Changes in version 2.1, 2005-09-16:
* Bison-generated parsers now support the translation of diagnostics like
"syntax error" into languages other than English. The default
language is still English. For details, please see the new
Internationalization section of the Bison manual. Software
distributors should also see the new PACKAGING file. Thanks to
Bruno Haible for this new feature.
* Wording in the Bison-generated parsers has been changed slightly to
simplify translation. In particular, the message "memory exhausted"
has replaced "parser stack overflow", as the old message was not
always accurate for modern Bison-generated parsers.
* Destructors are now called when the parser aborts, for all symbols left
behind on the stack. Also, the start symbol is now destroyed after a
successful parse. In both cases, the behavior was formerly inconsistent.
* When generating verbose diagnostics, Bison-generated parsers no longer
quote the literal strings associated with tokens. For example, for
a syntax error associated with '%token NUM "number"' they might
print 'syntax error, unexpected number' instead of 'syntax error,
unexpected "number"'.
* Version 0.3.8 (released 2005-09-27)
** New: buildSmbNtlmAuthRequest_noatsplit, buildSmbNtlmAuthResponse_noatsplit.
These functions can be used when your username contains '@' and you do
not wish that the remainder of the string is treated as a REALM value.
Suggested by David Leonard <davidl@vintela.com>.
** Uses memset instead of bzero.
** Fix license typo in libntlm.spec.
** Made the COPYING file contain LGPL, not GPL.
** Updated gnulib files.
** API and ABI modifications.
buildSmbNtlmAuthRequest_noatsplit: ADD.
buildSmbNtlmAuthResponse_noatsplit: ADD.
Possibly incompatible changes:
- hook_note_commit and hook_note_netsync_revision_received
take a new argument containing the text of the revision that
was received. (Timothy Brownawell <tbrownaw@gmail.com>)
- 'cat FILENAME' now acts like the old 'cat file REV
FILENAME'; use new commands 'automate get_revision',
'automate get_manifest', 'automate get_file' to fetch
objects by hash. (Grahame Bowland <grahame@angrygoats.net>)
General improvements:
- .mt-ignore support (Martin Dvorak
<jezek2@advel.cz>, Timothy Brownawell <tbrownaw@gmail.com>)
- much work on making monotone more i18n friendly (Benoît
Dejean <benoit@placenet.org>, Matt Johnston
<matt@ucc.asn.au>)
- support for more interactive merge tools:
- FileMerge.app (comes with OS X) (Marcel van der Boom
<marcel@hsdev.com>)
- TortoiseMerge (Win32; comes with TortoiseSVN) (Matthew
Gregan <kinetik@orcon.net.nz>)
- rename and drop now actually perform the specified rename or
deletion when the argument --execute is passed. (Richard
Levitte <richard@levitte.org>)
- 'help' command, same as --help (Matt Johnston
<matt@ucc.asn.au>).
- 'usher' support: experimental method for proxying multiple
netsync servers through a single port (similar concept to
vhosts) (Timothy Brownawell <tbrownaw@gmail.com>)
- support long passphrases (Matt Johnston <matt@ucc.asn.au>)
- Faster binary file detection (Eric Anderson
<anderse-monotone@cello.hpl.hp.com>)
- netsync speedups:
- when handling large files (Eric Anderson
<anderse-monotone@cello.hpl.hp.com>)
- when handling many branches (Marcel van der Boom
<marcel@hsdev.com>)
- new system to allow crash logs to contain not just execution
traces, but also dumps of data being handled when the error
was detected -- greatly improves debuggability of user
crashes.
- complete rework of path handling code, for clarity,
robustness, and speed. No user visible changes, except for
the many bugs fixed. (Special thanks to Matthew Gregan
<kinetik@orcon.net.nz> and Grahame Bowland
<grahame@angrygoats.net>.)
- however, if you have non-normalized paths in your history
(symptom: fresh pulls with 0.18 work, but fresh pulls with
0.19 do not), then 0.23 will report an error and refuse to
handle the affected revisions. Since it is believed this
only affects one project, and to conserve core developer
time, implementing a migration was put off for now. If
this causes problems or for more details, please send an
email to monotone-devel@nongnu.org.
- as always, many small bug fixes, speedups, and improved
messages.
New translations:
- fr (Benoît Dejean <benoit@placenet.org>)
- ja (Satoru SATOH <ss@gnome.gr.jp>)
Other new monotone-related projects since 0.22:
- mtsh by Timothy Brownawell:
https://netfiles.uiuc.edu/brownawe/www/mtsh/
GTK+ wrapper for monotone focusing on working copy
operations -- add/drop/revert/rename/commit/update/diff and
browsing. Has a mechanism for per-file commit comments.
