContext.empty() ||
!global.empty()) {
m_out << endl;
}
if (!perPeer.empty()) {
m_out << "# per-peer (unshared) properties: " << boost::join(perPeer, ", ") << endl;
}
if (!perContext.empty()) {
m_out << "# shared by peers in same context: " << boost::join(perContext, ", ") << endl;
}
if (!global.empty()) {
m_out << "# global properties: " << boost::join(global, ", ") << endl;
}
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
void Cmdline::dumpComment(ostream &stream,
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
const string &prefix,
const string &comment)
{
list<string> commentLines;
ConfigProperty::splitComment(comment, commentLines);
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
BOOST_FOREACH(const string &line, commentLines) {
stream << prefix << line << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
}
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
void Cmdline::usage(bool full, const string &error, const string &param)
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
{
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
ostream &out(error.empty() ? m_out : m_err);
improved command line and its documented More explicit description of the different operations. Added --run|-r switch: if properties are on the command line, then an operation must be specified explicitly, with --run standing for a sync run. Otherwise a sync might accidentally be done when a configuration change was intented, but the --configure switch was forgotten. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@696 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-08-03 15:00:51 +02:00
out << "Show available sources:" << endl;
out << " " << m_argv[0] << endl;
log handling: added access to previous sync sessions EvolutionSyncClient::getSessions() returns the full pathnames of the log dirs of the previous sync sessions. --print-sessions shows those on the command line. Right now this feature is only useful to remove these directories or (for the user) to find backup files. Further information has to written to and extracted from them to make --print-sessions and its graphical pendant more useful.
2009-04-15 15:58:05 +02:00
out << "Show information about configuration(s) and sync sessions:" << endl;
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
out << " " << m_argv[0] << " --print-servers|--print-configs|--print-peers" << endl;
out << " " << m_argv[0] << " --print-config [--quiet] <config> [main|<source ...]" << endl;
out << " " << m_argv[0] << " --print-sessions [--quiet] <config>" << endl;
improved command line and its documented More explicit description of the different operations. Added --run|-r switch: if properties are on the command line, then an operation must be specified explicitly, with --run standing for a sync run. Otherwise a sync might accidentally be done when a configuration change was intented, but the --configure switch was forgotten. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@696 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-08-03 15:00:51 +02:00
out << "Show information about SyncEvolution:" << endl;
out << " " << m_argv[0] << " --help|-h" << endl;
out << " " << m_argv[0] << " --version" << endl;
out << "Run a synchronization:" << endl;
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
out << " " << m_argv[0] << " <config> [<source> ...]" << endl;
out << " " << m_argv[0] << " --run <options for run> <config> [<source> ...]" << endl;
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
out << "Restore data from the automatic backups:" << endl;
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
out << " " << m_argv[0] << " --restore <session directory> --before|--after [--dry-run] <config> <source> ..." << endl;
config handling: added removing of a configuration Only files which are known to belong to SyncEvolution are deleted. This list is hard-coded in FileConfigTree.cpp and includes the hidden .synthesis meta files in addition to the normal config files. The --remove command line option exposes this new feature. It removes the configuration *without* asking for confirmation. Log dirs are not removed yet. This needs to be done via a different API which will be added as part of access to old session information.
2009-04-15 14:02:09 +02:00
out << "Remove a configuration:" << endl;
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
out << " " << m_argv[0] << " --remove <config>" << endl;
improved command line and its documented More explicit description of the different operations. Added --run|-r switch: if properties are on the command line, then an operation must be specified explicitly, with --run standing for a sync run. Otherwise a sync might accidentally be done when a configuration change was intented, but the --configure switch was forgotten. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@696 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-08-03 15:00:51 +02:00
out << "Modify configuration:" << endl;
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
out << " " << m_argv[0] << " --configure <options for configuration> <config> [<source> ...]" << endl;
out << " " << m_argv[0] << " --migrate <config>" << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
if (full) {
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
out << endl <<
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
"Options:" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"--sync|-s <mode>" << endl <<
"--sync|-s ?" << endl <<
" Temporarily synchronize the active sources in that mode. Useful" << endl <<
" for a \"refresh-from-server\" or \"refresh-from-client\" sync which" << endl <<
" clears all data at one end and copies all items from the other." << endl <<
"" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
"--print-servers|--print-configs|--print-peers" << endl <<
" Prints the names of all configured peers to stdout." << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"" << endl <<
"--print-config|-p" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" Prints the complete configuration for the selected peer" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" to stdout, including up-to-date comments for all properties. The" << endl <<
" format is the normal .ini format with source configurations in" << endl <<
" different sections introduced with [<source>] lines. Can be combined" << endl <<
" with --sync-property and --source-property to modify the configuration" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" on-the-fly. When one or more sources are listed after the <config>" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" name on the command line, then only the configs of those sources are" << endl <<
" printed. Using --quiet suppresses the comments for each property." << endl <<
" When setting a --template, then the reference configuration for" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" that peer is printed instead of an existing configuration." << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"" << endl <<
log handling: added access to previous sync sessions EvolutionSyncClient::getSessions() returns the full pathnames of the log dirs of the previous sync sessions. --print-sessions shows those on the command line. Right now this feature is only useful to remove these directories or (for the user) to find backup files. Further information has to written to and extracted from them to make --print-sessions and its graphical pendant more useful.
2009-04-15 15:58:05 +02:00
"--print-sessions" << endl <<
log handling: SyncReport is stored in log dir and can be retrieved Source specific values are stored with a key that is composed of the prefix "source-" followed by the source name (with - escaped) and other components representing the various properties. Empty values are not stored, which is fairly efficient for the statistics array. When reading back, only sources which have any value are recreated. Because the sync mode is always set, this ensures that the set of active sources is restored. Inactive sources are not part of that set! --print-sessions now prints the retrieved sync report just like --sync does. Combined with --quiet only the directory name is printed.
2009-04-16 12:07:49 +02:00
" Prints a list of all previous log directories. Unless --quiet is used, each" << endl <<
" file name is followed by the original sync report." << endl <<
log handling: added access to previous sync sessions EvolutionSyncClient::getSessions() returns the full pathnames of the log dirs of the previous sync sessions. --print-sessions shows those on the command line. Right now this feature is only useful to remove these directories or (for the user) to find backup files. Further information has to written to and extracted from them to make --print-sessions and its graphical pendant more useful.
2009-04-15 15:58:05 +02:00
"" << endl <<
syncevolution --help: fixed typos (MB#6450) "--configure" contained a long dash instead of two dashes. "--sync-option" -> "--sync-property"
2009-09-28 11:55:24 +02:00
"--configure|-c" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" Modify the configuration files for the selected peer. If no such" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" configuration exists, then a new one is created using one of the" << endl <<
" template configurations (see --template option). When creating" << endl <<
" a new configuration only the active sources will be set to active" << endl <<
" in the new configuration, i.e. \"syncevolution -c scheduleworld addressbook\"" << endl <<
" followed by \"syncevolution scheduleworld\" will only synchronize the" << endl <<
" address book. The other sources are created in a disabled state." << endl <<
" When modifying an existing configuration and sources are specified," << endl <<
" then the source properties of only those sources are modified." << endl <<
"" << endl <<
"--migrate" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" In older SyncEvolution releases a different layout of configuration files" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" was used. Using --migrate will automatically migrate to the new" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" layout and rename the <config> into <config>.old to prevent accidental use" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" of the old configuration. WARNING: old SyncEvolution releases cannot" << endl <<
" use the new configuration!" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
"" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" The switch can also be used to migrate a configuration in the current" << endl <<
" configuration directory: this preserves all property values, discards" << endl <<
" obsolete properties and sets all comments exactly as if the configuration" << endl <<
" had been created from scratch. WARNING: custom comments in the" << endl <<
" configuration are not preserved." << endl <<
"" << endl <<
added --restore and fixed --status --status was broken by the last commit: XDG_DATA_DIR was not expanded. Fixed it as part of the improvements for printing changes in --restore. Also made some other minor changes to utility classes as part of these improvements. The new --restore option can restore data from a data dump, identified by the directory and a before/after flag. The corresponding API is EvolutionSyncClient::restore().
2009-04-23 16:47:07 +02:00
"--restore" << endl <<
" Restores the data of the selected sources to the state from before or after the" << endl <<
" selected synchronization. The synchronization is selected via its log directory" << endl <<
" (see --print-sessions). Other directories can also be given as long as" << endl <<
" they contain database dumps in the format created by SyncEvolution." << endl <<
" The output includes information about the changes made during the" << endl <<
" restore, both in terms of item changes and content changes (which is" << endl <<
" not always the same, see manual for details). This output can be suppressed" << endl <<
" with --quiet." << endl <<
" In combination with --dry-run, the changes to local data are only simulated." << endl <<
" This can be used to check that --restore will not remove valuable information." << endl <<
"" << endl <<
command line: added long description of --remove option
2009-04-16 08:42:11 +02:00
"--remove" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" Deletes the configuration. If the <config> refers to a specific" << endl <<
" peer, only that peer's configuration is removed. If it refers to" << endl <<
" a context, that context and all peers inside it are removed." << endl <<
" Note that there is no confirmation question. Neither local data" << endl <<
" referenced by the configuration nor the content of log dirs are" << endl <<
" deleted." << endl <<
command line: added long description of --remove option
2009-04-16 08:42:11 +02:00
"" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"--sync-property|-y <property>=<value>" << endl <<
"--sync-property|-y ?" << endl <<
"--sync-property|-y <property>=?" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" Overrides a source-independent configuration property for the" << endl <<
" current synchronization run or permanently when --configure is used" << endl <<
" to update the configuration. Can be used multiple times. Specifying" << endl <<
" an unused property will trigger an error message." << endl <<
"" << endl <<
" When using the configuration layout introduced with 1.0, some of the" << endl <<
" sync properties are shared between peers, for example the directory" << endl <<
" where sessions are logged. Permanently changing such a shared" << endl <<
" property for one peer will automatically update the property for all" << endl <<
" other peers in the same context because the property is stored in a" << endl <<
" shared config file." << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"" << endl <<
"--source-property|-z <property>=<value>" << endl <<
"--source-property|-z ?" << endl <<
"--source-property|-z <property>=?" << endl <<
syncevolution --help: fixed typos (MB#6450) "--configure" contained a long dash instead of two dashes. "--sync-option" -> "--sync-property"
2009-09-28 11:55:24 +02:00
" Same as --sync-property, but applies to the configuration of all active" << endl <<
syncevolution --help: --source-option => --source-property (MB#6450) Missed one instance of this typo in the earlier commit...
2009-10-01 15:59:48 +02:00
" sources. \"--sync <mode>\" is a shortcut for \"--source-property sync=<mode>\"." << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
"--template|-l <peer name>|default|?|?<device>" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" Can be used to select from one of the built-in default configurations" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" for known SyncML peers. Defaults to the <config> name, so --template" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" only has to be specified when creating multiple different configurations" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" for the same peer, or when using a template that is named differently" << endl <<
" than the peer. \"default\" is an alias for \"scheduleworld\" and can be" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" used as the starting point for servers which do not have a built-in" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" template." << endl <<
"" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
" Each template contains a pseudo-random device ID. Therefore setting the" << endl <<
" \"deviceId\" sync property is only necessary when manually recreating a" << endl <<
" configuration or when a more descriptive name is desired." << endl <<
"" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" The available templates for different known SyncML servers are listed when" << endl <<
" using a single question mark instead of template name. When using the" << endl <<
" ?<device> format, a fuzzy search for a template that might be" << endl <<
" suitable for talking to such a device is done. The matching works best" << endl <<
" when using <device> = <Manufacturer>_<Model>. If you don't know the" << endl <<
" manufacturer, you can just keep it as empty. The output in this mode" << endl <<
" gives the template name followed by a short description and a rating how well" << endl <<
" the template matches the device (higher is better)." << endl <<
"" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"--status|-t" << endl <<
" The changes made to local data since the last synchronization are" << endl <<
" shown without starting a new one. This can be used to see in advance" << endl <<
documentation + usage: shared configs, 1.0 features (MB #8497) Mostly replaced <server> with <config> and added new text about the new config layout. Did some general updating of the documentation to match the improvements found in 1.0. Because sharing of properties is not documented, I added code which prints that as part of "--print-config". Lists of sync properties are printed beneath the main sync section in a comment, for source properties once after all sources. Added --print-configs and --print-peers as alias for --print-servers.
2010-01-20 11:36:19 +01:00
" whether the local data needs to be synchronized with the peer." << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"" << endl <<
command line + D-Bus: --status help and output (MB #5043) Added "--status without config name" comment to help output. Changed the "No running session(s) just now" into a full sentence, with a slightly more "positive" twist to it: "Background sync daemon is idle."
2010-03-29 10:58:31 +02:00
" When used without configuration name, it shows the status of the background" << endl <<
" sync daemon or an error if no such daemon exists." << endl <<
"" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"--quiet|-q" << endl <<
" Suppresses most of the normal output during a synchronization. The" << endl <<
" log file still contains all the information." << endl <<
"" << endl <<
command line + daemon: usability improvements (MB #5043) This patch changes what is shown to the user and how the user interacts with the command line. Details below. "--use-daemon [yes/no]" implies that yes/no is optional (square brackets!), with "yes" being the option that could be expected for a plain "--use-daemon" parameter. But the implementation always expects a "yes" or "no" parameter. The original format suggested was "--use-daemon[=yes/no]" This patch switches to that format, changes --use-daemon into --daemon (to be consistent with --keyring) and enables the same format for --keyring. Although not documented, 0/f/false and 1/t/true are also accepted. Because the value becomes part of the parameter, m_argc checks had to be adapted. The documentation for "--use-daemon" was inserted in the middle of the "--keyring documentation". "Parameter not set" has to be available to the Cmdline caller in the command line too, in addition to true/false. This was done originally with a string which is empty, "yes" or "no". Using a tri-state Cmdline::Bool class makes this a bit more explicit and removes the ambiguity around what useDaemon() returns (values weren't documented). When running without --daemon and no daemon available, the first lines of output were: ERROR: org.<cryptic error> [INFO] itodo20: inactive .... => The command line should fall back *silently* to running in-process. => Error messages should be formatted and logged as such, using SE_LOG_ERROR(). Old code might not have done that and we need to preserve that for compatibility with frontends, but new code should use [ERROR] as output. => D-Bus error messages (as in the case above) must have some user (and developer!) comprehensible explanation what went wrong. The D-Bus error class itself is no enough. Although I haven't tested it, I suspect that the code also would have re-run the operation in-process after the D-Bus server already executed it and failed. I rewrote this so that a check for "daemon available" without error messages is done first before committing to using the daemon. Once that decision is made, the command line will not fall back to in-process execution. Rewrote several error messages. Telling a user of a distro's binary to "re-configure" is misleading (he didn't configure himself). "can't" => "cannot", punctuation changes. Not sure whether is always an improvement, comments welcome. Comment on coding style: I've used "if ()" instead of "if()" because that is the GNU recommendation.
2010-03-26 18:47:41 +01:00
"--keyring|-k[=yes/no/...]" << endl <<
command line: add keyring support (MB#3604) --keyring|-k Save or retrieve passwords from the GNOME keyring when modifying the configuration or running a synchronization. Note that using this option applies to *all* passwords in a configuration, so setting a single password as follows moves the other passwords into the keyring, if they were not stored there already: --keyring --configure --sync-property proxyPassword=foo When passwords were stored in the keyring, their value is set to "-" in the configuration. This means that when running a synchronization without the --keyring argument, the password has to be entered interactively. The implementation introduces new virtual methods for password handling to properties and iterates over all properties to activate that special behavior. Another change is no longer to use cached strings in the SyncEvolutionConfig to store retrieved passwords. Instead, they are saved as the filter in the config node. This could help their users to get them but not flush to files. The main purpose of this change is to make it flexible to meet the requirements of dynamically added passwords in the backend.
2009-09-21 10:07:05 +02:00
" Save or retrieve passwords from the GNOME keyring when modifying the" << endl <<
" configuration or running a synchronization. Note that using this option" << endl <<
" applies to *all* passwords in a configuration, so setting a single" << endl <<
" password as follows moves the other passwords into the keyring, if" << endl <<
" they were not stored there already:" << endl <<
" --keyring --configure --sync-property proxyPassword=foo" << endl <<
"" << endl <<
" When passwords were stored in the keyring, their value is set to '-'" << endl <<
" in the configuration. This means that when running a synchronization" << endl <<
" without the --keyring argument, the password has to be entered" << endl <<
command line --keyring: documented interaction with --print-config As discussed on the mailing list and in MB#3604, the passwords are not retrieved from the keyring when --print-config is used.
