This reverts commit 7d527c6dd8.
It causes link errors on Fedora, see
https://bugzilla.redhat.com/show_bug.cgi?id=1926932
This might be a compiler bug, but as this is a not particular
important size optimization, removing it is the easiest fix.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
This addresses two different warnings from Fedora Rawhide:
/srv/runtests/work/sources/syncevolution/src/syncevo/SyncContext.cpp: In member function 'std::string SyncEvo::XMLFiles::get(SyncEvo::XMLFiles::Category)':
/srv/runtests/work/sources/syncevolution/src/syncevo/SyncContext.cpp:2390:28: error: loop variable 'entry' of type 'const StringPair&' {aka 'const std::pair<std::__cxx11::basic_string<char>, std::__cxx11::basic_string<char> >&'} binds to a temporary constructed from type 'std::pair<const std::__cxx11::basic_string<char>, std::__cxx11::basic_string<char> >' [-Werror=range-loop-construct]
2390 | for (const StringPair &entry: m_files[category]) {
| ^~~~~
/srv/runtests/work/sources/syncevolution/src/syncevo/SyncContext.cpp:2390:28: note: use non-reference type 'const StringPair' {aka 'const std::pair<std::__cxx11::basic_string<char>, std::__cxx11::basic_string<char> >'} to make the copy explicit or 'const std::pair<const std::__cxx11::basic_string<char>, std::__cxx11::basic_string<char> >&' to prevent copying
This fails because StringPair has non-const members. By using "auto",
we get rid of the need to define and pick exactly the right type.
/srv/runtests/work/sources/syncevolution/src/syncevo/SyncConfig.cpp: In member function 'void SyncEvo::SyncConfig::removeSyncSource(const string&)':
/srv/runtests/work/sources/syncevolution/src/syncevo/SyncConfig.cpp:2552:36: error: loop variable 'peer' creates a copy from type 'const string' {aka 'const std::__cxx11::basic_string<char>'} [-Werror=range-loop-construct]
2552 | for (const std::string peer: m_tree->getChildren(m_contextPath + "/peers")) {
| ^~~~
/srv/runtests/work/sources/syncevolution/src/syncevo/SyncConfig.cpp:2552:36: note: use reference type to prevent copying
2552 | for (const std::string peer: m_tree->getChildren(m_contextPath + "/peers")) {
| ^~~~
| &
We could have used "auto" also instead of "std::string", but here it
doesn't save that much typing and is more readable. We just have to
use a reference.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Recent g++ on Fedora Rawhide warns that the dynamic cast result
is used without NULL check. We know that the cast must succeed,
so a static cast is more appropriate.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Some older version of libsecret.h lacked `extern "C"`. Adding
that manually now causes compile errors on Fedora Rawhide and thus
has to be removed:
/usr/include/c++/11/type_traits:480:3: error: template with C linkage
480 | template<typename _Tp>
| ^~~~~~~~
/srv/runtests/work/sources/syncevolution/src/backends/gnome/GNOMEPlatform.cpp:24:1: note: 'extern "C"' linkage started here
24 | extern "C" {
| ^~~~~~~~~~
In file included from /usr/include/glib-2.0/glib/gmacros.h:241,
from /usr/lib64/glib-2.0/include/glibconfig.h:9,
from /usr/include/glib-2.0/glib/gtypes.h:32,
from /usr/include/glib-2.0/glib/galloca.h:32,
from /usr/include/glib-2.0/glib.h:30,
from /usr/include/libsecret-1/libsecret/secret.h:18,
from /srv/runtests/work/sources/syncevolution/src/backends/gnome/GNOMEPlatform.cpp:25:
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
The builtin GtkApplication support is almost equivalent and allows
getting rid of the deprecated libunique which isn't available anymore
in recent Linux distros.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
The --as-needed linker flag didn't work anymore on recent Linux
distros, with the result that libcppunit became a library dependency
of the syncevolution binaries although they didn't need it.
A better approach anyway is to only link the lib when it is expected
to be used (unit testing or in client-test).
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Developed originally by Milan Crha for Fedora, copied and updated
by Patrick Ohly.
From Milan:
There are going to be made huge libecal API changes, as huge as it
deserved a version bump from 1.2 to 2.0, and together with it a small
libebook API changes, most likely being part of the evolution-data-
server 3.33.2 release, which is planned for May 20. More about this can
be found here:
https://mail.gnome.org/archives/desktop-devel-list/2019-April/msg00016.html
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Newer clang (or was it gcc?) warn about catching exceptions by value
which have virtual methods. This shouldn't have mattered here because
the exception values where not really used, but using a const
reference is better nonetheless.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
The signond pipe helper class uses deprecated glib methods. Not sure
whether that wasn't an option at the time that it was written, but
nowadays glib has classes which can be used instead.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Having to specify the type of an iterator is annoying and does not
really add clarity to the code. With C++11 we can use "auto"
instead and in some cases remove helper typedefs.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
This saves some space (total number of blocks for SyncEvolution object
files when building statically down from 421588 to 413300) and
presumably also build times (not measured).