- "dumb server" support by Nathaniel Smith (share your
monotone repositories via HTTP/FTP, no netsync daemon
needed):
http://viewmtn.angrygoats.net//branch.psp?branch=net.venge.monotone.dumb
Still needs a command-line frontend to be usable, among
other things. Help wanted. In python.
- m7 by Larry Hastings <larry@hastings.org>
http://www.midwinter.com/~lch/programming/m7/
Experimental drop-in command-line wrapper for monotone.
Uses certs to add local incrementing version numbers, and an
enhanced annotate front-end.
* Version 0.5.19 (released 2005-09-19)
** The test for setlocale and nl_langinfo has now been separated.
The autoconf script now test for locale.h, setlocale and
nl_langinfo(CODESET) independently.
** Gnulib updates, fixes for getopt.
** Java manuals in doc/java/ are now generated by Gjdoc from GNU Classpath.
** Kaffe is used to link the pre-built libidn-*.jar file.
** Translation updates.
** API and ABI is backwards compatible with the previous version.
Version 6.4 05-Sep-05
---------------------
1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
"--" to be printed when multiple files were scanned, even when none of the
-A, -B, or -C options were used. This is not compatible with Gnu grep, so I
consider it to be a bug, and have restored the previous behaviour.
2. A couple of code tidies to get rid of compiler warnings.
3. The pcretest program used to cheat by referring to symbols in the library
whose names begin with _pcre_. These are internal symbols that are not
really supposed to be visible externally, and in some environments it is
possible to suppress them. The cheating is now confined to including
certain files from the library's source, which is a bit cleaner.
4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
file's purpose clearer.
5. Reorganized pcre_ucp_findchar().
ACR tries to replace autoconf functionality generating a full-compatible
'configure' script (runtime flags). But using shell-script instead of m4.
This means that ACR is faster, smaller and easy to use.
parallel port. This package is for Byron Jeff's adapted version for Brian
Lane's original software, and supports the Trivial Low Voltage Programmer
(TLVP).
change (it breaks for callbacks).
Inspired from the commit to solve the same issue on the 3.4 branch of RT.
Bump PKGREVISION and Mason version requirement.
makes -avoid-version work properly on platforms using $versuffix2 (e.g., OpenBSD,
FreeBSD, DragonFly). Bump PKGREVISION (and will bump LIBTOOL_REQD shortly).
MirOS change by Benny Siegert <bsiegert@mirbsd.org>.
Noted in private mail by Thorsten Glaser <tg@66h.42h.de>.
makefile issues. There are still some problems on solaris, but this
gets you further. Don't mark as available yet on solaris since it
still doesn't compile.
optimize_gcc.mk needs to know the name of the package that is
currently being built. Unfortunately with pkgsrc there is no way
to do this, and its previous attempts to derive PKGNAME (if not
set) from DISTNAME broke the php usage ext.mk file. Use a temp
variable to do the same thing without breaking other pkgsrc
assumptions
no longer correct since update to libevent 1.x; it now uses libtool and
generates a shlib.
Remove the offending bl3 line, and bump all dependents' PKGREVISIONs, since
the binary pkg changes for any OS that doesn't have a sufficient builtin
libevent version (or the package has requested a non-builtin version).
Changes:
- Major security fix (iDEFENSE Security Advisory IDEF1099 - Stack Overflow
Vulnerability)
- Major security fix from Palasik Sandor (LZX decompression buffer overrun)
- Bugfix/enhancement from David Huseby to make the "what" flags to
chm_enumerate work correctly, and to pass the flags along to the callback
function (via the chmUnitInfo structure) so that the callback doesn't
need to re-parse the filename.
- Compilation fixes for x86-64 from Vitaly V. Bursov.
- Miscellaneous fixes to the configure script, including some significant
cleanup by Vadim Zeitlin. The changes from Vadim should also allow the
configure script to correctly configure the build on OS X, where it was
previously failing to note that pread64 doesn't work.
- Minor update to the Makefile.in to do a mkdir before the install, in case
the specified INSTALLPREFIX directory is non-existent
* The search paths for allegro's cfg files and modules are hardcoded,
add the proper pkgsrc paths to them.
* allegro uses 'ps' internally with '-f'. NetBSD doesn't have a '-f'.
Remove it from the ps flags on NetBSD.
Bump PKGREVISION.
packages.
o Drop the frontend- and backend- options names and replace them with
simpler ones, now that we have the PKG_OPTIONS_GROUP mechanism it's
much more easier.
o Brick me for not having done that earlier.