2009-09-23 07:25:28 +02:00
" interactively. The --print-config output always shows '-' instead of" << endl <<
" retrieving the password from the keyring." << endl <<
command line: add keyring support (MB#3604) --keyring|-k Save or retrieve passwords from the GNOME keyring when modifying the configuration or running a synchronization. Note that using this option applies to *all* passwords in a configuration, so setting a single password as follows moves the other passwords into the keyring, if they were not stored there already: --keyring --configure --sync-property proxyPassword=foo When passwords were stored in the keyring, their value is set to "-" in the configuration. This means that when running a synchronization without the --keyring argument, the password has to be entered interactively. The implementation introduces new virtual methods for password handling to properties and iterates over all properties to activate that special behavior. Another change is no longer to use cached strings in the SyncEvolutionConfig to store retrieved passwords. Instead, they are saved as the filter in the config node. This could help their users to get them but not flush to files. The main purpose of this change is to make it flexible to meet the requirements of dynamically added passwords in the backend.
2009-09-21 10:07:05 +02:00
"" << endl <<
command line + daemon: usability improvements (MB #5043) This patch changes what is shown to the user and how the user interacts with the command line. Details below. "--use-daemon [yes/no]" implies that yes/no is optional (square brackets!), with "yes" being the option that could be expected for a plain "--use-daemon" parameter. But the implementation always expects a "yes" or "no" parameter. The original format suggested was "--use-daemon[=yes/no]" This patch switches to that format, changes --use-daemon into --daemon (to be consistent with --keyring) and enables the same format for --keyring. Although not documented, 0/f/false and 1/t/true are also accepted. Because the value becomes part of the parameter, m_argc checks had to be adapted. The documentation for "--use-daemon" was inserted in the middle of the "--keyring documentation". "Parameter not set" has to be available to the Cmdline caller in the command line too, in addition to true/false. This was done originally with a string which is empty, "yes" or "no". Using a tri-state Cmdline::Bool class makes this a bit more explicit and removes the ambiguity around what useDaemon() returns (values weren't documented). When running without --daemon and no daemon available, the first lines of output were: ERROR: org.<cryptic error> [INFO] itodo20: inactive .... => The command line should fall back *silently* to running in-process. => Error messages should be formatted and logged as such, using SE_LOG_ERROR(). Old code might not have done that and we need to preserve that for compatibility with frontends, but new code should use [ERROR] as output. => D-Bus error messages (as in the case above) must have some user (and developer!) comprehensible explanation what went wrong. The D-Bus error class itself is no enough. Although I haven't tested it, I suspect that the code also would have re-run the operation in-process after the D-Bus server already executed it and failed. I rewrote this so that a check for "daemon available" without error messages is done first before committing to using the daemon. Once that decision is made, the command line will not fall back to in-process execution. Rewrote several error messages. Telling a user of a distro's binary to "re-configure" is misleading (he didn't configure himself). "can't" => "cannot", punctuation changes. Not sure whether is always an improvement, comments welcome. Comment on coding style: I've used "if ()" instead of "if()" because that is the GNU recommendation.
2010-03-26 18:47:41 +01:00
"--daemon[=yes/no/...]" << endl <<
" Run operations in cooperation with the background sync daemon;" << endl <<
" enabled by default if it is installed." << endl <<
"" << endl <<
updated --help text git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@582 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-10 21:36:57 +02:00
"--help|-h" << endl <<
" Prints usage information." << endl <<
"" << endl <<
"--version" << endl <<
" Prints the SyncEvolution version." << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
if (error != "") {
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
out << endl << "ERROR: " << error << endl;
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
}
if (param != "") {
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
out << "INFO: use '" << param << (param[param.size() - 1] == '=' ? "" : " ") <<
next step in the configuration rewrite - added sanity checking of properties - added listing of properties and help for individual properties - use user-visible string as main alias for sources git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@524 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-19 12:29:14 +01:00
"?' to get a list of valid parameters" << endl;
}
}
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
SyncContext* Cmdline::createSyncClient() {
sync source handling: implemented per-source property filtering, Cmdline uses it Selecting active sources during a sync or status check was done with a combination of setting a sync mode via a source config filter and setting a list of active sources. Now the SyncConfig supports a source filter which is applied to all sources and source filters for each source. The latter override the former. This is powerful enough to start syncs with full control over which sources are active in which mode, as described in the new D-Bus API. As part of this patch, the command line semantic is implemented entirely using a combination of different source filters.
2009-09-14 21:54:43 +02:00
return new SyncContext(m_server, true);
command line: add keyring support (MB#3604) --keyring|-k Save or retrieve passwords from the GNOME keyring when modifying the configuration or running a synchronization. Note that using this option applies to *all* passwords in a configuration, so setting a single password as follows moves the other passwords into the keyring, if they were not stored there already: --keyring --configure --sync-property proxyPassword=foo When passwords were stored in the keyring, their value is set to "-" in the configuration. This means that when running a synchronization without the --keyring argument, the password has to be entered interactively. The implementation introduces new virtual methods for password handling to properties and iterates over all properties to activate that special behavior. Another change is no longer to use cached strings in the SyncEvolutionConfig to store retrieved passwords. Instead, they are saved as the filter in the config node. This could help their users to get them but not flush to files. The main purpose of this change is to make it flexible to meet the requirements of dynamically added passwords in the backend.
2009-09-21 10:07:05 +02:00
}
unit tests were unintentionally enabled when only integration tests should have been on git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@590 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-17 20:27:53 +02:00
#ifdef ENABLE_UNIT_TESTS
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
/** simple line-by-line diff */
static string diffStrings(const string &lhs, const string &rhs)
{
ostringstream res;
typedef boost::split_iterator<string::const_iterator> string_split_iterator;
string_split_iterator lit =
boost::make_split_iterator(lhs, boost::first_finder("\n", boost::is_iequal()));
string_split_iterator rit =
boost::make_split_iterator(rhs, boost::first_finder("\n", boost::is_iequal()));
while (lit != string_split_iterator() &&
rit != string_split_iterator()) {
if (*lit != *rit) {
res << "< " << *lit << endl;
res << "> " << *rit << endl;
}
++lit;
++rit;
}
while (lit != string_split_iterator()) {
res << "< " << *lit << endl;
++lit;
}
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
while (rit != string_split_iterator()) {
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
res << "> " << *rit << endl;
++rit;
}
return res.str();
}
# define CPPUNIT_ASSERT_EQUAL_DIFF( expected, actual ) \
do { \
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
string expected_ = (expected); \
string actual_ = (actual); \
if (expected_ != actual_) { \
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
CPPUNIT_NS::Message cpputMsg_(string("expected:\n") + \
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
expected_); \
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
cpputMsg_.addDetail(string("actual:\n") + \
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
actual_); \
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
cpputMsg_.addDetail(string("diff:\n") + \
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
diffStrings(expected_, actual_)); \
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
CPPUNIT_NS::Asserter::fail( cpputMsg_, \
CPPUNIT_SOURCELINE() ); \
} \
} while ( false )
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
// returns last line, including trailing line break, empty if input is empty
static string lastLine(const string &buffer)
{
if (buffer.size() < 2) {
return buffer;
}
size_t line = buffer.rfind("\n", buffer.size() - 2);
if (line == buffer.npos) {
return buffer;
}
return buffer.substr(line + 1);
}
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
// true if <word> =
static bool isPropAssignment(const string &buffer) {
size_t start = 0;
while (start < buffer.size() &&
!isspace(buffer[start])) {
start++;
}
if (start + 3 <= buffer.size() &&
buffer.substr(start, 3) == " = ") {
return true;
} else {
return false;
}
}
// remove pure comment lines from buffer,
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
// also empty lines,
// also defaultPeer (because reference properties do not include global props)
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
static string filterConfig(const string &buffer)
{
ostringstream res;
typedef boost::split_iterator<string::const_iterator> string_split_iterator;
for (string_split_iterator it =
boost::make_split_iterator(buffer, boost::first_finder("\n", boost::is_iequal()));
it != string_split_iterator();
++it) {
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
string line = boost::copy_range<string>(*it);
if (!line.empty() &&
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
line.find("defaultPeer =") == line.npos &&
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
(!boost::starts_with(line, "# ") ||
isPropAssignment(line.substr(2)))) {
res << line << endl;
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
}
}
return res.str();
}
username/password: empty in templates We used to have the strings "your SyncML server account name" and "your SyncML server password" hard-coded in the code which generated templates. As Jussi pointed out, this is a problem for GUIs which have to hide these strings from their users. The purpose of these strings was to have them in configs in cases where the user forgot to set them, as a reminder. This path moves the strings into the command line config setup, thus achieving both of these goals.
2009-11-19 16:30:46 +01:00
static string injectValues(const string &buffer)
{
string res = buffer;
// username/password not set in templates, only in configs created via
// the command line
boost::replace_first(res,
"# username = ",
"username = your SyncML server account name");
boost::replace_first(res,
"# password = ",
"password = your SyncML server password");
return res;
}
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
// remove lines indented with spaces
static string filterIndented(const string &buffer)
{
ostringstream res;
bool first = true;
typedef boost::split_iterator<string::const_iterator> string_split_iterator;
for (string_split_iterator it =
boost::make_split_iterator(buffer, boost::first_finder("\n", boost::is_iequal()));
it != string_split_iterator();
++it) {
if (!boost::starts_with(*it, " ")) {
if (!first) {
res << endl;
} else {
first = false;
}
res << *it;
}
}
return res.str();
}
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
// sort lines by file, preserving order inside each line
static void sortConfig(string &config)
{
// file name, line number, property
typedef pair<string, pair<int, string> > line_t;
vector<line_t> lines;
typedef boost::split_iterator<string::iterator> string_split_iterator;
int linenr = 0;
for (string_split_iterator it =
boost::make_split_iterator(config, boost::first_finder("\n", boost::is_iequal()));
it != string_split_iterator();
++it, ++linenr) {
string line(it->begin(), it->end());
if (line.empty()) {
continue;
}
size_t colon = line.find(':');
string prefix = line.substr(0, colon);
lines.push_back(make_pair(prefix, make_pair(linenr, line.substr(colon))));
}
// stable sort because of line number
sort(lines.begin(), lines.end());
size_t len = config.size();
config.resize(0);
config.reserve(len);
BOOST_FOREACH(const line_t &line, lines) {
config += line.first;
config += line.second.second;
config += "\n";
}
}
// convert the internal config dump to .ini style (--print-config)
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
static string internalToIni(const string &config)
{
ostringstream res;
string section;
typedef boost::split_iterator<string::const_iterator> string_split_iterator;
for (string_split_iterator it =
boost::make_split_iterator(config, boost::first_finder("\n", boost::is_iequal()));
it != string_split_iterator();
++it) {
string line(it->begin(), it->end());
if (line.empty()) {
continue;
}
size_t colon = line.find(':');
string prefix = line.substr(0, colon);
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
// internal values are not part of the --print-config output
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
if (boost::contains(prefix, ".internal.ini") ||
boost::contains(line, "= internal value")) {
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
continue;
}
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
// --print-config also doesn't duplicate the "type" property
// => remove the shared property
if (boost::contains(line, ":type = ") &&
boost::starts_with(line, "sources/")) {
continue;
}
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
// sources/<name>/config.ini or
// spds/sources/<name>/config.ini
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
size_t endslash = prefix.rfind('/');
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
if (endslash != line.npos && endslash > 1) {
size_t slash = prefix.rfind('/', endslash - 1);
if (slash != line.npos) {
string newsource = prefix.substr(slash + 1, endslash - slash - 1);
if (newsource != section &&
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
prefix.find("/sources/") != prefix.npos &&
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
newsource != "syncml") {
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
res << "[" << newsource << "]" << endl;
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
section = newsource;
}
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
}
}
string assignment = line.substr(colon + 1);
// substitude aliases with generic values
boost::replace_first(assignment, "= F", "= 0");
boost::replace_first(assignment, "= T", "= 1");
config: enable clientAuthType (Bugzilla #3242) This is now mapped to the Synthesis defaultauth config option. It seems that the option only has an effect during the initial sync of a client. Later the engine seems to use the setting that was negotiated with the server during that sync, without the possibility to override that later on.
2009-06-10 14:27:16 +02:00
boost::replace_first(assignment, "= syncml:auth-md5", "= md5");
boost::replace_first(assignment, "= syncml:auth-basix", "= basic");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
res << assignment << endl;
}
return res.str();
}
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
/**
* Testing is based on a text representation of a directory
* hierarchy where each line is of the format
* <file path>:<line in file>
*
* The order of files is alphabetical, of lines in the file as
* in the file. Lines in the file without line break cannot
* be represented.
*
* The root of the hierarchy is not part of the representation
* itself.
*/
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
class CmdlineTest : public CppUnit::TestFixture {
CPPUNIT_TEST_SUITE(CmdlineTest);
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
CPPUNIT_TEST(testFramework);
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
CPPUNIT_TEST(testSetupScheduleWorld);
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
CPPUNIT_TEST(testSetupDefault);
CPPUNIT_TEST(testSetupRenamed);
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
CPPUNIT_TEST(testSetupFunambol);
CPPUNIT_TEST(testSetupSynthesis);
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
CPPUNIT_TEST(testPrintServers);
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
CPPUNIT_TEST(testPrintConfig);
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
CPPUNIT_TEST(testPrintFileTemplates);
templates: also search in .config/syncevolution-templates (MBC #1208) With this patch, template files are also found in .config/syncevolution-templates (or the corresponding directory pointed to by the XDG env variables). The tests had to be adapted a) to not find templates installed by the user in his own home and b) to cover finding templates via XDG_CONFIG_HOME.
2010-05-03 21:41:36 +02:00
CPPUNIT_TEST(testPrintFileTemplatesConfig);
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
CPPUNIT_TEST(testTemplate);
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
CPPUNIT_TEST(testMatchTemplate);
command line: support creating new sources (MB #8424) "syncevolution --configure" works by copying the config with modified properties set as filters. This did not work for adding sources because SyncContext::copy() only worked on *existing* sources. Changed the semantic of the sourceFilter parameter in SyncContext::copy() so that it specifies *exactly* which sources are to be copied instead of acting as a filter. If the given set is a subset of the existing sources, this does not change anything (SyncContext.cpp dummy.copy(*this, &activeSources)). If the given set is a superset, then now the additional sources will be copied where they were ignored before (Cmdline.cpp). There are no other calls to SyncConfig::copy(), so this change is safe. Added a new CmdlineTest to cover this situation.
2009-12-03 16:01:32 +01:00
CPPUNIT_TEST(testAddSource);
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
CPPUNIT_TEST(testSync);
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
CPPUNIT_TEST(testConfigure);
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
CPPUNIT_TEST(testOldConfigure);
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
CPPUNIT_TEST(testListSources);
CPPUNIT_TEST(testMigrate);
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
CPPUNIT_TEST_SUITE_END();
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
public:
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
CmdlineTest() :
m_testDir("CmdlineTest"),
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
// properties sorted by the order in which they are defined
// in the sync and sync source property registry
m_scheduleWorldConfig("peers/scheduleworld/.internal.ini:# HashCode = 0\n"
"peers/scheduleworld/.internal.ini:# ConfigDate = \n"
"peers/scheduleworld/.internal.ini:# lastNonce = \n"
"peers/scheduleworld/.internal.ini:# deviceData = \n"
"peers/scheduleworld/config.ini:syncURL = http://sync.scheduleworld.com/funambol/ds\n"
"peers/scheduleworld/config.ini:username = your SyncML server account name\n"
"peers/scheduleworld/config.ini:password = your SyncML server password\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"config.ini:# logdir = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/config.ini:# loglevel = 0\n"
"peers/scheduleworld/config.ini:# printChanges = 1\n"
logging: enable log dirs by default When the "logdir" sync property is unset, then SyncEvolution now maintains log directories in $XDG_DATA_HOME/applications/syncevolution. The "maxlogdirs" property defaults to 10 now because of this change. When the base directory for log dirs ends in "syncevolution", then the directories created inside it start with the server name, followed by a date and time (as before). Otherwise "SyncEvolution-" is added. This is a safety-mechanism that ensures that we have a good chance of finding the right log directories again with a non-standard base directory. The previous default (a single log dir in /tmp with the user name as part of the log dir name) is no longer supported: when configuring a non-standard log dir, that directory has to be chosen so that files from different users aren't mixed up.
2009-04-16 17:22:31 +02:00
"config.ini:# maxlogdirs = 10\n"
DBusServer: fix 4 issues for automatic sync 1) change the registration order of the properties: before 'preventSlowSync'. 2) newly added 3 properties break unit tests in CmdlineTest. Fix them. 3) prevent automatic termination when there is any auto sync task enabled by configs. 4) don't regularly checking auto sync if a config's 'autoSyncInterval' is set as value '0'. Only do auto sync when changes are detected in this scenario.