However, it did not work for all templates, leading to link errors
when trying to add std::map and std::pair of strings. It probably also
does not make sense for templates where only some functionality is
used.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
NULL is ambiguous (can be integer and pointer) and using it as
terminator for vararg list of pointers without explicit casting to a
pointer was downright incorrect. nullptr fixes that.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Several headers were no longer needed resp. could be replaced by more
specific ones like noncopyable.hpp.
boost::assign mostly can be replaced with initialization lists and
boost::tuple with std::tuple.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
This allows us to get rid of an external dependency. Mostly std::regex
works, but there are limitations that have to be worked around:
- no multiline support in C++11
- conversion of groups to non-string types has to be done manually
While at it, use raw strings to get rid of excessive backslash
escaping.
pcrecpp::StringPiece was used as a general-purpose utility class. Now
we have our own implementation.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
std::unique_ptr usually can be used instead. std::vector also works
for arrays and even has a data() method as part of the official API in
C++11.
For historic reasons, the functions creating SyncSources returned
plain pointers. We are breaking the API now, so we might as well fix
that.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
This was committed to SVN, probably because it might have become
relevant again. Now it is long obsolete.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
We can use std::shared_ptr and std::function instead now.
Lambdas are usually a better alternative to boost/std::bind. The
downside is the need to explicitly specify parameters completely. When
inlining callbacks entirely with lambdas, duplication of that
parameter list can be avoided.
Whenever possible, use std::make_shared to construct objects that are
tracked by std::shared_ptr.
Some objects need a std::weak_ptr during object destruction. For that
we have to use our own implementation of std::enable_shared_from_this,
with a matching creator function. The additional benefit is that we
can get rid of explicit static "create" methods by making that create
function a friend.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
boost/foreach.hpp is no longer needed, range-based loops work
the same. With some helpers, even reverse iteration and
boost::make_split_iterator() can be handled the same way.
"auto" makes it possible to avoid explicitly spelling out the
expected type.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Lambdas without variable capture are guaranteed to be compatible with
plain C functions, so we can use them as callbacks. That keeps the
code closer together and avoids having to declare helper methods as
part of the public class.
In some cases the static method is the actual code, in which case only
"nothrow()" gets replaced with "noexcept" because it's cleaner and to
mark that the code was looked at and intentionally left as-is.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Universal references and template methods make it possible to pass
through arbitrary callbacks right through to the point where we need
to create a std::function for a plain-C lambda.
Boost is not needed anymore.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
By specifying the list of signal types as template parameters
it becomes possible to use a single implementation. Lambdas
can replace explicit callback methods.
The reimplementation is more flexible and does not enforce
the use of a boost::function. This matches how connectSignal()
was used in practice. Thanks to universal references, the boost::bind
instances get moved directly into the allocated instances that are
attached to the signal handler.
The downside is that the call syntax changes slightly, because
partially specifying template parameters does not work.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Using templates with a varying number of types allows removing
duplicated code for the different cases.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
The OperationWrapperSwitch class can be reduced to just two cases,
which reduces code duplication considerably. Conditional compilation
with "if constexpr" could be used to eliminate the helper class
entirely, but that's C++17.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
The code becomes a lot more readable. One can also set breakpoints
inside the callbacks.
Exception handling in GRunInMain() is better now, with the ability to
rethrow arbitrary exceptions thanks to std::exception_ptr.
Only some usage of boost::lambda in the Akonadi backend remains.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
This aspect of syncevo-dbus-server had been developed for use in an
IVI head unit. It's currently neither used nor tested and no longer
compiles (timeout.h moved, API changes in libfolks).
Instead of trying to keep it working while enhancing the usage of C++,
removing it entirely is easier.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
GIO D-Bus is a more modern and capable implementation. The older one
was only needed on certain old Linux distros (Maemo) which did not
have a recent enough glib.
The reason for removing the old one is that it allows making API
incompatible changes for the C++ D-Bus binding without having to do
that in two places.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
cppcheck started complaining about std::string::find() being
inefficient. boost::starts_with() would indeed be better, but even simpler
and more readable is a whole-string comparison.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Commit 5bafef3957 (included in v1.4)
broke the printing of the messages about Ctrl-C handling, in
particular the hint that using it quickly will switch from suspending
to aborting.
The message code was getting overwritten by a new default case due to
a missing break.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
The SoupTransportAgent modernization in 1.5.3 led to SSL checking
always being enabled because the default changed from disabled to
enabled and SyncEvolution did not set it.
Worse, in older versions it probably (untested) was always disabled
because it was not enabled either.
Now the checking of SSL is always set explicitly and thus always
mirrors the SyncEvolution configuration.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
set/getenv() are not thread-safe, and a recent bug report via private
email shows that this does cause segfaults:
Thread 4 (Thread 19251):
....
Thread 1 (Thread 19311):
...
In this case, DLT was used and the setenv call was setting the
LIBSYNTHESIS_<context> variables.
The solution is to avoid setenv() in code which might run in parallel
to other threads:
- DLT-related variables are set at the beginning of
syncevo-dbus-server startup which then gets inherited
by syncevo-dbus-helper and in the environment prepared
for syncevo-local-sync (because the latter might run with
a different log level)
- the default for SYNCEVOLUTION_PBAP_SYNC is now "incremental"
everywhere and SyncContext is told about the special
mode where it needs to keep photo data differently, i.e. setting
SYNCEVOLUTION_PBAP_SYNC in dbus-sync.cpp for PBAP syncing is
no longer necessary
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Jonas Smedegaard
Patrick Ohly
Milan Crha
deloptes