* libglade support (see manpage) (Nicolas Brugier)
* Fixes to property parsing and allow user created
parameters (Jean-Yves Lefort)
* Add support for prealloc and allow bitfields (K. Haley)
* Don't reference TYPE_SELF when generating privates (Big Thor)
* Added support for direct color 8-bpp surfaces
* Altivec optimized blitters
* SDL.dll works on Windows 95 again
* Dropping a document onto an SDL app passes it as a command line parameter in MacOS X
* Fixed crash trying to allocate hardware surfaces on MacOS X
* Fixed short read problem with SDL_RWFromMem()
* Fixed gcc parse errors in SDL_audio.h on Windows
* Added support for Tru64 UNIX 4.X
* Improved RISC OS support
* Numerous improvements to the Atari port
* SDL_OPENGLBLIT is officially obsolete
Version 6.3 15-Aug-05
---------------------
1. The file libpcre.pc.in did not have general read permission in the tarball.
2. There were some problems when building without C++ support:
(a) If C++ support was not built, "make install" and "make test" still
tried to test it.
(b) There were problems when the value of CXX was explicitly set. Some
changes have been made to try to fix these, and ...
(c) --disable-cpp can now be used to explicitly disable C++ support.
(d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
backslash in a target when C++ was disabled. This confuses some
versions of "make", apparently. Using an intermediate variable solves
this. (Same for CPP_LOBJ.)
3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
(non-Windows) now includes $(CFLAGS) because these flags are sometimes
necessary on certain architectures.
4. Added a setting of -export-symbols-regex to the link command to remove
those symbols that are exported in the C sense, but actually are local
within the library, and not documented. Their names all begin with
"_pcre_". This is not a perfect job, because (a) we have to except some
symbols that pcretest ("illegally") uses, and (b) the facility isn't always
available (and never for static libraries). I have made a note to try to
find a way round (a) in the future.
*1.1.1* (11 July, 2005)
* Added more efficient implementation of the development mode reset of classes #1638 [Chris McGrath]
*1.1.0* (6 July, 2005)
* Fixed conflict with Glue gem #1606 [Rick Olson]
* Added new rules to the Inflector to deal with more unusual plurals mouse/louse => mice/lice, information => information, ox => oxen, virus => viri, archive => archives #1571, #1583, #1490, #1599, #1608 [foamdino@gmail.com/others]
* Fixed memory leak with Object#remove_subclasses_of, which inflicted a Rails application running in development mode with a ~20KB leak per request #1289 [c.r.mcgrath@gmail.com]
* Made 1.year == 365.25.days to account for leap years. This allows you to do User.find(:all, :conditions => ['birthday > ?', 50.years.ago]) without losing a lot of days. #1488 [tuxie@dekadance.se]
* Added an exception if calling id on nil to WhinyNil #584 [kevin-temp@writesoon.com]
* Added Fix/Bignum#multiple_of? which returns true on 14.multiple_of?(7) and false on 16.multiple_of?(7) #1464 [Thomas Fuchs]
* Added even? and odd? to work with Bignums in addition to Fixnums #1464 [Thomas Fuchs]
* Fixed Time#at_beginning_of_week returned the next Monday instead of the previous one when called on a Sunday #1403 [jean.helou@gmail.com]
* Increased the speed of indifferent hash access by using Hash#default. #1436 [Nicholas Seckar]
* Added that " " is now also blank? (using strip if available)
* Fixed Dependencies so all modules are able to load missing constants #1173 [Nicholas Seckar]
* Fixed the Inflector to underscore strings containing numbers, so Area51Controller becomes area51_controller #1176 [Nicholas Seckar]
* Fixed that HashWithIndifferentAccess stringified all keys including symbols, ints, objects, and arrays #1162 [Nicholas Seckar]
* Fixed Time#last_year to go back in time, not forward #1278 [fabien@odilat.com]
* Fixed the pluralization of analysis to analyses #1295 [seattle@rootimage.msu.edu]
* Fixed that Time.local(2005,12).months_since(1) would raise "ArgumentError: argument out of range" #1311 [jhahn@niveon.com]
* Added silencing to the default Logger class
*1.0.4* (19th April, 2005)
* Fixed that in some circumstances controllers outside of modules may have hidden ones inside modules. For example, admin/content might have been hidden by /content. #1075 [Nicholas Seckar]
* Fixed inflection of perspectives and similar words #1045 [thijs@vandervossen.net]
* Added Fixnum#even? and Fixnum#odd?