2010-03-02 06:34:42 +01:00
"peers/scheduleworld/config.ini:# autoSync = 0\n"
autoSyncInterval/autoSyncDelay/RetryDuration/RetryInterval: consistent value handling (MB #10257) All duration config properties now use the same parsing code and store their values as seconds. This changes the interpretation of the auto sync durations, which used minutes before, but because they were only introduced very recently, this change is still acceptable. It would be nice to store values in config files with a hint about the unit, but we cannot do that because it would break compatibility with older software expecting plain number of seconds. The format accepted for all of these options is following ISO 8601, with some relaxed rules: /** * Time interval >= 0. Values are formatted as number of seconds * and accepted in a variety of formats, following ISO 8601: * - x = x seconds * - d = x[YyWwDdHhMmSs] = x years/weeks/days/hours/minutes/seconds * - d[+]d... = combination of the previous durations * * As an extension of ISO 8601, white spaces are silently ignored, * suffix checks are case-insensitive and s (or S) for seconds * can be omitted. */ class SecondsConfigProperty ...
2010-03-22 17:45:03 +01:00
"peers/scheduleworld/config.ini:# autoSyncInterval = 30M\n"
"peers/scheduleworld/config.ini:# autoSyncDelay = 5M\n"
client-test Cmdline: adapted to recent config changes The SyncEvo::CmdlineTest failed because the default values of preventSlowSync and maxMsgSize were changed recently. The deviceId prefix change broke another part of the test.
2010-02-25 17:59:51 +01:00
"peers/scheduleworld/config.ini:# preventSlowSync = 1\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/config.ini:# useProxy = 0\n"
"peers/scheduleworld/config.ini:# proxyHost = \n"
"peers/scheduleworld/config.ini:# proxyUsername = \n"
"peers/scheduleworld/config.ini:# proxyPassword = \n"
"peers/scheduleworld/config.ini:# clientAuthType = md5\n"
autoSyncInterval/autoSyncDelay/RetryDuration/RetryInterval: consistent value handling (MB #10257) All duration config properties now use the same parsing code and store their values as seconds. This changes the interpretation of the auto sync durations, which used minutes before, but because they were only introduced very recently, this change is still acceptable. It would be nice to store values in config files with a hint about the unit, but we cannot do that because it would break compatibility with older software expecting plain number of seconds. The format accepted for all of these options is following ISO 8601, with some relaxed rules: /** * Time interval >= 0. Values are formatted as number of seconds * and accepted in a variety of formats, following ISO 8601: * - x = x seconds * - d = x[YyWwDdHhMmSs] = x years/weeks/days/hours/minutes/seconds * - d[+]d... = combination of the previous durations * * As an extension of ISO 8601, white spaces are silently ignored, * suffix checks are case-insensitive and s (or S) for seconds * can be omitted. */ class SecondsConfigProperty ...
2010-03-22 17:45:03 +01:00
"peers/scheduleworld/config.ini:# RetryDuration = 5M\n"
RetryInterval: 0 for Funambol, 2M for everyone else (MB #10451) A user reported that the default 1 minute resend interval did not work for him with his own Funambol server installation. The Funambol developers are notified, but won't be able to do much about this in the near future. Therefore this patch disables message resend in the Funambol server template. Existing configurations have to be updated manually, both to suppress the resending and to enable it again later. The default resend interval is also increased from 1 minute to 2 minutes, to help other slow servers which might have problems handling resends when they arrive to quickly.
2010-04-19 13:56:07 +02:00
"peers/scheduleworld/config.ini:# RetryInterval = 2M\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/config.ini:# remoteIdentifier = \n"
"peers/scheduleworld/config.ini:# PeerIsClient = 0\n"
SAN 1.0/1.1 generation support (MB#9312) Support generating a legacy version SAN package leverating the new API in libsynthesis. Added a new configuration option 'SyncMLVersion'(1.0|1.1|1.2), on the server side, this is used to select the SAN version to send to the client (default to empty which means it will first try SAN 1.2 and fallback to SAN 1.1 automatically). On the client side, this is used to select the SyncML version talking to remote server, default to empty which means it will try the latest commonly supported SyncML version by the server, when setting a value the client will only try the specifid version.
2010-02-26 11:20:52 +01:00
"peers/scheduleworld/config.ini:# SyncMLVersion = \n"
PeerName sync property: arbitrary name for config (MB #8350) This property is meant to be used by the sync-UI to store the intended spelling of the peer (mixed case) and store characters which are not valid for config names. Therefore the value needs to use escaping rules.
2010-01-13 18:05:45 +01:00
"peers/scheduleworld/config.ini:# PeerName = \n"
cleaned up comments on properties git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@557 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:39:15 +02:00
"config.ini:deviceId = fixed-devid\n" /* this is not the default! */
configuration: updated to match server use cases (MB #7710) remoteDevId must be user-visible, because it is necessary to configure it in advance when the peer is an HTTP SyncML client. Also updated several property documentation strings. We have to explain how they apply when the local side acts as SyncML server and/or OBEX is involved.
2009-11-27 19:47:47 +01:00
"peers/scheduleworld/config.ini:# remoteDeviceId = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/config.ini:# enableWBXML = 1\n"
client-test Cmdline: adapted to recent config changes The SyncEvo::CmdlineTest failed because the default values of preventSlowSync and maxMsgSize were changed recently. The deviceId prefix change broke another part of the test.
2010-02-25 17:59:51 +01:00
"peers/scheduleworld/config.ini:# maxMsgSize = 150000\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/config.ini:# maxObjSize = 4000000\n"
"peers/scheduleworld/config.ini:# enableCompression = 0\n"
"peers/scheduleworld/config.ini:# SSLServerCertificates = \n"
"peers/scheduleworld/config.ini:# SSLVerifyServer = 1\n"
"peers/scheduleworld/config.ini:# SSLVerifyHost = 1\n"
Merge branch 'syncevolution-0-9-branch' Conflicts: NEWS configure-post.in configure-pre.in src/Makefile-gen.am src/syncevo/Cmdline.cpp src/syncevo/SyncConfig.cpp
2010-01-19 16:01:10 +01:00
"peers/scheduleworld/config.ini:WebURL = http://www.scheduleworld.com\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/config.ini:# IconURI = \n"
"peers/scheduleworld/config.ini:ConsumerReady = 1\n"
"peers/scheduleworld/sources/addressbook/.internal.ini:# adminData = \n"
<dbtypeid>: handle hash collisisions The Synthesis engine requires that each <datastore> has a unique integer <dbtypeid>. This is part of the API for progress events and configuring the <datastore>s. SyncEvolution uses the source name as unique key and mapped that to an integer via Hash(), without checking for hash collisions. This patch fixes that by assigning an ID the same way the first time a source is used. Hash collisions are checked at that point. Without hash collisions (the assumption which had to hold already anyway), the same IDs will be used as before. The ID is recorded in the new internal "synthesisID", which means that it won't change even when there are collisions and sources are added and removed. Another benefit of this patch is that the source lookup can be done easily via the unique ID, which is useful for an upcoming patch (which will rename some sources inside the Synthesis engine). Because SourceList assigns the unique ID, it has to prevent that users of it add sources, normal or virtual, without going through SourceList. Therefore the m_virtualDS and vector base class were made private. This patch further clarifies what exactly the role of SourceList is and that normal sources and virtual sources are handled differently. This is important so that code can iterate over one or the other set.
2010-02-15 16:50:06 +01:00
"peers/scheduleworld/sources/addressbook/.internal.ini:# synthesisID = 0\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/addressbook/config.ini:sync = two-way\n"
fixed creating and testing of default configs ScheduleWorld must explicitly select text/vcard for uri3, the others do not (and should not) specify it explicitly and thus use text/x-vcard by default. The renaming of uuid- to sc-pim- broke regression testing; fixed. git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@671 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-12 21:45:50 +02:00
"sources/addressbook/config.ini:type = addressbook:text/vcard\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/addressbook/config.ini:type = addressbook:text/vcard\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"sources/addressbook/config.ini:# evolutionsource = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/addressbook/config.ini:uri = card3\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"sources/addressbook/config.ini:# evolutionuser = \n"
"sources/addressbook/config.ini:# evolutionpassword = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/calendar/.internal.ini:# adminData = \n"
<dbtypeid>: handle hash collisisions The Synthesis engine requires that each <datastore> has a unique integer <dbtypeid>. This is part of the API for progress events and configuring the <datastore>s. SyncEvolution uses the source name as unique key and mapped that to an integer via Hash(), without checking for hash collisions. This patch fixes that by assigning an ID the same way the first time a source is used. Hash collisions are checked at that point. Without hash collisions (the assumption which had to hold already anyway), the same IDs will be used as before. The ID is recorded in the new internal "synthesisID", which means that it won't change even when there are collisions and sources are added and removed. Another benefit of this patch is that the source lookup can be done easily via the unique ID, which is useful for an upcoming patch (which will rename some sources inside the Synthesis engine). Because SourceList assigns the unique ID, it has to prevent that users of it add sources, normal or virtual, without going through SourceList. Therefore the m_virtualDS and vector base class were made private. This patch further clarifies what exactly the role of SourceList is and that normal sources and virtual sources are handled differently. This is important so that code can iterate over one or the other set.
2010-02-15 16:50:06 +01:00
"peers/scheduleworld/sources/calendar/.internal.ini:# synthesisID = 0\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/calendar/config.ini:sync = two-way\n"
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"sources/calendar/config.ini:type = calendar\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/calendar/config.ini:type = calendar\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"sources/calendar/config.ini:# evolutionsource = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/calendar/config.ini:uri = cal2\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"sources/calendar/config.ini:# evolutionuser = \n"
"sources/calendar/config.ini:# evolutionpassword = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/memo/.internal.ini:# adminData = \n"
<dbtypeid>: handle hash collisisions The Synthesis engine requires that each <datastore> has a unique integer <dbtypeid>. This is part of the API for progress events and configuring the <datastore>s. SyncEvolution uses the source name as unique key and mapped that to an integer via Hash(), without checking for hash collisions. This patch fixes that by assigning an ID the same way the first time a source is used. Hash collisions are checked at that point. Without hash collisions (the assumption which had to hold already anyway), the same IDs will be used as before. The ID is recorded in the new internal "synthesisID", which means that it won't change even when there are collisions and sources are added and removed. Another benefit of this patch is that the source lookup can be done easily via the unique ID, which is useful for an upcoming patch (which will rename some sources inside the Synthesis engine). Because SourceList assigns the unique ID, it has to prevent that users of it add sources, normal or virtual, without going through SourceList. Therefore the m_virtualDS and vector base class were made private. This patch further clarifies what exactly the role of SourceList is and that normal sources and virtual sources are handled differently. This is important so that code can iterate over one or the other set.
2010-02-15 16:50:06 +01:00
"peers/scheduleworld/sources/memo/.internal.ini:# synthesisID = 0\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/memo/config.ini:sync = two-way\n"
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"sources/memo/config.ini:type = memo\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/memo/config.ini:type = memo\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"sources/memo/config.ini:# evolutionsource = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/memo/config.ini:uri = note\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"sources/memo/config.ini:# evolutionuser = \n"
"sources/memo/config.ini:# evolutionpassword = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/todo/.internal.ini:# adminData = \n"
<dbtypeid>: handle hash collisisions The Synthesis engine requires that each <datastore> has a unique integer <dbtypeid>. This is part of the API for progress events and configuring the <datastore>s. SyncEvolution uses the source name as unique key and mapped that to an integer via Hash(), without checking for hash collisions. This patch fixes that by assigning an ID the same way the first time a source is used. Hash collisions are checked at that point. Without hash collisions (the assumption which had to hold already anyway), the same IDs will be used as before. The ID is recorded in the new internal "synthesisID", which means that it won't change even when there are collisions and sources are added and removed. Another benefit of this patch is that the source lookup can be done easily via the unique ID, which is useful for an upcoming patch (which will rename some sources inside the Synthesis engine). Because SourceList assigns the unique ID, it has to prevent that users of it add sources, normal or virtual, without going through SourceList. Therefore the m_virtualDS and vector base class were made private. This patch further clarifies what exactly the role of SourceList is and that normal sources and virtual sources are handled differently. This is important so that code can iterate over one or the other set.
2010-02-15 16:50:06 +01:00
"peers/scheduleworld/sources/todo/.internal.ini:# synthesisID = 0\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/todo/config.ini:sync = two-way\n"
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"sources/todo/config.ini:type = todo\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/todo/config.ini:type = todo\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"sources/todo/config.ini:# evolutionsource = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/todo/config.ini:uri = task2\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"sources/todo/config.ini:# evolutionuser = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"sources/todo/config.ini:# evolutionpassword = ")
SSL certificates: include path for Moblin and fix command line test (Bugzilla #4220) Moblin is going to integrate the ca-certificates package from Fedora, which puts the certificate file into /etc/pki/tls/certs/ca-bundle.crt Added that path to the default search path in SyncEvolution for libsoup and also fixed the SyncEvolutionCmdlineTest.
2009-07-17 16:21:25 +02:00
{
#ifdef ENABLE_LIBSOUP
// path to SSL certificates has to be set only for libsoup
boost::replace_first(m_scheduleWorldConfig,
"SSLServerCertificates = ",
"SSLServerCertificates = /etc/ssl/certs/ca-certificates.crt:/etc/pki/tls/certs/ca-bundle.crt:/usr/share/ssl/certs/ca-bundle.crt");
#endif
}
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
protected:
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
/** verify that createFiles/scanFiles themselves work */
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
void testFramework() {
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
const string root(m_testDir);
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
const string content("baz:line\n"
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
"caz/subdir:booh\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"caz/subdir2/sub:# comment\n"
"caz/subdir2/sub:# foo = bar\n"
"caz/subdir2/sub:# empty = \n"
"caz/subdir2/sub:# another comment\n"
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
"foo:bar1\n"
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"foo:\n"
"foo: \n"
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
"foo:bar2\n");
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
const string filtered("baz:line\n"
"caz/subdir:booh\n"
"caz/subdir2/sub:# foo = bar\n"
"caz/subdir2/sub:# empty = \n"
"foo:bar1\n"
"foo: \n"
"foo:bar2\n");
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
createFiles(root, content);
string res = scanFiles(root);
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
CPPUNIT_ASSERT_EQUAL_DIFF(filtered, res);
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
}
deviceId is now a pseudo UUID in new configurations git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@556 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:02:32 +02:00
void removeRandomUUID(string &buffer) {
client-test Cmdline: adapted to recent config changes The SyncEvo::CmdlineTest failed because the default values of preventSlowSync and maxMsgSize were changed recently. The deviceId prefix change broke another part of the test.
2010-02-25 17:59:51 +01:00
string uuidstr = "deviceId = syncevolution-";
deviceId is now a pseudo UUID in new configurations git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@556 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:02:32 +02:00
size_t uuid = buffer.find(uuidstr);
CPPUNIT_ASSERT(uuid != buffer.npos);
size_t end = buffer.find("\n", uuid + uuidstr.size());
CPPUNIT_ASSERT(end != buffer.npos);
buffer.replace(uuid, end - uuid, "deviceId = fixed-devid");
}
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
/** create new configurations */
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
void testSetupScheduleWorld() { doSetupScheduleWorld(false); }
void doSetupScheduleWorld(bool shared) {
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
string root;
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
root = m_testDir;
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
root += "/syncevolution/default";
string peer;
if (shared) {
peer = root + "/peers/scheduleworld";
} else {
peer = root;
}
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
{
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
rm_r(peer);
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
TestCmdline cmdline("--configure",
"--sync-property", "proxyHost = proxy",
"scheduleworld",
"addressbook",
NULL);
cmdline.doit();
string res = scanFiles(root);
deviceId is now a pseudo UUID in new configurations git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@556 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:02:32 +02:00
removeRandomUUID(res);
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
string expected = ScheduleWorldConfig();
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
sortConfig(expected);
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
boost::replace_first(expected,
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"# proxyHost = ",
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
"proxyHost = proxy");
boost::replace_all(expected,
"sync = two-way",
"sync = disabled");
boost::replace_first(expected,
"addressbook/config.ini:sync = disabled",
"addressbook/config.ini:sync = two-way");
CPPUNIT_ASSERT_EQUAL_DIFF(expected, res);
}
{
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
rm_r(peer);
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
TestCmdline cmdline("--configure",
deviceId is now a pseudo UUID in new configurations git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@556 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:02:32 +02:00
"--sync-property", "deviceID = fixed-devid",
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
"scheduleworld",
NULL);
cmdline.doit();
string res = scanFiles(root);
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
string expected = ScheduleWorldConfig();
sortConfig(expected);
CPPUNIT_ASSERT_EQUAL_DIFF(expected, res);
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
}
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
}
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
void testSetupDefault() {
string root;
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
root = m_testDir;
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
root += "/syncevolution/default";
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
rm_r(root);
TestCmdline cmdline("--configure",
"--template", "default",
deviceId is now a pseudo UUID in new configurations git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@556 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:02:32 +02:00
"--sync-property", "deviceID = fixed-devid",
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
"some-other-server",
NULL);
cmdline.doit();
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
string res = scanFiles(root, "some-other-server");
string expected = ScheduleWorldConfig();
sortConfig(expected);
boost::replace_all(expected, "/scheduleworld/", "/some-other-server/");
CPPUNIT_ASSERT_EQUAL_DIFF(expected, res);
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
}
void testSetupRenamed() {
string root;
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
root = m_testDir;
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
root += "/syncevolution/default";
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
rm_r(root);
TestCmdline cmdline("--configure",
"--template", "scheduleworld",
deviceId is now a pseudo UUID in new configurations git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@556 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:02:32 +02:00
"--sync-property", "deviceID = fixed-devid",
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
"scheduleworld2",
NULL);
cmdline.doit();
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
string res = scanFiles(root, "scheduleworld2");
string expected = ScheduleWorldConfig();
sortConfig(expected);
boost::replace_all(expected, "/scheduleworld/", "/scheduleworld2/");
CPPUNIT_ASSERT_EQUAL_DIFF(expected, res);
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
}
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
void testSetupFunambol() { doSetupFunambol(false); }
void doSetupFunambol(bool shared) {
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
string root;
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
root = m_testDir;
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
root += "/syncevolution/default";
string peer;
if (shared) {
peer = root + "/peers/funambol";
} else {
peer = root;
}
rm_r(peer);
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
const char * const argv_fixed[] = {
"--configure",
"--sync-property", "deviceID = fixed-devid",
// templates are case-insensitive
"FunamBOL",
NULL
}, * const argv_shared[] = {
"--configure",
"FunamBOL",
NULL
};
TestCmdline cmdline(shared ? argv_shared : argv_fixed);
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
cmdline.doit();
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
string res = scanFiles(root, "funambol");
string expected = FunambolConfig();
sortConfig(expected);
CPPUNIT_ASSERT_EQUAL_DIFF(expected, res);
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
}
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
void testSetupSynthesis() { doSetupSynthesis(false); }
void doSetupSynthesis(bool shared) {
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
string root;
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
root = m_testDir;
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
root += "/syncevolution/default";
string peer;
if (shared) {
peer = root + "/peers/synthesis";
} else {
peer = root;
}
rm_r(peer);
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
const char * const argv_fixed[] = {
"--configure",
"--sync-property", "deviceID = fixed-devid",
"synthesis",
NULL
}, * const argv_shared[] = {
"--configure",
"synthesis",
NULL
};
TestCmdline cmdline(shared ? argv_shared : argv_fixed);
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
cmdline.doit();
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
string res = scanFiles(root, "synthesis");
string expected = SynthesisConfig();
sortConfig(expected);
CPPUNIT_ASSERT_EQUAL_DIFF(expected, res);
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
}
void testTemplate() {
TestCmdline failure("--template", NULL);
CPPUNIT_ASSERT(!failure.m_cmdline->parse());
CPPUNIT_ASSERT_EQUAL_DIFF("", failure.m_out.str());
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
CPPUNIT_ASSERT_EQUAL(string("ERROR: missing parameter for '--template'\n"), lastLine(failure.m_err.str()));
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
TestCmdline help("--template", "? ", NULL);
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
help.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("Available configuration templates:\n"
Cmdline Test: testMatchTemplate fix (MB#10004) The syncevolutionclient template was changed to only match with 'syncevolutionclient' exactly to avoid collision with the server side template 'syncevolution'. The cmdline test has to adapt accordingly, added a header line to describe the meaning of each output column.
2010-03-03 09:24:58 +01:00
" template name = template description\n"
configuration: template and server names are mixed case with case-insensitive comparison now The mixed case names look nicer, in particular in a GUI. To stay compatible with previous releases and to simplify command line usage, files and directories user lower case server names.
2009-04-08 16:32:05 +02:00
" Funambol = http://my.funambol.com\n"
Google+vcard21: Add Google sync template configuration; bugzilla #2423
2009-06-18 15:25:53 +02:00
" Google = http://m.google.com/sync\n"
Goosync: added template (MB #9113) Goosync provides access to Google calendar, tasks and contacts via SyncML. Google itself only supports contacts. Memos are not supported by either of them. This patch only adds the template with the necessary configuration parameters. We have done no testing with the service, so it is not marked as "consumer ready" at this time.
2010-01-18 10:07:18 +01:00
" Goosync = http://www.goosync.com/\n"
configuration: template and server names are mixed case with case-insensitive comparison now The mixed case names look nicer, in particular in a GUI. To stay compatible with previous releases and to simplify command line usage, files and directories user lower case server names.
2009-04-08 16:32:05 +02:00
" Memotoo = http://www.memotoo.com\n"
Mobical: add configure template for mobical (Bug #3009) add template for mobical
2009-08-07 09:11:43 +02:00
" Mobical = http://www.mobical.net\n"
Oracle: added template The template uses an unspecific "https://your.company/mobilesync/server" syncURL. The hope is that users just have to replace "your.company" in many cases. https is chosen because apparently, many installations are restricted to that.
2009-10-29 16:13:31 +01:00
" Oracle = http://www.oracle.com/technology/products/beehive/index.html\n"
Add ovi template
2010-02-09 04:03:20 +01:00
" Ovi = http://www.ovi.com\n"
ScheduleWorld: changed webURL As requested by Mark, the webURL was changed from sync.scheduleworld.com to www.scheduleworld.com, which is the official entry point for browerser. IMHO sync.scheduleworld.com has always worked, though, and currently still does.
2010-01-18 21:58:31 +01:00
" ScheduleWorld = http://www.scheduleworld.com\n"
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
" SyncEvolution = http://www.syncevolution.org\n"
ZYB: Add template for zyb (Bug #2424) Add zyb template. Since zyb only supports contact and calendar, disable todo and memo by default
2009-07-18 16:40:29 +02:00
" Synthesis = http://www.synthesis.ch\n"
" ZYB = http://www.zyb.com\n",
added --template testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@547 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:04:26 +01:00
help.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", help.m_err.str());
}
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
void testMatchTemplate() {
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "testcases/templates");
templates: also search in .config/syncevolution-templates (MBC #1208) With this patch, template files are also found in .config/syncevolution-templates (or the corresponding directory pointed to by the XDG env variables). The tests had to be adapted a) to not find templates installed by the user in his own home and b) to cover finding templates via XDG_CONFIG_HOME.
2010-05-03 21:41:36 +02:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", "/dev/null");
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
Configuration template: fingerprint changes The first entry in the fingerprint is used to uniqely identify the template (a.k.a template name), it was formulated like 'nokia_7210c' which is not friendly to user. This patch changed the naming to 'Nokia 7210c', other existing templates is also updated accordingly.
2010-03-19 06:19:53 +01:00
TestCmdline help1("--template", "?nokia 7210c", NULL);
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
help1.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("Available configuration templates:\n"
MatchTemplate better output (MB#10004) Instead of a 1-5 valued rank, convert to percentage. ./syncevolution -l \?nokia_7210c Available configuration templates: template name = template description matching score in percent (100% = exact match) Nokia_7210c = Template for Nokia 7210c phone and as default template for all Nokia phones 100% SyncEvolutionClient = SyncEvolution server side template 40% ServerDefault = server side default template 20%
2010-03-04 03:26:03 +01:00
" template name = template description matching score in percent (100% = exact match)\n"
templates: keep fingerPrint a list of models We decided against "abusing" the first entry in the fingerPrint list as the name of the device family/class. The reason is that there might be templates which have no such class, which would be impossible to determine if only "fingerPrint" is available. Instead the plan is to introduce a separate "templateName" property. Note that the sync-ui currently still assumes that the first fingerPrint entry can be shown to the user. To avoid confusion (device is called "Nokia N85", shown as "Nokia 7210c") we have to keep one template per model.
2010-03-24 10:23:20 +01:00
" Nokia 7210c = Template for Nokia S40 series Phone 100%\n"
Configuration template: fingerprint changes The first entry in the fingerprint is used to uniqely identify the template (a.k.a template name), it was formulated like 'nokia_7210c' which is not friendly to user. This patch changed the naming to 'Nokia 7210c', other existing templates is also updated accordingly.
2010-03-19 06:19:53 +01:00
" SyncEvolution Client = SyncEvolution server side template 40%\n",
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
help1.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", help1.m_err.str());
TestCmdline help2("--template", "?nokia", NULL);
help2.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("Available configuration templates:\n"
MatchTemplate better output (MB#10004) Instead of a 1-5 valued rank, convert to percentage. ./syncevolution -l \?nokia_7210c Available configuration templates: template name = template description matching score in percent (100% = exact match) Nokia_7210c = Template for Nokia 7210c phone and as default template for all Nokia phones 100% SyncEvolutionClient = SyncEvolution server side template 40% ServerDefault = server side default template 20%
2010-03-04 03:26:03 +01:00
" template name = template description matching score in percent (100% = exact match)\n"
templates: keep fingerPrint a list of models We decided against "abusing" the first entry in the fingerPrint list as the name of the device family/class. The reason is that there might be templates which have no such class, which would be impossible to determine if only "fingerPrint" is available. Instead the plan is to introduce a separate "templateName" property. Note that the sync-ui currently still assumes that the first fingerPrint entry can be shown to the user. To avoid confusion (device is called "Nokia N85", shown as "Nokia 7210c") we have to keep one template per model.
2010-03-24 10:23:20 +01:00
" Nokia 7210c = Template for Nokia S40 series Phone 100%\n"
Configuration template: fingerprint changes The first entry in the fingerprint is used to uniqely identify the template (a.k.a template name), it was formulated like 'nokia_7210c' which is not friendly to user. This patch changed the naming to 'Nokia 7210c', other existing templates is also updated accordingly.
2010-03-19 06:19:53 +01:00
" SyncEvolution Client = SyncEvolution server side template 40%\n",
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
help2.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", help2.m_err.str());
TestCmdline help3("--template", "?7210c", NULL);
help3.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("Available configuration templates:\n"
MatchTemplate better output (MB#10004) Instead of a 1-5 valued rank, convert to percentage. ./syncevolution -l \?nokia_7210c Available configuration templates: template name = template description matching score in percent (100% = exact match) Nokia_7210c = Template for Nokia 7210c phone and as default template for all Nokia phones 100% SyncEvolutionClient = SyncEvolution server side template 40% ServerDefault = server side default template 20%
2010-03-04 03:26:03 +01:00
" template name = template description matching score in percent (100% = exact match)\n"
templates: keep fingerPrint a list of models We decided against "abusing" the first entry in the fingerPrint list as the name of the device family/class. The reason is that there might be templates which have no such class, which would be impossible to determine if only "fingerPrint" is available. Instead the plan is to introduce a separate "templateName" property. Note that the sync-ui currently still assumes that the first fingerPrint entry can be shown to the user. To avoid confusion (device is called "Nokia N85", shown as "Nokia 7210c") we have to keep one template per model.
2010-03-24 10:23:20 +01:00
" Nokia 7210c = Template for Nokia S40 series Phone 60%\n"
Configuration template: fingerprint changes The first entry in the fingerprint is used to uniqely identify the template (a.k.a template name), it was formulated like 'nokia_7210c' which is not friendly to user. This patch changed the naming to 'Nokia 7210c', other existing templates is also updated accordingly.
2010-03-19 06:19:53 +01:00
" SyncEvolution Client = SyncEvolution server side template 20%\n",
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
help3.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", help3.m_err.str());
Configuration template: fingerprint changes The first entry in the fingerprint is used to uniqely identify the template (a.k.a template name), it was formulated like 'nokia_7210c' which is not friendly to user. This patch changed the naming to 'Nokia 7210c', other existing templates is also updated accordingly.
2010-03-19 06:19:53 +01:00
TestCmdline help4("--template", "?syncevolution client", NULL);
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
help4.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("Available configuration templates:\n"
MatchTemplate better output (MB#10004) Instead of a 1-5 valued rank, convert to percentage. ./syncevolution -l \?nokia_7210c Available configuration templates: template name = template description matching score in percent (100% = exact match) Nokia_7210c = Template for Nokia 7210c phone and as default template for all Nokia phones 100% SyncEvolutionClient = SyncEvolution server side template 40% ServerDefault = server side default template 20%
2010-03-04 03:26:03 +01:00
" template name = template description matching score in percent (100% = exact match)\n"
Configuration template: fingerprint changes The first entry in the fingerprint is used to uniqely identify the template (a.k.a template name), it was formulated like 'nokia_7210c' which is not friendly to user. This patch changed the naming to 'Nokia 7210c', other existing templates is also updated accordingly.
2010-03-19 06:19:53 +01:00
" SyncEvolution Client = SyncEvolution server side template 100%\n"
templates: keep fingerPrint a list of models We decided against "abusing" the first entry in the fingerPrint list as the name of the device family/class. The reason is that there might be templates which have no such class, which would be impossible to determine if only "fingerPrint" is available. Instead the plan is to introduce a separate "templateName" property. Note that the sync-ui currently still assumes that the first fingerPrint entry can be shown to the user. To avoid confusion (device is called "Nokia N85", shown as "Nokia 7210c") we have to keep one template per model.
2010-03-24 10:23:20 +01:00
" Nokia 7210c = Template for Nokia S40 series Phone 40%\n",
Configuration templates matching: match templates based on metadata Introduced TemplateConfig to abstracting the template configuration structure, the template metadata used for matching is also parsed here. The fields introduced in the metadata are: PeerIsClient: identify whether this is a server side configuration or a client side configuration. Fingerprint: the matching string for this template, it is a comma separated string with each string modeled as: "Manufacture_Model". The first substring is also used as the name to identify this template so that user can select the template by this name. eg: Nokia 7210c: Nokia_7210c SyncEvolution server: SyncEvolutionServer, SyncEvolution ScheduleWorld: ScheduleWorld,default SyncEvolution client: SyncEvolutionClient, SyncEvolution Description: this is a just a descriptive string not used for matching. GetServerTemplates is changed to add another "devices" parameter to identify it is asking for templates for a list of "devices". Each device is a tuple <matchstring (devicename), matchMode (server/client/all)>. TemplateList as the return type, which is a list of class TemplateDescription so that we can also return enough information for corresponding templates. This list is sorted by the 3-tuple <finger, rank, name>. Add MatchServerTemplates method which will iterating all templates inside the folder and match against the input parameter and finally return a sorted list of matched templates. The atcually fuzzy match algorithm is based on a LCS (added in the following commit). Cmdline interface is changed accordingly: --template ? is changed to --template ?[string], so that user use the former case to match all templates for a tradiontial SyncML client and the latter case to match templates related to an input string. SyncConfig API is also renamed (Server -> Peer) because both server/client configuration/template are handled. The original configuration template (Funambol and ScheduleWorld) has been moved to the new template structure (under servers), they also have a .template.ini file added so that they can be matched and picked up. All templates for supported servers still have built-in template support in the code as before. Templates for SyncEvolution based server is also added. Server side templates are added (Nokia default, Nokia_7210c and SyncEvolutionServer). Add unit test for the new template match use case.