* Fixed problem with classes being required twice. Object#const_missing now uses require_dependency to load files. It used to use require_or_load which would cause models to be loaded twice, which was not good for validations and other class methods #971 [Nicholas Seckar]
= Cmd Changelog
== Version 0.7.2
* Breaking the assignment to READLINE_SUPPORTED over multiple lines did not
appease RDoc. Fortunately Dave Thomas showed me several ways that I could
accomplish the same thing whilst appeasing RDoc. Sadly another minor bug fix
release :\ [Dave Thomas]
== Version 0.7.1
* Broke assignment to READLINE_SUPPORTED over several lines as removing the
semicolons resulted in the incorrect behavior, and having the semicolons made
RDoc fail to parse the file. [Jim Weirich]
= Rake Changelog
== Version 0.5.4
* Added double quotes to the test runner.
* Added .svn to default ignore list.
* Updated FileList#include to support nested arrays and filelists.
Remove the files/ant script, by patching the bin/ant script, as ant people
worked on it.
I'm taking maintainership for this one
Fix PR 29038 and PR 29338, apache-ant now can use plugins
Fix PR 30332 (update to 1.6.5)
Approved by jwise@
Dates are complex enough without times and timezones. This module may
be used to create simple date objects. It handles:
Validation:
Reject 1999-02-29 but accept 2000-02-29.
Interval arithmetic:
How many days were between two given dates? What date comes N days
after today?
Day-of-week calculation:
What day of the week is a given date?
It does NOT deal with hours, minutes, seconds, and time zones.
According to http://pvs.csl.sri.com/announcements/pvs3.2-release-notes.shtml
this contains a number of bug fixes as well as the following improvements
and added features:
-Startup Script Update
-Theory Interpretation Enhancements
-References to Mapped Entities
-Cleaning up Specifications
-Binary Files
-Generating HTML
-Default Strategies
-Better handling of TCCs in Proofs
-typepred! rule and all-typepreds strategy
-grind-with-ext and reduce-with-ext
-New forward chain commands
-TeX Substitutions
-add-declaration and IMPORTINGs
-Prelude additions
1) Simplify the way how an emacs version is picked when no emacs
is installed, but a user try to install an Emacs Lisp package.
Just pick up the version set as EMACS_TYPE than searching for
versions already installed etc. If the EMACS_TYPE version is
not supported by the Emacs Lisp Package, just fail. EMACS_TYPE
be default to GNU Emacs 21.
(In other words, users should set EMACS_TYPE as they want.
Otherwise GNU Emacs 21 is used.)
2) All Emacs Lisp Packages *must* prepend EMACS_PKGNAME_PREFIX to
a) the PKGNAME itself, and b) PKGNAME in its dependency lines.
EMACS_PKGNAME_PREFIX is expanded to "xemacs-" when XEmacs is
used. This keeps dependency graph of Emacs-Lisp-packages-
installed-for-XEmacs consistent.
3) Document EMACS_* variables as much as possible.
4) Provide more cookies for PLIST. Maybe utilized later.
Note that the 2) change doesn't affect the default, GNU Emacs 21
behaviour. So no version / revision bumps in this commit.
changes:
-made GSF_CLASS_FULL and GSF_DYNAMIC_CLASS_FULL really full (add
_finalize and class_finalize arguments).
-Handle 0x8000 and 0x8001 codepages.
-bugfixes
also fix installation of gconf schema
build and run on other platforms when MAINTAINER knows what magic
Makefile glue is required. This is from maintainer's discussion
on tech-pkg.
Remove patch-af. Use LD_LIBS instead, which the build already knows
about.
Add custom settings for above platforms so they install correctly.
Idea provided by maintainer on tech-pkg. I tweaked it more.
I tested on NetBSD 2.0.2, Linux and DragonFly.
Also remove blank line from end of Makefile.
following packages because they don't use "it" and "setup check"
as BUILD_TARGET and INSTALL_TARGET, respectively:
devel/bglibs
mail/mailfront
mail/qmail-qfilter
sysutils/memtester
sysutils/qlogtools
Since there's only one consumer of BGWARE_INSTALLER, and future
bgware will use bg-installer from bglibs, move the do-install
routine directly into qlogtools.
- Fixed numberic export glitch in vpwentry_export.
- Fixed dependancies to let "make -j" work properly.
- Added some pre-computed .lib files (such as socket.lib).
backslashes anymore. A single backslash is enough. Changed the
definition in all affected packages. For those that are not caught, an
additional check is placed into bsd.pkginstall.mk.
are identical (e.g., msgid == msgid_plural) by ignoring the duplicates.
In fact, this is a rewrite of the script, since I couldn't understand the
old one (ew).
Also change the way we use it to only pull it in the build if the real
msgfmt does not support plurals (i.e., it's older than 0.10.36).
Fixes PR pkg/30596 and PR pkg/30938 (both related to epiphany).
It also fixes the build of evolution-data-server and probably others.
- Fixed negative indexing bugs in str_findprev* and ibuf_getstr_crlf
functions. Thanks Gerrit Pape.
- Added the capability to bg-installer to optionally run commands based
on the existance of the source (or another) file.
- Fixed bg-installer to not create directories in "dry-run" mode.
This is not necessary for normal scons use, but there is an extension
"bksys" which needs XML.
Since scons users do not necessarily use Python theirselfes, a buildtime
dependency on px-xml-for-the-python-version-used-by-scons would be needed
which I don't see an easy and robust way for.
bump PKGREVISION
- Added the capability to bg-installer to handle multiple destinations
within the same source list.
- Renamed the "setmode" function in bg-installer that was causing
compile problems on *BSD systems.
- Added a sysdep for struct timespec and clock_gettime, which are
defined if not present.
* Fixed bugs in the exiv2, OpenOffice, and OLE2 plug-ins.
* Static relocatable glib no longer required.
* getKeywords2 function is now included in the code.
Changes:
* Fixes for LWP-2.0
* Reduced size of the medium packets to 1500.
* Explicitly clear the EchoTimestamp when we receive an out-of-order
SFTP data packet to avoid sending a stale timestamp value.
* Make sure we retransmit at least one packet once we hit EOF so that we
can continue to make progress.
New in 2.0:
* Use ucontext.h provided functions for thread creation and switching.
In case those are not available use sigaltstack for thread creation
and sigsetjmp/siglongjump for thread switching, and if sigaltstack
doesn't exist fall back on the old process.S assembly code.
* Simplified pthread support a bit, now it builds and runs on cygwin.
* RPM .spec file fix to allow non-root users to build (Matthew Rich).
* Included AIX process.S assembly, since I don't know the right ifdefs,
it isn't actually part of process.S yet.
* This library is binary compatible with previous releases. But some of
the prototypes of callback functions got cleaned up a bit so that
building against this version of LWP will give some trivially fixable
compile warnings or errors.
New in 1.13:
* Stack switching code for AMD64 processors (Brett Lymn).
New Libraries
* Iostreams Library: Framework for defining streams, stream
buffers and i/o filters, from Jonathan Turkanis.
* Functional/Hash Library: A TR1 hash function object that can be
extended to hash user defined types, from Daniel James.
* Parameter Library: Write functions that accept arguments by
name: especially useful when a function has more than one
argument with a useful default value, since named arguments can
be passed in any order.
* Pointer Container Library: Containers for storing heap-allocated
polymorphic objects to ease OO-programming, from Thorsten Ottosen.
* Wave: Standards conformant implementation of the mandated
C99/C++ preprocessor functionality packed behind an easy to use
iterator interface, from Hartmut Kaiser.
Updated Libraries
* Assignment Library: Support for Pointer Container Library and
new efficient functions ref_list_of() and cref_list_of() for
generating anonymous ranges.
* Bind Library: Bind expressions now support comparisons and
negation. Example: bind(&X::name, _1) < bind(&X::name, _2).
* Date-Time Library:
o Added local time and time zone classes.
o Added format-based Input/Output facets.
o For a complete list of changes, see the library change history.
* Graph Library: Introduced several new algorithms and improved
existing algorithms:
o Experimental Python bindings, from Doug Gregor and Indiana
University.
o floyd_warshall_all_pairs_shortest_paths, from Lauren Foutz
and Scott Hill.
o astar_search, from Kristopher Beevers and Jufeng Peng.
o fruchterman_reingold_force_directed_layout, from Doug
Gregor and Indiana University.
o biconnected_components and articulation_points, from
Jeremy Siek, Janusz Piwowarski, and Doug Gregor.
o sequential_vertex_coloring has been updated, tested, and
documented.
o gursoy_atun_layout, from Jeremiah Willcock and Doug Gregor
of Indiana University.
o king_ordering, from D. Kevin McGrath of Indiana University.
o cuthill_mckee ordering has been recast as an invocation of
breadth_first_search and now supports graphs with multiple
components.
o dijkstra_shortest_paths now uses a relaxed heap as
its priority queue, improving its complexity to O(V log V) and
improving real-world performance for larger graphs.
o read_graphviz now has a new, Spirit-based parser that
works for all graph types and supports arbitrary
properties on the graph, from Ron Garcia. The old,
Bison-based GraphViz reader has been deprecated and will
be removed in a future Boost release. write_graphviz also
supports dynamic properties.
o subgraph: get_property now refers to the subgraph
property, not the root graph's property.
o See the history for additional changes and bug fixes.
* Multi-index Containers Library:
o New hashed indices.
o Added serialization support.
o For a complete list of changes, see the library release notes.