2010-01-19 08:01:05 +01:00
help4.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", help4.m_err.str());
}
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
void testPrintServers() {
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
rm_r(m_testDir);
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
doSetupScheduleWorld(false);
doSetupSynthesis(true);
doSetupFunambol(true);
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
TestCmdline cmdline("--print-servers", NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("Configured servers:\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
" funambol = CmdlineTest/syncevolution/default/peers/funambol\n"
" scheduleworld = CmdlineTest/syncevolution/default/peers/scheduleworld\n"
" synthesis = CmdlineTest/syncevolution/default/peers/synthesis\n",
test additional aspects of --template and --print-servers git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@548 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 18:26:23 +01:00
cmdline.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
}
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
void testPrintConfig() {
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
rm_r(m_testDir);
testSetupFunambol();
{
TestCmdline failure("--print-config", NULL);
CPPUNIT_ASSERT(failure.m_cmdline->parse());
CPPUNIT_ASSERT(!failure.m_cmdline->run());
CPPUNIT_ASSERT_EQUAL_DIFF("", failure.m_out.str());
CPPUNIT_ASSERT_EQUAL(string("ERROR: --print-config requires either a --template or a server name.\n"),
lastLine(failure.m_err.str()));
}
{
TestCmdline failure("--print-config", "foo", NULL);
CPPUNIT_ASSERT(failure.m_cmdline->parse());
CPPUNIT_ASSERT(!failure.m_cmdline->run());
CPPUNIT_ASSERT_EQUAL_DIFF("", failure.m_out.str());
CPPUNIT_ASSERT_EQUAL(string("ERROR: server 'foo' has not been configured yet.\n"),
lastLine(failure.m_err.str()));
}
{
TestCmdline failure("--print-config", "--template", "foo", NULL);
CPPUNIT_ASSERT(failure.m_cmdline->parse());
CPPUNIT_ASSERT(!failure.m_cmdline->run());
CPPUNIT_ASSERT_EQUAL_DIFF("", failure.m_out.str());
CPPUNIT_ASSERT_EQUAL(string("ERROR: no configuration template for 'foo' available.\n"),
lastLine(failure.m_err.str()));
}
{
TestCmdline cmdline("--print-config", "--template", "scheduleworld", NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
string actual = cmdline.m_out.str();
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
// deviceId must be the one from Funambol
CPPUNIT_ASSERT(boost::contains(actual, "deviceId = fixed-devid"));
username/password: empty in templates We used to have the strings "your SyncML server account name" and "your SyncML server password" hard-coded in the code which generated templates. As Jussi pointed out, this is a problem for GUIs which have to hide these strings from their users. The purpose of these strings was to have them in configs in cases where the user forgot to set them, as a reminder. This path moves the strings into the command line config setup, thus achieving both of these goals.
2009-11-19 16:30:46 +01:00
string filtered = injectValues(filterConfig(actual));
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
CPPUNIT_ASSERT_EQUAL_DIFF(filterConfig(internalToIni(ScheduleWorldConfig())),
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
filtered);
// there should have been comments
CPPUNIT_ASSERT(actual.size() > filtered.size());
}
command line --template new config in non-standard context: wrong properties written (MB #10007) First, "syncevolution --configure --template scheduleworld@context ..." should trigger an error. The context, if any, must be specified as part of the config name. Second, "syncevolution --configure template scheduleworld foo@bar", when foo@bar did not exist yet, copied the shared properties from the default context into foo@bar and the context @bar. Determining the context had to distinguish between "--template scheduleworld" and "scheduleworld@bar", which implies the "scheduleworld" template.
2010-03-02 21:28:28 +01:00
{
TestCmdline cmdline("--print-config", "--template", "scheduleworld@nosuchcontext", NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
string actual = cmdline.m_out.str();
// deviceId must *not* be the one from Funambol because of the new context
CPPUNIT_ASSERT(!boost::contains(actual, "deviceId = fixed-devid"));
}
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
{
configuration: template and server names are mixed case with case-insensitive comparison now The mixed case names look nicer, in particular in a GUI. To stay compatible with previous releases and to simplify command line usage, files and directories user lower case server names.
2009-04-08 16:32:05 +02:00
TestCmdline cmdline("--print-config", "--template", "Default", NULL);
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
username/password: empty in templates We used to have the strings "your SyncML server account name" and "your SyncML server password" hard-coded in the code which generated templates. As Jussi pointed out, this is a problem for GUIs which have to hide these strings from their users. The purpose of these strings was to have them in configs in cases where the user forgot to set them, as a reminder. This path moves the strings into the command line config setup, thus achieving both of these goals.
2009-11-19 16:30:46 +01:00
string actual = injectValues(filterConfig(cmdline.m_out.str()));
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
CPPUNIT_ASSERT(boost::contains(actual, "deviceId = fixed-devid"));
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
CPPUNIT_ASSERT_EQUAL_DIFF(filterConfig(internalToIni(ScheduleWorldConfig())),
deviceId is now a pseudo UUID in new configurations git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@556 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:02:32 +02:00
actual);
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
}
{
TestCmdline cmdline("--print-config", "funambol", NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF(filterConfig(internalToIni(FunambolConfig())),
username/password: empty in templates We used to have the strings "your SyncML server account name" and "your SyncML server password" hard-coded in the code which generated templates. As Jussi pointed out, this is a problem for GUIs which have to hide these strings from their users. The purpose of these strings was to have them in configs in cases where the user forgot to set them, as a reminder. This path moves the strings into the command line config setup, thus achieving both of these goals.
2009-11-19 16:30:46 +01:00
injectValues(filterConfig(cmdline.m_out.str())));
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
}
{
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
// override context and template properties
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
TestCmdline cmdline("--print-config", "--template", "scheduleworld",
"--sync-property", "syncURL=foo",
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
"--source-property", "evolutionsource=Personal",
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
"--source-property", "sync=disabled",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
string expected = filterConfig(internalToIni(ScheduleWorldConfig()));
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
boost::replace_first(expected,
fixed wrong syncURL in ScheduleWorld template git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@637 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-06-28 15:27:07 +02:00
"syncURL = http://sync.scheduleworld.com/funambol/ds",
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
"syncURL = foo");
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
boost::replace_all(expected,
"# evolutionsource = ",
"evolutionsource = Personal");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
boost::replace_all(expected,
"sync = two-way",
"sync = disabled");
username/password: empty in templates We used to have the strings "your SyncML server account name" and "your SyncML server password" hard-coded in the code which generated templates. As Jussi pointed out, this is a problem for GUIs which have to hide these strings from their users. The purpose of these strings was to have them in configs in cases where the user forgot to set them, as a reminder. This path moves the strings into the command line config setup, thus achieving both of these goals.
2009-11-19 16:30:46 +01:00
string actual = injectValues(filterConfig(cmdline.m_out.str()));
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
CPPUNIT_ASSERT(boost::contains(actual, "deviceId = fixed-devid"));
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
CPPUNIT_ASSERT_EQUAL_DIFF(expected,
deviceId is now a pseudo UUID in new configurations git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@556 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:02:32 +02:00
actual);
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
}
{
TestCmdline cmdline("--print-config", "--quiet",
"--template", "scheduleworld",
"funambol",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
deviceId is now a pseudo UUID in new configurations git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@556 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:02:32 +02:00
string actual = cmdline.m_out.str();
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
CPPUNIT_ASSERT(boost::contains(actual, "deviceId = fixed-devid"));
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
CPPUNIT_ASSERT_EQUAL_DIFF(internalToIni(ScheduleWorldConfig()),
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
injectValues(filterConfig(actual)));
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
}
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
{
// change shared source properties, then check template again
TestCmdline cmdline("--configure",
"--source-property", "evolutionsource=Personal",
"funambol",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
}
{
TestCmdline cmdline("--print-config", "--quiet",
"--template", "scheduleworld",
"funambol",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
string expected = filterConfig(internalToIni(ScheduleWorldConfig()));
// from modified Funambol config
boost::replace_all(expected,
"# evolutionsource = ",
"evolutionsource = Personal");
string actual = injectValues(filterConfig(cmdline.m_out.str()));
CPPUNIT_ASSERT(boost::contains(actual, "deviceId = fixed-devid"));
CPPUNIT_ASSERT_EQUAL_DIFF(expected,
actual);
}
command line --template new config in non-standard context: wrong properties written (MB #10007) First, "syncevolution --configure --template scheduleworld@context ..." should trigger an error. The context, if any, must be specified as part of the config name. Second, "syncevolution --configure template scheduleworld foo@bar", when foo@bar did not exist yet, copied the shared properties from the default context into foo@bar and the context @bar. Determining the context had to distinguish between "--template scheduleworld" and "scheduleworld@bar", which implies the "scheduleworld" template.
2010-03-02 21:28:28 +01:00
{
// print config => must not use settings from default context
TestCmdline cmdline("--print-config", "--template", "scheduleworld@nosuchcontext", NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
// source settings *not* from modified Funambol config
string expected = filterConfig(internalToIni(ScheduleWorldConfig()));
string actual = injectValues(filterConfig(cmdline.m_out.str()));
CPPUNIT_ASSERT(!boost::contains(actual, "deviceId = fixed-devid"));
removeRandomUUID(actual);
CPPUNIT_ASSERT_EQUAL_DIFF(expected,
actual);
}
{
// create config => again, must not use settings from default context
TestCmdline cmdline("--configure", "--template", "scheduleworld", "other@other", NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
}
{
TestCmdline cmdline("--print-config", "other@other", NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
// source settings *not* from modified Funambol config
string expected = filterConfig(internalToIni(ScheduleWorldConfig()));
string actual = injectValues(filterConfig(cmdline.m_out.str()));
CPPUNIT_ASSERT(!boost::contains(actual, "deviceId = fixed-devid"));
removeRandomUUID(actual);
CPPUNIT_ASSERT_EQUAL_DIFF(expected,
actual);
}
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
}
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
void testPrintFileTemplates() {
templates: also search in .config/syncevolution-templates (MBC #1208) With this patch, template files are also found in .config/syncevolution-templates (or the corresponding directory pointed to by the XDG env variables). The tests had to be adapted a) to not find templates installed by the user in his own home and b) to cover finding templates via XDG_CONFIG_HOME.
2010-05-03 21:41:36 +02:00
rm_r(m_testDir);
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
// use local copy of templates in build dir (no need to install)
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "./templates");
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
templates: also search in .config/syncevolution-templates (MBC #1208) With this patch, template files are also found in .config/syncevolution-templates (or the corresponding directory pointed to by the XDG env variables). The tests had to be adapted a) to not find templates installed by the user in his own home and b) to cover finding templates via XDG_CONFIG_HOME.
2010-05-03 21:41:36 +02:00
doPrintFileTemplates();
}
void testPrintFileTemplatesConfig() {
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
rm_r(m_testDir);
templates: also search in .config/syncevolution-templates (MBC #1208) With this patch, template files are also found in .config/syncevolution-templates (or the corresponding directory pointed to by the XDG env variables). The tests had to be adapted a) to not find templates installed by the user in his own home and b) to cover finding templates via XDG_CONFIG_HOME.
2010-05-03 21:41:36 +02:00
mkdir_p(m_testDir);
symlink("../templates", (m_testDir + "/syncevolution-templates").c_str());
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
doPrintFileTemplates();
}
void doPrintFileTemplates() {
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
testSetupFunambol();
{
TestCmdline cmdline("--print-config", "--template", "scheduleworld", NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
string actual = cmdline.m_out.str();
// deviceId must be the one from Funambol
CPPUNIT_ASSERT(boost::contains(actual, "deviceId = fixed-devid"));
string filtered = injectValues(filterConfig(actual));
CPPUNIT_ASSERT_EQUAL_DIFF(filterConfig(internalToIni(ScheduleWorldConfig())),
filtered);
// there should have been comments
CPPUNIT_ASSERT(actual.size() > filtered.size());
}
{
TestCmdline cmdline("--print-config", "funambol", NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF(filterConfig(internalToIni(FunambolConfig())),
injectValues(filterConfig(cmdline.m_out.str())));
}
}
command line: support creating new sources (MB #8424) "syncevolution --configure" works by copying the config with modified properties set as filters. This did not work for adding sources because SyncContext::copy() only worked on *existing* sources. Changed the semantic of the sourceFilter parameter in SyncContext::copy() so that it specifies *exactly* which sources are to be copied instead of acting as a filter. If the given set is a subset of the existing sources, this does not change anything (SyncContext.cpp dummy.copy(*this, &activeSources)). If the given set is a superset, then now the additional sources will be copied where they were ignored before (Cmdline.cpp). There are no other calls to SyncConfig::copy(), so this change is safe. Added a new CmdlineTest to cover this situation.
2009-12-03 16:01:32 +01:00
void testAddSource() {
string root;
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
testSetupScheduleWorld();
root = m_testDir;
root += "/syncevolution/default";
{
TestCmdline cmdline("--configure",
"--source-property", "uri = dummy",
"scheduleworld",
"xyz",
NULL);
cmdline.doit();
string res = scanFiles(root);
string expected = ScheduleWorldConfig();
expected += "\n"
"peers/scheduleworld/sources/xyz/.internal.ini:# adminData = \n"
<dbtypeid>: handle hash collisisions The Synthesis engine requires that each <datastore> has a unique integer <dbtypeid>. This is part of the API for progress events and configuring the <datastore>s. SyncEvolution uses the source name as unique key and mapped that to an integer via Hash(), without checking for hash collisions. This patch fixes that by assigning an ID the same way the first time a source is used. Hash collisions are checked at that point. Without hash collisions (the assumption which had to hold already anyway), the same IDs will be used as before. The ID is recorded in the new internal "synthesisID", which means that it won't change even when there are collisions and sources are added and removed. Another benefit of this patch is that the source lookup can be done easily via the unique ID, which is useful for an upcoming patch (which will rename some sources inside the Synthesis engine). Because SourceList assigns the unique ID, it has to prevent that users of it add sources, normal or virtual, without going through SourceList. Therefore the m_virtualDS and vector base class were made private. This patch further clarifies what exactly the role of SourceList is and that normal sources and virtual sources are handled differently. This is important so that code can iterate over one or the other set.
2010-02-15 16:50:06 +01:00
"peers/scheduleworld/sources/xyz/.internal.ini:# synthesisID = 0\n"
config: adding sources affects peers in the same context (MB #9329) When adding a source "foo" in peer "xyz" which doesn't exist in peer "abc" using the same context, the new source appeared as new source in "abc" (okay, by design because of shared properties) and was enabled (not okay). It was meant to be disabled via the "sync = disabled" default value, except that this default had not been changed when implementing MB Changed that default now. Because "sync" is set explicitly when configuring sources, this should have no side effects.
2010-01-29 21:16:00 +01:00
"peers/scheduleworld/sources/xyz/config.ini:# sync = disabled\n"
command line: support creating new sources (MB #8424) "syncevolution --configure" works by copying the config with modified properties set as filters. This did not work for adding sources because SyncContext::copy() only worked on *existing* sources. Changed the semantic of the sourceFilter parameter in SyncContext::copy() so that it specifies *exactly* which sources are to be copied instead of acting as a filter. If the given set is a subset of the existing sources, this does not change anything (SyncContext.cpp dummy.copy(*this, &activeSources)). If the given set is a superset, then now the additional sources will be copied where they were ignored before (Cmdline.cpp). There are no other calls to SyncConfig::copy(), so this change is safe. Added a new CmdlineTest to cover this situation.
2009-12-03 16:01:32 +01:00
"peers/scheduleworld/sources/xyz/config.ini:# type = select backend\n"
"peers/scheduleworld/sources/xyz/config.ini:uri = dummy\n"
"sources/xyz/config.ini:# type = select backend\n"
"sources/xyz/config.ini:# evolutionsource = \n"
"sources/xyz/config.ini:# evolutionuser = \n"
"sources/xyz/config.ini:# evolutionpassword = ";
sortConfig(expected);
CPPUNIT_ASSERT_EQUAL_DIFF(expected, res);
}
}
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
void testSync() {
TestCmdline failure("--sync", NULL);
CPPUNIT_ASSERT(!failure.m_cmdline->parse());
CPPUNIT_ASSERT_EQUAL_DIFF("", failure.m_out.str());
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
CPPUNIT_ASSERT_EQUAL(string("ERROR: missing parameter for '--sync'\n"), lastLine(failure.m_err.str()));
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
TestCmdline failure2("--sync", "foo", NULL);
CPPUNIT_ASSERT(!failure2.m_cmdline->parse());
CPPUNIT_ASSERT_EQUAL_DIFF("", failure2.m_out.str());
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
CPPUNIT_ASSERT_EQUAL(string("ERROR: '--sync foo': not one of the valid values (two-way, slow, refresh-from-client = refresh-client, refresh-from-server = refresh-server = refresh, one-way-from-client = one-way-client, one-way-from-server = one-way-server = one-way, disabled = none)\n"), lastLine(failure2.m_err.str()));
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
testing of/fixed creating config with sources specified; improved white space handling in parameters git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@552 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:11:45 +02:00
TestCmdline help("--sync", " ?", NULL);
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
help.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("--sync\n"
" requests a certain synchronization mode:\n"
" two-way = only send/receive changes since last sync\n"
" slow = exchange all items\n"
" refresh-from-client = discard all remote items and replace with\n"
" the items on the client\n"
" refresh-from-server = discard all local items and replace with\n"
" the items on the server\n"
" one-way-from-client = transmit changes from client\n"
" one-way-from-server = transmit changes from server\n"
cleaned up comments on properties git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@557 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:39:15 +02:00
" none (or disabled) = synchronization disabled\n",
added testing of --sync property storing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@549 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 11:02:07 +02:00
help.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", help.m_err.str());
TestCmdline filter("--sync", "refresh-from-server", NULL);
CPPUNIT_ASSERT(filter.m_cmdline->parse());
CPPUNIT_ASSERT(!filter.m_cmdline->run());
CPPUNIT_ASSERT_EQUAL_DIFF("", filter.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF("sync = refresh-from-server",
string(filter.m_cmdline->m_sourceProps));
CPPUNIT_ASSERT_EQUAL_DIFF("",
string(filter.m_cmdline->m_syncProps));
TestCmdline filter2("--source-property", "sync=refresh", NULL);
CPPUNIT_ASSERT(filter2.m_cmdline->parse());
CPPUNIT_ASSERT(!filter2.m_cmdline->run());
CPPUNIT_ASSERT_EQUAL_DIFF("", filter2.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF("sync = refresh",
string(filter2.m_cmdline->m_sourceProps));
CPPUNIT_ASSERT_EQUAL_DIFF("",
string(filter2.m_cmdline->m_syncProps));
}
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
void testConfigure() {
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
rm_r(m_testDir);
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
testSetupScheduleWorld();
shared layout: fix for showing and setting "type" property (MB #9939) The "type" property is special because it must be set in the per-peer config node (backend and data format) and in the shared config node (only backend relevant). The MultiplexConfigNode picks the right node, but that was based on the (invalid) assumption that the peer nodes would be NULL when not in use. What really happens is that dummy instances are used, to allow writing into these nodes. Because of this false assumption, the "type" for the context was always printed as the default "select backend". Fixed by telling the MultiplexConfigNode that its peer nodes are dummies. In addition, setting the "type" property for a context didn't work because copyProperties() treated "type" as a per-peer property which had to be skipped. Fixed by teaching it about the special case. The D-Bus and Cmdline tests were extended to cover these cases. They failed (as expected) before fixing the implementation.