* Program Options Library:
o Option descriptions are now printed with word wrapping.
o Command line parser can bypass unregistered options,
instead of throwing.
o Removed support for "implicit" (optional) values.
o New customization method
'command_line_parser::extra_style_parser'. Unlike
'additional_parser', allows the user to parse several
tokens and return a vector of options, not just a single
option.
o Work with disabled exceptions.
* Property Map Library: Introduced the dynamic properties class,
which provides dynamically-typed access to a set of property maps.
* Random Number Library: improved initialization for
mersenne_twister, algorithm by Makoto Matsumoto and Takuji
Nishimura, implemented for Boost by Jens Maurer.
Note: All test vectors for mersenne_twisters constructed or
seeded without parameters or with a single unsigned int parameter
become invalid.
* Range Library: Minor addition of convenience functions to
iterator range like front(), back() and operator[]().
* Regex Library:
o Rewritten front end parser now supports (?imsx-imsx)
constructs, plus lookbehind assertions and conditional
expressions.
o Thin wrapper classes improve integration with MFC/ATL code.
o Full (optional) Unicode support via the ICU library.
Refer to the regex history page for more information on these
and other small changes.
* Serialization Library:
o DLL version.
o Auto-linking.
o Serialization of variants.
o Improved seialization of shared pointers.
* Signals Library: added slot blocking/unblocking, from Frantz
Maerten. Huge improvements to signal invocation performance from
Robert Zeh.
This update has been tested on NetBSD 2.0.2, 3.0_BETA and current.
changes:
* fix a potential mem leak content_type and osip_message_parse.
* fix osip_body_clone method (add terminating NULL)
* fix for binary support when multipart is used.
* automatic check for reliable protocol SCTP and TLS.
* avoid <winsock.h> inclusion so user can (and must) now include either
<winsock.h> or <winsock2.h> before including <osip/osip2.h>
file's sole purpose was to provide a dependency on pkg-config and set
some environment variables. Instead, turn pkg-config into a "tool"
in the tools framework, where the pkg-config wrapper automatically
adds PKG_CONFIG_LIBDIR to the environment before invoking the real
pkg-config.
For all package Makefiles that included pkg-config/buildlink3.mk, remove
that inclusion and replace it with USE_TOOLS+=pkg-config.
* Fixed build for SUN Forte C++ 5.5
* Fixed build for MSVC++ 7.1
* Fixed crash when using --no-inline with g++.
2.0.15:
* g++ 3.2 (and Mac OS X g++ 3.3) build fix.
(Paul Pogonyshev)
* Compose: Fix slot lifetime regression introduced in
2.0.9. (Philip Langdale)
* tests: Small ISO C++ correctness fix (Marek Rouchal)
* Don't specify unused function parameter names.
(Andris Pavenis)
2.0.14:
* SUN Forte 5.7 build fix for ambiguity when using
inner template class. However, you still need the
patch in bug #302098 to finish the build.
2.0.13:
* signal_emit::emit(): Ensure the correct order of
destruction of the member variables, to avoid a leak.
(Andreas Ames, bug #306249)
* Allow recursive signal emission again.
(Neal E. Coombes, bug #303896)
* SUN Forte CC 5.5 build fixes:
- test_compatibility minor fix.
- visit_each() template specializations:
Mention the bool I_derives_trackable template type,
(Friedemann Kleint, bug #305647)
- Check for the non-standard SUN reverse_iterator,
and use alternative code if necessary.
(Murray Cumming)
2.0.12:
* Fixes crashes when using virtual inheritance, particularly
with bound by-reference parameters, caused by casting
from derived to base when the derived destructor has run.
(Régis Duchesne)
This might affect non-g++ compilers, so do tell us about
any problems.
2.0.11:
* Build fixes for SUN Forte, Tru64
(Murray Cumming), and MSVC++ (Cedric Gustin).
2.0.10:
* tests: Include <new> to avoid unresolved symbols on Tru64.
(Tim Mooney)
* When signal handlers are connected made during an emit
of the same signal, prevent them from being called in the
same emit, to prevent infinite loops.
(Neal E. Coombes)
* Performance improvement in a corner case.
(Neal E. Coombes).
2.0.9:
* sigc::bind() now works with the AIX and Tru64 compilers.
See the comments in sigc++/visit_each.h: visit_each_type()
if you have compilation problems.
(Murray Cumming)
* sigc::var() is now documented. (Roger Ferrer Ibáñez)
2.0.8:
* Maybe avoid (incorrect) warning with g++ 3.3.5.
(Murray Cumming)
* Fix namespace ambiguity when using multiple
major versions of libsigc++. (Liza Klerck)
2.0.7:
* Now builds with the following compilers, in addition to
the existing GNU g++, SUN Forte CC 5.5, MSVC++ .Net 2003,
and Intel compilers:
- IBM AIX xlC v7
- Tru64 C++ V6.5-042
- IRIX MIPSpro 7.4.2m
(Older versions of all these compilers might also work.)
(Murray Cumming, www.thewrittenword.com)
* MSVC++ .Net 2003 build improvements.
(Cedric Gustin, Timothy M. Shead)
* Replace C-style casts with reinterpret_cast<> and
static_cast<>. (e97_far at e.kth.se).
* Documentation: Added manual, based on the manual in
libsigc++ 1.2, but updated for the new API.
(Murray Cumming)
2.0.6:
* Fixed a memory leak in sigc::slot.
* Fixed compilation for gcc-3.4.
* Fixed compilation for Intel C++ compiler (upgraded libtool).
* Fixed project files for MSVC .Net (Timothy M. Shead).
* Fixed segfaults when compiled with MSVC .Net 2003 (moved
all calls to new and delete into non-inline library code).
* In the compatibility module use correct bound_mem_functor
variants for const (volatile) methods when creating a slot.
* Minor documentation fix.
* Resolved bugs: #152327#148744#152323#151404#153143
2.0.5:
* Distribute pregenerated configuration header for MSVC .Net.
2.0.4:
* Fixed warnings and compiler errors in the test cases.
* Added a new test case (Murray Cumming).
* Fixed 'hello_world' example.
* Don't test optional features that fail with the Sun FORTE.
* Fixes for the Sun FORTE to compile out-of-the-box
(Damien Carbery, Murray Cumming, Martin Schulze).
* Fixes for MSVC to build a DLL out-of-the-box (James Lin).
* Improved compiler specific configuration during 'configure'.
* Added rmp description file libsigc++-2.0.spec (Eric Bourque).
* Minor documentation improvements (Murray Cumming).
* Resolved bugs: #147311#147313#147391#144846#145541
2.0.3:
* Fix segfault on emission of unconnected signal.
* Test emission of unconnected signals in the test case.
* Suppress compiler warning at dynamic_cast<>-test for good.
(Help from Christof Petig and Timothy M. Shead.)
2.0.2:
* Suppress compiler warning in compatibility module at
dynamic_cast<>-test (fix suggested by Timothy M. Shead).
* If a custom accumulator is specified invoke it on signal
emission even if the signal's slot list is empty. (This used
to be the case in libsigc++-1.2 as pointed out by Timothy.)
2.0.1:
* Fixed serious bug in reference counting in sigc::signal_base::impl().
* Fixed SigC::Object-derivation check in SigC::slot() compatibility module.
* Fixed compilation on Apple gcc 3.3 (assisted by Spundun Bhatt).
* Fixed configure check for gcc 3.4 (Murray Cumming).
2.0.0:
* Implemented sigc::connection::blocked() (Murray Cumming).
* Added the scripts directory to the make dist target (Murray Cumming).
* Added more documentation (Martin Schulze).
1.9.16:
* Fixed compiler warning in sigc::connection (Alexander Nedotsukov, Murray Cumming).
* Fixed examples and made them part of the regular build (Murray Cumming).
* Added header sigc++config.h for configure time checks (Murray Cumming).
* Added configure time checks to determine the correct syntax
for explicit template method specializations (Murray Cumming).
* Removed code using partial specializations of overloaded template methods
from test cases. SUN Forte doesn't support this feature (Martin Schulze).
* Fixed compilation for gcc 3.4 (Murray Cumming).
1.9.15:
API additions:
* Add numbered slot# templates.
* Allow for methods of the object's base types to be passed into sigc::mem_fun().
Other fixes and cleanups:
* Make is_base_and_derived template compatible with the SUN Forte.
* Non-template code moved from .m4 macro source to .h/.cc files (Murray Cumming).
* Implementation moved to .cc files (Murray Cumming).
* More fixes for the SUN Forte. Make some more ctors explicit.
1.9.14:
* Added sigc::slot_base::operator bool() (Murray Cumming).
* Build docs directory by default (Murray Cumming).
* Fixed minor doxygen issues (Murray Cumming).
* Fixed compiler warning in signal.h (Murray Cumming).
1.9.13:
* Fixed passing references through sigc::slot (Reported by Jeff Franks).
* Enabled binding of objects to method slots through sigc::bind().
* Reworked sigc::bind() API: Made the template argument for the
parameter position zero-based and optional. Added overloads for
binding of up to 7 arguments at a time when no position is specified.
* Reworked sigc::hide() API: Made the template argument for the
parameter position zero-based and optional.