2010-03-02 16:30:10 +01:00
string expected = doConfigure(ScheduleWorldConfig(), "sources/addressbook/config.ini:");
{
// updating type for peer must also update type for context
TestCmdline cmdline("--configure",
"--source-property", "type=file:text/vcard:3.0",
"scheduleworld", "addressbook",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
boost::replace_all(expected,
"type = addressbook:text/vcard",
"type = file:text/vcard:3.0");
CPPUNIT_ASSERT_EQUAL_DIFF(expected,
filterConfig(printConfig("scheduleworld")));
string shared = filterConfig(printConfig("@default"));
CPPUNIT_ASSERT(shared.find("type = file:text/vcard:3.0") != shared.npos);
}
{
// updating type for context must not affect peer
TestCmdline cmdline("--configure",
"--source-property", "type=file:text/vcard:2.1",
"@default", "addressbook",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
CPPUNIT_ASSERT_EQUAL_DIFF(expected,
filterConfig(printConfig("scheduleworld")));
string shared = filterConfig(printConfig("@default"));
CPPUNIT_ASSERT(shared.find("type = file:text/vcard:2.1") != shared.npos);
}
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
string syncProperties("syncURL:\n"
"\n"
"username:\n"
- all passwords can be specified as env variables (${foo}) or read from stdin (-) - fixed config access in sync sources (broke during rewrite) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@562 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-03 22:01:56 +02:00
"\n"
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"password:\n"
"\n"
"logdir:\n"
"\n"
"loglevel:\n"
"\n"
logging: allow detailed control over printing of database changes with --printChanges printChanges: enables or disables the detailed (and sometimes slow) comparison of database content before and after a sync session This option can be overriden temporarily by classes derived from EvolutionSyncClient in the prepare() call with setPrintChanges().
2009-04-21 11:22:32 +02:00
"printChanges:\n"
"\n"
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"maxlogdirs:\n"
"\n"
DBusServer: fix 4 issues for automatic sync 1) change the registration order of the properties: before 'preventSlowSync'. 2) newly added 3 properties break unit tests in CmdlineTest. Fix them. 3) prevent automatic termination when there is any auto sync task enabled by configs. 4) don't regularly checking auto sync if a config's 'autoSyncInterval' is set as value '0'. Only do auto sync when changes are detected in this scenario.
2010-03-02 06:34:42 +01:00
"autoSync:\n"
"\n"
"autoSyncInterval:\n"
"\n"
"autoSyncDelay:\n"
"\n"
preventSlowSync: updated Cmdline config tests Adding the "preventSlowSync" config option broke the config tests in Cmdline.cpp. Fixed.
2010-01-26 17:56:02 +01:00
"preventSlowSync:\n"
"\n"
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"useProxy:\n"
"\n"
"proxyHost:\n"
"\n"
"proxyUsername:\n"
- all passwords can be specified as env variables (${foo}) or read from stdin (-) - fixed config access in sync sources (broke during rewrite) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@562 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-03 22:01:56 +02:00
"\n"
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"proxyPassword:\n"
"\n"
"clientAuthType:\n"
"\n"
command line unit test: adapted to new RetryDuration/Interval config properties The test hadn't been updated to include the new properties, leading to diffs between expected and actual configuration in some cases.
2009-09-23 15:16:42 +02:00
"RetryDuration:\n"
"\n"
"RetryInterval:\n"
"\n"
client-test: added remoteIdentifier + PeerIsClient The two properties were added to the config, but the test wasn't updated accordingly.
2009-11-19 16:33:33 +01:00
"remoteIdentifier:\n"
"\n"
"PeerIsClient:\n"
"\n"
SAN 1.0/1.1 generation support (MB#9312) Support generating a legacy version SAN package leverating the new API in libsynthesis. Added a new configuration option 'SyncMLVersion'(1.0|1.1|1.2), on the server side, this is used to select the SAN version to send to the client (default to empty which means it will first try SAN 1.2 and fallback to SAN 1.1 automatically). On the client side, this is used to select the SyncML version talking to remote server, default to empty which means it will try the latest commonly supported SyncML version by the server, when setting a value the client will only try the specifid version.
2010-02-26 11:20:52 +01:00
"SyncMLVersion:\n"
"\n"
PeerName sync property: arbitrary name for config (MB #8350) This property is meant to be used by the sync-UI to store the intended spelling of the peer (mixed case) and store characters which are not valid for config names. Therefore the value needs to use escaping rules.
2010-01-13 18:05:45 +01:00
"PeerName:\n"
"\n"
cleaned up comments on properties git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@557 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 20:39:15 +02:00
"deviceId:\n"
"\n"
configuration: updated to match server use cases (MB #7710) remoteDevId must be user-visible, because it is necessary to configure it in advance when the peer is an HTTP SyncML client. Also updated several property documentation strings. We have to explain how they apply when the local side acts as SyncML server and/or OBEX is involved.
2009-11-27 19:47:47 +01:00
"remoteDeviceId:\n"
"\n"
configuration: removed obsolete config options and added WBXML Removed the "encoding" source option. It was of dubious usefulness with Funambol and has meaning with Synthesis at all. Removed dead code in EvolutionSyncConfig. Added "enableWBXML" sync option. WBXML is used by default, but for debugging disabling it might be useful: dumping WBXML doesn't seem to work reliably. Most Client::Sync tests use the default encoding, usually WBXML unless changed via CLIENT_TEST_XML=1. Client::Sync::*::testItemsXML always uses XML and Client::Sync::*::testItems always WBXML.
2009-03-17 14:16:49 +01:00
"enableWBXML:\n"
"\n"
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"maxMsgSize:\n"
"maxObjSize:\n"
"\n"
added SSL options also to regression testing git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@588 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-04-16 22:33:38 +02:00
"enableCompression:\n"
"\n"
"SSLServerCertificates:\n"
"\n"
"SSLVerifyServer:\n"
"\n"
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
"SSLVerifyHost:\n"
"\n"
"WebURL:\n"
"\n"
configuration: added "ConsumerReady" option A property of server template configs. True if the server is ready for use by "normal" users (everyone can get an account and some kind of support, we have tested the server well enough, ...). Currently the ScheduleWorld and Funambol configs qualify. Synthesis is not generally available, Memotoo has to be retested and included in the nightly regression testing.
2009-06-29 20:32:26 +02:00
"IconURI:\n"
"\n"
config: added "defaultPeer" global property + SaveConfigNode "hidden" flags A global property is not useful yet at the moment because there is no way of sharing or reading it outside of a specific peer configuration. This property will be useful once shared properties truly work. Adding it now simplifies testing. The property uses an escape mechanism to store unusual characters, although that shouldn't be necessary in practice. The motivation is that it is typically read and written by frontends, which benefit from the escape mechanism (in contrast to normal users editing the config file directly). The property is user-visible, just in case that someone is interested. That SafeConfigProperty automatically sets the "hidden" flag was unexpected. Removed that behaviour, in particular given that all other such nodes already set the flag explicitly anyway.
2009-11-13 17:25:30 +01:00
"ConsumerReady:\n"
"\n"
"defaultPeer:\n");
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
string sourceProperties("sync:\n"
"\n"
"type:\n"
"\n"
"evolutionsource:\n"
"\n"
"uri:\n"
"\n"
"evolutionuser:\n"
configuration: removed obsolete config options and added WBXML Removed the "encoding" source option. It was of dubious usefulness with Funambol and has meaning with Synthesis at all. Removed dead code in EvolutionSyncConfig. Added "enableWBXML" sync option. WBXML is used by default, but for debugging disabling it might be useful: dumping WBXML doesn't seem to work reliably. Most Client::Sync tests use the default encoding, usually WBXML unless changed via CLIENT_TEST_XML=1. Client::Sync::*::testItemsXML always uses XML and Client::Sync::*::testItems always WBXML.
2009-03-17 14:16:49 +01:00
"evolutionpassword:\n");
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
{
TestCmdline cmdline("--sync-property", "?",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF(syncProperties,
filterIndented(cmdline.m_out.str()));
}
{
TestCmdline cmdline("--source-property", "?",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF(sourceProperties,
filterIndented(cmdline.m_out.str()));
}
{
TestCmdline cmdline("--source-property", "?",
"--sync-property", "?",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF(sourceProperties + syncProperties,
filterIndented(cmdline.m_out.str()));
}
{
TestCmdline cmdline("--sync-property", "?",
"--source-property", "?",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF(syncProperties + sourceProperties,
filterIndented(cmdline.m_out.str()));
}
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
}
void testOldConfigure() {
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
string oldConfig = OldScheduleWorldConfig();
InitList<string> props = InitList<string>("serverNonce") +
"clientNonce" +
"devInfoHash" +
testing: fixed SyncEvolutionCmdlineTest after adding HashCode/ConfigDate Adding new properties affects the config handling tests. Updated.
2009-06-25 10:55:22 +02:00
"HashCode" +
"ConfigDate" +
server config: fixed tests and code for new config options The Cmdline tests hadn't been updated together with adding the new options and therefore failed. New properties must be added to the reference config m_scheduleWorldConfig and new internal properties also to the props list in testOldConfigure. While checking the test failures, the following problems were found and fixed: - "adminData" property name was used for both sync and source property. The sync property stores Synthesis device information and was renamed to "deviceData" to avoid confusion. SyncConfig API calls renamed accordingly. Property instance now follows syncProp* naming convention. - The source "adminData" property was not declared as "hidden" and therefore showed up in the user-visible config.ini files and --print-config output. The real value set/get in SyncSourceAdmin was done correctly via the m_hiddenNode.
2009-11-12 21:19:03 +01:00
"deviceData" +
"adminData" +
<dbtypeid>: handle hash collisisions The Synthesis engine requires that each <datastore> has a unique integer <dbtypeid>. This is part of the API for progress events and configuring the <datastore>s. SyncEvolution uses the source name as unique key and mapped that to an integer via Hash(), without checking for hash collisions. This patch fixes that by assigning an ID the same way the first time a source is used. Hash collisions are checked at that point. Without hash collisions (the assumption which had to hold already anyway), the same IDs will be used as before. The ID is recorded in the new internal "synthesisID", which means that it won't change even when there are collisions and sources are added and removed. Another benefit of this patch is that the source lookup can be done easily via the unique ID, which is useful for an upcoming patch (which will rename some sources inside the Synthesis engine). Because SourceList assigns the unique ID, it has to prevent that users of it add sources, normal or virtual, without going through SourceList. Therefore the m_virtualDS and vector base class were made private. This patch further clarifies what exactly the role of SourceList is and that normal sources and virtual sources are handled differently. This is important so that code can iterate over one or the other set.
2010-02-15 16:50:06 +01:00
"synthesisID" +
server config: fixed tests and code for new config options The Cmdline tests hadn't been updated together with adding the new options and therefore failed. New properties must be added to the reference config m_scheduleWorldConfig and new internal properties also to the props list in testOldConfigure. While checking the test failures, the following problems were found and fixed: - "adminData" property name was used for both sync and source property. The sync property stores Synthesis device information and was renamed to "deviceData" to avoid confusion. SyncConfig API calls renamed accordingly. Property instance now follows syncProp* naming convention. - The source "adminData" property was not declared as "hidden" and therefore showed up in the user-visible config.ini files and --print-config output. The real value set/get in SyncSourceAdmin was done correctly via the m_hiddenNode.
2009-11-12 21:19:03 +01:00
"lastNonce" +
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
"last";
BOOST_FOREACH(string &prop, props) {
boost::replace_all(oldConfig,
prop + " = ",
prop + " = internal value");
}
rm_r(m_testDir);
createFiles(m_testDir + "/.sync4j/evolution/scheduleworld", oldConfig);
doConfigure(oldConfig, "spds/sources/addressbook/config.txt:");
}
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
shared layout: fix for showing and setting "type" property (MB #9939) The "type" property is special because it must be set in the per-peer config node (backend and data format) and in the shared config node (only backend relevant). The MultiplexConfigNode picks the right node, but that was based on the (invalid) assumption that the peer nodes would be NULL when not in use. What really happens is that dummy instances are used, to allow writing into these nodes. Because of this false assumption, the "type" for the context was always printed as the default "select backend". Fixed by telling the MultiplexConfigNode that its peer nodes are dummies. In addition, setting the "type" property for a context didn't work because copyProperties() treated "type" as a per-peer property which had to be skipped. Fixed by teaching it about the special case. The D-Bus and Cmdline tests were extended to cover these cases. They failed (as expected) before fixing the implementation.
2010-03-02 16:30:10 +01:00
string doConfigure(const string &SWConfig, const string &addressbookPrefix) {
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
string expected;
{
TestCmdline cmdline("--configure",
"--source-property", "sync = disabled",
"scheduleworld",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
expected = filterConfig(internalToIni(SWConfig));
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
boost::replace_all(expected,
"sync = two-way",
"sync = disabled");
CPPUNIT_ASSERT_EQUAL_DIFF(expected,
filterConfig(printConfig("scheduleworld")));
}
{
TestCmdline cmdline("--configure",
"--source-property", "sync = one-way-from-server",
"scheduleworld",
"addressbook",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
expected = SWConfig;
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
boost::replace_all(expected,
"sync = two-way",
"sync = disabled");
boost::replace_first(expected,
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
addressbookPrefix + "sync = disabled",
addressbookPrefix + "sync = one-way-from-server");
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
expected = filterConfig(internalToIni(expected));
CPPUNIT_ASSERT_EQUAL_DIFF(expected,
filterConfig(printConfig("scheduleworld")));
}
{
TestCmdline cmdline("--configure",
"--sync", "two-way",
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"-z", "evolutionsource=source",
logging: enable log dirs by default When the "logdir" sync property is unset, then SyncEvolution now maintains log directories in $XDG_DATA_HOME/applications/syncevolution. The "maxlogdirs" property defaults to 10 now because of this change. When the base directory for log dirs ends in "syncevolution", then the directories created inside it start with the server name, followed by a date and time (as before). Otherwise "SyncEvolution-" is added. This is a safety-mechanism that ensures that we have a good chance of finding the right log directories again with a non-standard base directory. The previous default (a single log dir in /tmp with the user name as part of the log dir name) is no longer supported: when configuring a non-standard log dir, that directory has to be chosen so that files from different users aren't mixed up.
2009-04-16 17:22:31 +02:00
"--sync-property", "maxlogdirs=20",
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"-y", "LOGDIR=logdir",
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
"scheduleworld",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
boost::replace_all(expected,
"sync = one-way-from-server",
"sync = two-way");
boost::replace_all(expected,
"sync = disabled",
"sync = two-way");
boost::replace_all(expected,
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"# evolutionsource = ",
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
"evolutionsource = source");
boost::replace_all(expected,
logging: enable log dirs by default When the "logdir" sync property is unset, then SyncEvolution now maintains log directories in $XDG_DATA_HOME/applications/syncevolution. The "maxlogdirs" property defaults to 10 now because of this change. When the base directory for log dirs ends in "syncevolution", then the directories created inside it start with the server name, followed by a date and time (as before). Otherwise "SyncEvolution-" is added. This is a safety-mechanism that ensures that we have a good chance of finding the right log directories again with a non-standard base directory. The previous default (a single log dir in /tmp with the user name as part of the log dir name) is no longer supported: when configuring a non-standard log dir, that directory has to be chosen so that files from different users aren't mixed up.
2009-04-16 17:22:31 +02:00
"# maxlogdirs = 10",
"maxlogdirs = 20");
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
boost::replace_all(expected,
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
"# logdir = ",
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
"logdir = logdir");
CPPUNIT_ASSERT_EQUAL_DIFF(expected,
filterConfig(printConfig("scheduleworld")));
}
shared layout: fix for showing and setting "type" property (MB #9939) The "type" property is special because it must be set in the per-peer config node (backend and data format) and in the shared config node (only backend relevant). The MultiplexConfigNode picks the right node, but that was based on the (invalid) assumption that the peer nodes would be NULL when not in use. What really happens is that dummy instances are used, to allow writing into these nodes. Because of this false assumption, the "type" for the context was always printed as the default "select backend". Fixed by telling the MultiplexConfigNode that its peer nodes are dummies. In addition, setting the "type" property for a context didn't work because copyProperties() treated "type" as a per-peer property which had to be skipped. Fixed by teaching it about the special case. The D-Bus and Cmdline tests were extended to cover these cases. They failed (as expected) before fixing the implementation.
2010-03-02 16:30:10 +01:00
return expected;
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
}
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
void testListSources() {
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
// pick the varargs constructor; NULL alone is ambiguous
TestCmdline cmdline(NULL, NULL);
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
// exact output varies, do not test
}
void testMigrate() {
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
rm_r(m_testDir);
string oldRoot = m_testDir + "/.sync4j/evolution/scheduleworld";
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
string newRoot = m_testDir + "/syncevolution/default";
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
string oldConfig = OldScheduleWorldConfig();
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
{
// migrate old config
createFiles(oldRoot, oldConfig);
string createdConfig = scanFiles(oldRoot);
TestCmdline cmdline("--migrate",
"scheduleworld",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
string migratedConfig = scanFiles(newRoot);
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
string expected = ScheduleWorldConfig();
sortConfig(expected);
CPPUNIT_ASSERT_EQUAL_DIFF(expected, migratedConfig);
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
string renamedConfig = scanFiles(oldRoot + ".old");
CPPUNIT_ASSERT_EQUAL_DIFF(createdConfig, renamedConfig);
}
{
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
// rewrite existing config with obsolete properties
// => these properties should get removed
//
// There is one limitation: shared nodes are not rewritten.
// This is acceptable.
createFiles(newRoot + "/peers/scheduleworld",
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"config.ini:# obsolete comment\n"
"config.ini:obsoleteprop = foo\n",
true);
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
string createdConfig = scanFiles(newRoot, "scheduleworld");
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
TestCmdline cmdline("--migrate",
"scheduleworld",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
string migratedConfig = scanFiles(newRoot, "scheduleworld");
string expected = ScheduleWorldConfig();
sortConfig(expected);
CPPUNIT_ASSERT_EQUAL_DIFF(expected, migratedConfig);
string renamedConfig = scanFiles(newRoot, "scheduleworld.old");
boost::replace_all(createdConfig, "/scheduleworld/", "/scheduleworld.old/");
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
CPPUNIT_ASSERT_EQUAL_DIFF(createdConfig, renamedConfig);
}
{
syncevolution --migrate: copy .synthesis directory (MB #8048) Without copying this directory, using the migrated configuration will lead to a slow sync. SyncEvo::CmdlineTest::testMigrate was extended to cover this problem. Manually testing shows that the slow sync is indeed avoided with this change.
2009-11-25 15:30:07 +01:00
// migrate old config with changes and .synthesis directory, a second time
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
createFiles(oldRoot, oldConfig);
createFiles(oldRoot,
syncevolution --migrate: copy .synthesis directory (MB #8048) Without copying this directory, using the migrated configuration will lead to a slow sync. SyncEvo::CmdlineTest::testMigrate was extended to cover this problem. Manually testing shows that the slow sync is indeed avoided with this change.
2009-11-25 15:30:07 +01:00
".synthesis/dummy-file.bfi:dummy = foobar\n"
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
"spds/sources/addressbook/changes/config.txt:foo = bar\n"
"spds/sources/addressbook/changes/config.txt:foo2 = bar2\n",
true);
string createdConfig = scanFiles(oldRoot);
rm_r(newRoot);
TestCmdline cmdline("--migrate",
"scheduleworld",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
string migratedConfig = scanFiles(newRoot);
string expected = m_scheduleWorldConfig;
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
sortConfig(expected);
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
boost::replace_first(expected,
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"peers/scheduleworld/sources/addressbook/config.ini",
"peers/scheduleworld/sources/addressbook/.other.ini:foo = bar\n"
"peers/scheduleworld/sources/addressbook/.other.ini:foo2 = bar2\n"
"peers/scheduleworld/sources/addressbook/config.ini");
syncevolution --migrate: copy .synthesis directory (MB #8048) Without copying this directory, using the migrated configuration will lead to a slow sync. SyncEvo::CmdlineTest::testMigrate was extended to cover this problem. Manually testing shows that the slow sync is indeed avoided with this change.
2009-11-25 15:30:07 +01:00
boost::replace_first(expected,
"peers/scheduleworld/config.ini",
"peers/scheduleworld/.synthesis/dummy-file.bfi:dummy = foobar\n"
"peers/scheduleworld/config.ini");
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
CPPUNIT_ASSERT_EQUAL_DIFF(expected, migratedConfig);
string renamedConfig = scanFiles(oldRoot + ".old.1");
CPPUNIT_ASSERT_EQUAL_DIFF(createdConfig, renamedConfig);
}
syncevolution --migrate: support migration into a certain context This depends on peer/context awareness in the code which looks for the old config and renames it. If the old config is not in the same context as the new one, it is searched by peer name alone. This works for all old-style configs as well as rewriting a peer inside the same context.
2009-11-25 16:10:44 +01:00
{
string otherRoot = m_testDir + "/syncevolution/other";
rm_r(otherRoot);
// migrate old config into non-default context
createFiles(oldRoot, oldConfig);
string createdConfig = scanFiles(oldRoot);
{
TestCmdline cmdline("--migrate",
"scheduleworld@other",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
}
string migratedConfig = scanFiles(otherRoot);
string expected = ScheduleWorldConfig();
sortConfig(expected);
CPPUNIT_ASSERT_EQUAL_DIFF(expected, migratedConfig);
string renamedConfig = scanFiles(oldRoot + ".old");
CPPUNIT_ASSERT_EQUAL_DIFF(createdConfig, renamedConfig);
// migrate the migrated config again inside the "other" context
{
TestCmdline cmdline("--migrate",
"scheduleworld@other",
NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_out.str());
}
migratedConfig = scanFiles(otherRoot, "scheduleworld");
expected = ScheduleWorldConfig();
sortConfig(expected);
CPPUNIT_ASSERT_EQUAL_DIFF(expected, migratedConfig);
renamedConfig = scanFiles(otherRoot, "scheduleworld.old");
boost::replace_all(expected, "/scheduleworld/", "/scheduleworld.old/");
CPPUNIT_ASSERT_EQUAL_DIFF(expected, renamedConfig);
}
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
}
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
const string m_testDir;
SSL certificates: include path for Moblin and fix command line test (Bugzilla #4220) Moblin is going to integrate the ca-certificates package from Fedora, which puts the certificate file into /etc/pki/tls/certs/ca-bundle.crt Added that path to the default search path in SyncEvolution for libsoup and also fixed the SyncEvolutionCmdlineTest.
2009-07-17 16:21:25 +02:00
string m_scheduleWorldConfig;
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
private:
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
/**
* vararg constructor with NULL termination,
* out and error stream into stringstream members
*/
class TestCmdline {
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
void init() {
m_argv.reset(new const char *[m_argvstr.size() + 1]);
m_argv[0] = "client-test";
for (size_t index = 0;
index < m_argvstr.size();
++index) {
m_argv[index + 1] = m_argvstr[index].c_str();
}
m_cmdline.set(new Cmdline(m_argvstr.size() + 1, m_argv.get(), m_out, m_err), "cmdline");
}
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
public:
TestCmdline(const char *arg, ...) {
va_list argList;
va_start (argList, arg);
for (const char *curr = arg;
curr;
curr = va_arg(argList, const char *)) {
m_argvstr.push_back(curr);
}
va_end(argList);
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
init();
}
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
TestCmdline(const char * const argv[]) {
for (int i = 0; argv[i]; i++) {
m_argvstr.push_back(argv[i]);
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
}
command line: preserve and show shared properties (MB #8048) When adding a new peer, existing shared properties must be preserved. The same applies to migrating (= copying) and old configuration. To be consistent, printing templates also needs to take existing shared properties into account. Special treatment is necessary for the "type" property: it must be taken from the peer, not from the existing config, so that the peer can choose a data format for himself. The peer config must be consistent with regards to the backend for this to work. The code is similar to syncevo-dbus-server, but not identical because temporarily overriding properties is done differently. Therefore SyncConfig::createServerTemplate() has been left as it is (= unaware of shared properties). The client-test SyncEvo::CmdlineTest was updated to cover some of these case.
2009-11-24 19:26:58 +01:00
init();
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
}
void doit() {
bool success;
success = m_cmdline->parse() &&
m_cmdline->run();
if (m_err.str().size()) {
stdout: never write to std::cout directly (MB #5041) Cmdline and SyncContext standard output should always go through a variable that can be pointed towards the actual output channel. Cmdline already had m_out for that, but it wasn't used everywhere. Fixed. SyncContext now has setOutput()/getOutput() and that is used in SyncContext.cpp. It is not set anywhere yet. To catch incorrect use of cout or cerr inside SyncEvolution, the SyncEvo namespace defines its own cout and cerr which cannot be used like std::cout/cerr, thus triggering compiler errors. Use "std::cout/cerr" when necessary.
2010-03-26 09:53:27 +01:00
m_out << endl << m_err.str();
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
}
CPPUNIT_ASSERT(success);
}
ostringstream m_out, m_err;
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
cxxptr<Cmdline> m_cmdline;
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
private:
vector<string> m_argvstr;
boost::scoped_array<const char *> m_argv;
};
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
string ScheduleWorldConfig() {
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
string config = m_scheduleWorldConfig;
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
#if 0
// Currently we don't have an icon for ScheduleWorld. If we
// had (MB #2062) one, then this code would ensure that the
// reference config also has the right path for it.
const char *templateDir = getenv("SYNCEVOLUTION_TEMPLATE_DIR");
if (!templateDir) {
templateDir = TEMPLATE_DIR;
}
if (isDir(string(templateDir) + "/ScheduleWorld")) {
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
boost::replace_all(config,
"# IconURI = ",
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
string("IconURI = file://") + templateDir + "/ScheduleWorld/icon.png");
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
}
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
#endif
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
return config;
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
}
string OldScheduleWorldConfig() {
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
// old style paths
string oldConfig =
"spds/syncml/config.txt:syncURL = http://sync.scheduleworld.com/funambol/ds\n"
"spds/syncml/config.txt:username = your SyncML server account name\n"
"spds/syncml/config.txt:password = your SyncML server password\n"
"spds/syncml/config.txt:# logdir = \n"
"spds/syncml/config.txt:# loglevel = 0\n"
"spds/syncml/config.txt:# printChanges = 1\n"
"spds/syncml/config.txt:# maxlogdirs = 10\n"
DBusServer: fix 4 issues for automatic sync 1) change the registration order of the properties: before 'preventSlowSync'. 2) newly added 3 properties break unit tests in CmdlineTest. Fix them. 3) prevent automatic termination when there is any auto sync task enabled by configs. 4) don't regularly checking auto sync if a config's 'autoSyncInterval' is set as value '0'. Only do auto sync when changes are detected in this scenario.
2010-03-02 06:34:42 +01:00
"spds/syncml/config.txt:# autoSync = 0\n"
autoSyncInterval/autoSyncDelay/RetryDuration/RetryInterval: consistent value handling (MB #10257) All duration config properties now use the same parsing code and store their values as seconds. This changes the interpretation of the auto sync durations, which used minutes before, but because they were only introduced very recently, this change is still acceptable. It would be nice to store values in config files with a hint about the unit, but we cannot do that because it would break compatibility with older software expecting plain number of seconds. The format accepted for all of these options is following ISO 8601, with some relaxed rules: /** * Time interval >= 0. Values are formatted as number of seconds * and accepted in a variety of formats, following ISO 8601: * - x = x seconds * - d = x[YyWwDdHhMmSs] = x years/weeks/days/hours/minutes/seconds * - d[+]d... = combination of the previous durations * * As an extension of ISO 8601, white spaces are silently ignored, * suffix checks are case-insensitive and s (or S) for seconds * can be omitted. */ class SecondsConfigProperty ...
2010-03-22 17:45:03 +01:00
"spds/syncml/config.txt:# autoSyncInterval = 30M\n"
"spds/syncml/config.txt:# autoSyncDelay = 5M\n"
client-test Cmdline: adapted to recent config changes The SyncEvo::CmdlineTest failed because the default values of preventSlowSync and maxMsgSize were changed recently. The deviceId prefix change broke another part of the test.
2010-02-25 17:59:51 +01:00
"spds/syncml/config.txt:# preventSlowSync = 1\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"spds/syncml/config.txt:# useProxy = 0\n"
"spds/syncml/config.txt:# proxyHost = \n"
"spds/syncml/config.txt:# proxyUsername = \n"
"spds/syncml/config.txt:# proxyPassword = \n"
"spds/syncml/config.txt:# clientAuthType = md5\n"
autoSyncInterval/autoSyncDelay/RetryDuration/RetryInterval: consistent value handling (MB #10257) All duration config properties now use the same parsing code and store their values as seconds. This changes the interpretation of the auto sync durations, which used minutes before, but because they were only introduced very recently, this change is still acceptable. It would be nice to store values in config files with a hint about the unit, but we cannot do that because it would break compatibility with older software expecting plain number of seconds. The format accepted for all of these options is following ISO 8601, with some relaxed rules: /** * Time interval >= 0. Values are formatted as number of seconds * and accepted in a variety of formats, following ISO 8601: * - x = x seconds * - d = x[YyWwDdHhMmSs] = x years/weeks/days/hours/minutes/seconds * - d[+]d... = combination of the previous durations * * As an extension of ISO 8601, white spaces are silently ignored, * suffix checks are case-insensitive and s (or S) for seconds * can be omitted. */ class SecondsConfigProperty ...
2010-03-22 17:45:03 +01:00
"spds/syncml/config.txt:# RetryDuration = 5M\n"
RetryInterval: 0 for Funambol, 2M for everyone else (MB #10451) A user reported that the default 1 minute resend interval did not work for him with his own Funambol server installation. The Funambol developers are notified, but won't be able to do much about this in the near future. Therefore this patch disables message resend in the Funambol server template. Existing configurations have to be updated manually, both to suppress the resending and to enable it again later. The default resend interval is also increased from 1 minute to 2 minutes, to help other slow servers which might have problems handling resends when they arrive to quickly.
2010-04-19 13:56:07 +02:00
"spds/syncml/config.txt:# RetryInterval = 2M\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"spds/syncml/config.txt:# remoteIdentifier = \n"
"spds/syncml/config.txt:# PeerIsClient = 0\n"
SAN 1.0/1.1 generation support (MB#9312) Support generating a legacy version SAN package leverating the new API in libsynthesis. Added a new configuration option 'SyncMLVersion'(1.0|1.1|1.2), on the server side, this is used to select the SAN version to send to the client (default to empty which means it will first try SAN 1.2 and fallback to SAN 1.1 automatically). On the client side, this is used to select the SyncML version talking to remote server, default to empty which means it will try the latest commonly supported SyncML version by the server, when setting a value the client will only try the specifid version.
2010-02-26 11:20:52 +01:00
"spds/syncml/config.txt:# SyncMLVersion = \n"
PeerName sync property: arbitrary name for config (MB #8350) This property is meant to be used by the sync-UI to store the intended spelling of the peer (mixed case) and store characters which are not valid for config names. Therefore the value needs to use escaping rules.
2010-01-13 18:05:45 +01:00
"spds/syncml/config.txt:# PeerName = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"spds/syncml/config.txt:deviceId = fixed-devid\n" /* this is not the default! */
configuration: updated to match server use cases (MB #7710) remoteDevId must be user-visible, because it is necessary to configure it in advance when the peer is an HTTP SyncML client. Also updated several property documentation strings. We have to explain how they apply when the local side acts as SyncML server and/or OBEX is involved.
2009-11-27 19:47:47 +01:00
"spds/syncml/config.txt:# remoteDeviceId = \n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"spds/syncml/config.txt:# enableWBXML = 1\n"
client-test Cmdline: adapted to recent config changes The SyncEvo::CmdlineTest failed because the default values of preventSlowSync and maxMsgSize were changed recently. The deviceId prefix change broke another part of the test.
2010-02-25 17:59:51 +01:00
"spds/syncml/config.txt:# maxMsgSize = 150000\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"spds/syncml/config.txt:# maxObjSize = 4000000\n"
"spds/syncml/config.txt:# enableCompression = 0\n"
#ifdef ENABLE_LIBSOUP
// path to SSL certificates is only set for libsoup
"spds/syncml/config.txt:# SSLServerCertificates = /etc/ssl/certs/ca-certificates.crt:/etc/pki/tls/certs/ca-bundle.crt:/usr/share/ssl/certs/ca-bundle.crt\n"
#else
"spds/syncml/config.txt:# SSLServerCertificates = \n"
#endif
"spds/syncml/config.txt:# SSLVerifyServer = 1\n"
"spds/syncml/config.txt:# SSLVerifyHost = 1\n"
resolved template merge bug Somehow after merging, src/templates/Funambol and src/templates/ScheduleWorld came back. The WebURL change for ScheduleWorld did not make it into all the right places on master.
2010-01-19 19:17:31 +01:00
"spds/syncml/config.txt:WebURL = http://www.scheduleworld.com\n"
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
"spds/syncml/config.txt:# IconURI = \n"
"spds/syncml/config.txt:ConsumerReady = 1\n"
"spds/sources/addressbook/config.txt:sync = two-way\n"
"spds/sources/addressbook/config.txt:type = addressbook:text/vcard\n"
"spds/sources/addressbook/config.txt:# evolutionsource = \n"
"spds/sources/addressbook/config.txt:uri = card3\n"
"spds/sources/addressbook/config.txt:# evolutionuser = \n"
"spds/sources/addressbook/config.txt:# evolutionpassword = \n"
"spds/sources/calendar/config.txt:sync = two-way\n"
"spds/sources/calendar/config.txt:type = calendar\n"
"spds/sources/calendar/config.txt:# evolutionsource = \n"
"spds/sources/calendar/config.txt:uri = cal2\n"
"spds/sources/calendar/config.txt:# evolutionuser = \n"
"spds/sources/calendar/config.txt:# evolutionpassword = \n"
"spds/sources/memo/config.txt:sync = two-way\n"
"spds/sources/memo/config.txt:type = memo\n"
"spds/sources/memo/config.txt:# evolutionsource = \n"
"spds/sources/memo/config.txt:uri = note\n"
"spds/sources/memo/config.txt:# evolutionuser = \n"
"spds/sources/memo/config.txt:# evolutionpassword = \n"
"spds/sources/todo/config.txt:sync = two-way\n"
"spds/sources/todo/config.txt:type = todo\n"
"spds/sources/todo/config.txt:# evolutionsource = \n"
"spds/sources/todo/config.txt:uri = task2\n"
"spds/sources/todo/config.txt:# evolutionuser = \n"
"spds/sources/todo/config.txt:# evolutionpassword = \n";
added testing of --configure on old-style config git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@555 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 18:13:08 +02:00
return oldConfig;
}
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
string FunambolConfig() {
string config = m_scheduleWorldConfig;
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
boost::replace_all(config, "/scheduleworld/", "/funambol/");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
boost::replace_first(config,
fixed wrong syncURL in ScheduleWorld template git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@637 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-06-28 15:27:07 +02:00
"syncURL = http://sync.scheduleworld.com/funambol/ds",
added memotoo to expected list of templates, fixed Funambol syncURL git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@641 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-06-28 23:32:16 +02:00
"syncURL = http://my.funambol.com/sync");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
boost::replace_first(config,
ScheduleWorld: changed webURL As requested by Mark, the webURL was changed from sync.scheduleworld.com to www.scheduleworld.com, which is the official entry point for browerser. IMHO sync.scheduleworld.com has always worked, though, and currently still does.
2010-01-18 21:58:31 +01:00
"WebURL = http://www.scheduleworld.com",
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
"WebURL = http://my.funambol.com");
WBXML: enabled by default, except for Funambol Using WBXML reduces message sizes and increases parsing performance. It was not enabled initially in the 0.9 releases in order to test this new feature more thoroughly. Old configs don't have an explicit enableWBXML setting and therefore will automatically use the new default. The Funambol server has a known issue with WBXML: https://forge.ow2.org/tracker/index.php?func=detail&aid=313729&group_id=96&atid=100096 When creating new configs for it, the default is overridden and WBXML is disabled. Old configs should be updated accordingly, just to be safe. The bug is unlikely to be triggered, but if it happens, sync sessions fail completely.
2009-06-25 11:06:07 +02:00
boost::replace_first(config,
"# enableWBXML = 1",
"enableWBXML = 0");
RetryInterval: 0 for Funambol, 2M for everyone else (MB #10451) A user reported that the default 1 minute resend interval did not work for him with his own Funambol server installation. The Funambol developers are notified, but won't be able to do much about this in the near future. Therefore this patch disables message resend in the Funambol server template. Existing configurations have to be updated manually, both to suppress the resending and to enable it again later. The default resend interval is also increased from 1 minute to 2 minutes, to help other slow servers which might have problems handling resends when they arrive to quickly.
2010-04-19 13:56:07 +02:00
boost::replace_first(config,
"# RetryInterval = 2M",
"RetryInterval = 0");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
boost::replace_first(config,
"addressbook/config.ini:uri = card3",
"addressbook/config.ini:uri = card");
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
boost::replace_all(config,
"addressbook/config.ini:type = addressbook:text/vcard",
"addressbook/config.ini:type = addressbook");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
boost::replace_first(config,
scheduleworld config template: calendar was using 'event2' instead of the correct 'cal2' URI git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@624 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-06-12 20:55:01 +02:00
"calendar/config.ini:uri = cal2",
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
"calendar/config.ini:uri = event");
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
boost::replace_all(config,
"calendar/config.ini:type = calendar",
"calendar/config.ini:type = calendar:text/calendar!");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
boost::replace_first(config,
"todo/config.ini:uri = task2",
"todo/config.ini:uri = task");
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
boost::replace_all(config,
"todo/config.ini:type = todo",
"todo/config.ini:type = todo:text/calendar!");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
return config;
}
string SynthesisConfig() {
string config = m_scheduleWorldConfig;
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
boost::replace_all(config, "/scheduleworld/", "/synthesis/");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
boost::replace_first(config,
fixed wrong syncURL in ScheduleWorld template git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@637 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-06-28 15:27:07 +02:00
"syncURL = http://sync.scheduleworld.com/funambol/ds",
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
"syncURL = http://www.synthesis.ch/sync");
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
boost::replace_first(config,
ScheduleWorld: changed webURL As requested by Mark, the webURL was changed from sync.scheduleworld.com to www.scheduleworld.com, which is the official entry point for browerser. IMHO sync.scheduleworld.com has always worked, though, and currently still does.
2010-01-18 21:58:31 +01:00
"WebURL = http://www.scheduleworld.com",
configuration: added support for configuration templates in /etc and server icons The server config.ini files and icon.[svg|png] in src/default will be installed in /etc/default/syncevolution. All configurations found there extend and/or override the builtin templates. The advantage is that file-based templates can have icons and that templates can be added/modified without recompiling SyncEvolution. This meta information is now part of EvolutionSyncConfig: - getWebURL() returns a URL with further information; this replaces the hard-coded URL string that was previously returned as comment for templates in the ServerList - getIconURI() currently returns absolute file paths to an icon file (example: icon.png for ScheduleWorld). The content and exact URI may vary, depending on how system administrators or distributions configure SyncEvolution. If possible, callers should be able to handle http:// and other URI access methods.
2009-04-07 20:33:18 +02:00
"WebURL = http://www.synthesis.ch");
configuration: added "ConsumerReady" option A property of server template configs. True if the server is ready for use by "normal" users (everyone can get an account and some kind of support, we have tested the server well enough, ...). Currently the ScheduleWorld and Funambol configs qualify. Synthesis is not generally available, Memotoo has to be retested and included in the nightly regression testing.
2009-06-29 20:32:26 +02:00
boost::replace_first(config,
"ConsumerReady = 1",
"# ConsumerReady = 0");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
boost::replace_first(config,
"addressbook/config.ini:uri = card3",
"addressbook/config.ini:uri = contacts");
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
boost::replace_all(config,
"addressbook/config.ini:type = addressbook:text/vcard",
"addressbook/config.ini:type = addressbook");
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
boost::replace_first(config,
scheduleworld config template: calendar was using 'event2' instead of the correct 'cal2' URI git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@624 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-06-12 20:55:01 +02:00
"calendar/config.ini:uri = cal2",
added testing of --print-config, fixed some of the error handling git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@551 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 13:43:36 +02:00
"calendar/config.ini:uri = events");
boost::replace_first(config,
"calendar/config.ini:sync = two-way",
"calendar/config.ini:sync = disabled");
boost::replace_first(config,
"memo/config.ini:uri = note",
"memo/config.ini:uri = notes");
boost::replace_first(config,
"todo/config.ini:uri = task2",
"todo/config.ini:uri = tasks");
boost::replace_first(config,
"todo/config.ini:sync = two-way",
"todo/config.ini:sync = disabled");
return config;
ScopedEnvChange: moved from Cmdline.cpp to util The class turned out to be necessary for SyncConfig testing, once that started to look for configs in the normal locations in $HOME and $XDG_CONFIG_HOME (separate patch for that).
2010-01-29 20:28:24 +01:00
}
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
/** create directory hierarchy, overwriting previous content */
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
void createFiles(const string &root, const string &content, bool append = false) {
if (!append) {
rm_r(root);
}
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
size_t start = 0;
ofstream out;
string outname;
out.exceptions(ios_base::badbit|ios_base::failbit);
while (start < content.size()) {
size_t delim = content.find(':', start);
size_t end = content.find('\n', start);
if (delim == content.npos ||
end == content.npos) {
// invalid content ?!
break;
}
string newname = content.substr(start, delim - start);
string line = content.substr(delim + 1, end - delim - 1);
if (newname != outname) {
if (out.is_open()) {
out.close();
}
string fullpath = root + "/" + newname;
size_t fileoff = fullpath.rfind('/');
mkdir_p(fullpath.substr(0, fileoff));
added testing of source listing and migration, minor fixes git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@554 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 17:15:00 +02:00
out.open(fullpath.c_str(),
append ? ios_base::out : (ios_base::out|ios_base::trunc));
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
outname = newname;
}
out << line << endl;
start = end + 1;
}
}
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
/** turn directory hierarchy into string
*
* @param root root path in file system
* @param peer if non-empty, then ignore all <root>/peers/<foo> directories
* where <foo> != peer
* @param onlyProps ignore lines which are comments
*/
string scanFiles(const string &root, const string &peer = "", bool onlyProps = true) {
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
ostringstream out;
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
scanFiles(root, "", peer, out, onlyProps);
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
return out.str();
}
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
void scanFiles(const string &root, const string &dir, const string &peer, ostringstream &out, bool onlyProps) {
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
string newroot = root;
newroot += "/";
newroot += dir;
ReadDir readDir(newroot);
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
sort(readDir.begin(), readDir.end());
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
BOOST_FOREACH(const string &entry, readDir) {
if (isDir(newroot + "/" + entry)) {
client-test: updated SyncEvo::CmdlineTest to work with shared config layout Root of the tested configuration files is the "default" context. The global property "defaultPeer" is thus not covered and must be ignored in the --print-config output (where it is included). Removing and scanning for properties of a specific peer must take the "peers" hierarchy into account. Some tests work with more than one peer configuration at a time. The "type" property is duplicated. This is only visible in a file system dump, not in the --print-config output. The reference ScheduleWorld config dump uses the order of properties as in the registry so that it can be converted into the --print-config .ini format automatically. When the order produced by scanning files is required, the lines have to be sorted first. The old sync4j-style config is no longer derived from the reference ScheduleWorld config. Instead it is stored as plain text; this was easier than updating the transformation. One problem found in testMigrate is that obsolete shared properties are not removed. Removal of obsolete properties was a side effect of rewriting a config under a new name as part of --migrate; it still works for per-peer properties. This is a minor feature, so I don't intend to reimplement this for shared properties. The tests do not yet cover proper creation of templates when there are already existing shared properties. Needs to be added together with a fix for the problem.
2009-11-24 15:02:21 +01:00
if (boost::ends_with(newroot, "/peers") &&
!peer.empty() &&
entry != peer) {
// skip different peer directory
continue;
} else {
scanFiles(root, dir + (dir.empty() ? "" : "/") + entry, peer, out, onlyProps);
}
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
} else {
ifstream in;
in.exceptions(ios_base::badbit /* failbit must not trigger exception because is set when reaching eof ?! */);
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
in.open((newroot + "/" + entry).c_str());
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
string line;
while (!in.eof()) {
getline(in, line);
if ((line.size() || !in.eof()) &&
(!onlyProps ||
default values are commmented out in .ini files; some important ones are always set normally git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@559 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 23:50:51 +02:00
(boost::starts_with(line, "# ") ?
isPropAssignment(line.substr(2)) :
!line.empty()))) {
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
if (dir.size()) {
out << dir << "/";
}
reduced explicit usage of iterators (mostly with BOOST_FOREACH) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@665 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-07-11 22:25:02 +02:00
out << entry << ":";
added testing of config creation, fixed setting of default values git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@543 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-29 15:16:34 +01:00
out << line << '\n';
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
}
}
}
}
}
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
string printConfig(const string &server) {
fixed reading templates from file (MB #8335) Broke while adding support for shared configs. We didn't have a test for it, therefore this went unnoticed. I have added a test for it (CmdlineTest::printFileTemplates) and fixed the issue (using the wrong file layout for templates). The test depends on having access to installed templates in a known location, which previously was a problem due to the hard-coded TEMPLATE_DIR. Now SyncContext::createServerTemplate() checks an env variable (SYNCEVOLUTION_TEMPLATE_DIR) which is set by most of the command line tests to prevent reading from there unintentionally and once to "./templates" to ensure reading from there.
2009-12-01 16:36:33 +01:00
ScopedEnvChange templates("SYNCEVOLUTION_TEMPLATE_DIR", "/dev/null");
added testing of configuring existing config; minor fixes (case sensitivity in filter, modifying only selected sources) git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@553 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-30 15:41:56 +02:00
ScopedEnvChange xdg("XDG_CONFIG_HOME", m_testDir);
ScopedEnvChange home("HOME", m_testDir);
TestCmdline cmdline("--print-config", server.c_str(), NULL);
cmdline.doit();
CPPUNIT_ASSERT_EQUAL_DIFF("", cmdline.m_err.str());
return cmdline.m_out.str();
}
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
};
files and classes renamed, include statements cleaned up The intention is to get rid of the historic and inconsistent naming of some classes and their corresponding files: * EvolutionSyncClient = class derived from Funambol's SyncClient, * SyncEvolutionConfig = SyncEvolution's config With the strict 'namespace SyncEvo' and the syncevo/ path prefix for most header files it is no longer necessary to have "SyncEvolution" or "Evolution" in the names. This patch thus renames as follows: EvolutionSyncClient => SyncContext EvolutionSmartPtr => SmartPtr SyncEvolutionCmdline => Cmdline SyncEvolutionConfig => SyncConfig SyncEvolutionUtil => util The former EvolutionSyncClient always had a role that went beyond just running a sync, for example it also provided config access. With the upcoming server support it also won't be just a client. Thus the new name "SyncContext". The 'syncevo/' prefix is used throughout the code now. removed whenever the prefix made it clear that the file belongs to SyncEvolution. This helps finding incorrect include paths. Quotes should be used exclusively for SyncEvolution files which don't have a specific prefix yet (test.h, config.h) to help identifying them.
2009-10-05 14:49:32 +02:00
SYNCEVOLUTION_TEST_SUITE_REGISTRATION(CmdlineTest);
infrastructure for config tree testing: scanFiles/createFiles git-svn-id: https://zeitsenke.de/svn/SyncEvolution/trunk@542 15ad00c4-1369-45f4-8270-35d70d36bdcd
2008-03-28 23:32:00 +01:00
#endif // ENABLE_UNIT_TESTS
introduced "namespace SyncEvo" consistently Added syncevo/declarations.h, which has This is now used for all SyncEvolution source files, except for the GTK UI, which is written in plain C. In the library it helps to avoid name clashes. The reason for using defines instead of spelling out "namespace SyncEvo" is twofold: 1. if that should ever become necessary, it is easier to rename the namespace via configure options by changing the define 2. editors don't indent the whole file content
2009-10-02 17:23:53 +02:00
SE_END_CXX