* Fixed compilation problems with MSVC .Net 2003 (Roel Vanhout).
* Distribute MSVC .Net 2003 project files in the tarballs.
* Improved and extended documentation.
* Minor cleanups.
1.9.12:
* Added adaptor retype(). With this final API addition all adaptors
are in place that are available in libsigc++-1.2.
* Added negation lambda operator. Use STL names for lambda actions.
* Remove formerly disabled support for gcc extension typeof().
* Added project files for MS Visual Studio .Net 2003. (Roel Vanhout)
* Make libsigc++2 compile with .Net 2003. (Roel Vanhout, Martin Schulze)
* Build shared version of libsigc++2 by default. (Cedric Gustin)
* Add support for win32 platform. (Cedric Gustin)
* Install .m4 files. (requested by Ron Steinke)
* Cleaned up functors.
* Restructured and completed documentation of the core library parts.
1.9.11:
API Additions and important bug fixes:
* Compatibility module completed. libsigc++-1.2 filenames are preserved.
* Fixed critical bug in auto-disconnection: don't defer detaching
of a slot from all referred trackables during signal emission.
* Reduced size of slots significantly.
* Fixed support for sigc::ref() in adaptors.
* Fixed sigc::visit_each(): only hit targets that are passed by
reference; pass bound members in bound_member_functor by reference.
* Add lambda actions sigc::{reinterpret,static,dynamic}_cast_
to support explicit parameter conversion.
* Add adaptors sigc::retype_return<>() and sigc::hide_return().
Minor fixes:
* Fixed return type deduction for bind<0>.
libsigc++-1.9.11 should compile with gcc-3.3.
* Fixed copy constructor and operator=() of slot template.
* Fixed a compiler warning in signal_emit#<>::emit().
* Improved test case.
1.9.10:
* Fix compiler issues with gcc-3.3.2 (patch from Jeff Franks).
* Remove compiler check for the gcc extension typeof().
* Simplify bind_functor templates.
* Move definition of struct nil into functor_trait.h.
1.9.9:
* Add a constructor to sigc::connection that takes a slot_base&
to support user defined slot lists like they are used in gtkmm.
* Fix compiler issues with gcc-3.3.2 (reported by Jeff Franks).
1.9.8:
* Add compatibility module that defines namespace SigC.
namespace SigC should be API compatible to libsigc++-1.2.
Currently only the core parts of the library are supported.
Adaptors are still to follow.
* Fix connection::operator=(). Include connection.h in sigc++.h.
* Get rid of namespace functor.
* Rename dependency to destroy_notify_callback.
* Rename trackable::clear() to trackable::notify_callbacks().
* Move slot_base, signal_base, slot_iterator[_buf], slot_list
out of namespace internal. They are public API.
* Add reference counter to signal_impl enabling signals
to share the underlying information.
* Add convenience function signal#::make_slot().
* Get rid of one-letter-parameter-names.
* Get rid of "using namespace ..." in the test cases.
* Add lambda operators subscript ([]) and assign (=).
* Fix is_base_and_derived<> for const types.
* New and updated documentation.
* Add previous announces to file NEWS.
1.9.7:
* Added sigc++/sigc++.h. (Murray Cumming)
* Added member_method example. (Murray Cumming)
* Renamed closure to slot.
* Fixed issues with gcc-3.3. (Adreas Rottmann)
* Removed unnecessary void specializations.
* Made adaptors' operator()() (overload with no arguments) return a value.
* Made visit_each() support adaptors.
* Overhauled return type deduction to make it work without typeof().
* Added convinience macros SIGC_FUNCTORS_HAVE_RESULT_TYPE and
SIGC_FUNCTOR_TRAIT(T_functor, T_result) to make return type deduction system
support 3rd-party funtors.
* Changed syntax of group adaptor from "[functor] % grp([lambdas])" to "group
([functor], [lambdas])".
* Made many fixes to lambda functionality.
* Added var() and constant() lambda creators.
* Added many lambda operators.
* Added ref() which creates a reference wrapper to enable storage of
references in bind and group adaptors.
* Expanded test suite.
* Added documentation. (Corrections by Murray Cumming)
1.9.6:
* First public release of the unstable 2.0 generation.
libsigc++ 2.0 uses modern C++ mechanisms to achieve a highly
flexible, yet typesafe callback system. It supports all features of
libsigc++ 1.2 and improves upon it by:
- No need to specify the number of arguments in signal definitions.
- Connection of any compatible (=implicitly convertable) functor
to a signal.
- Implicit type conversions of parameters during signal emission.
- Lambda adaptor for complete restructuring of functor parameter
lists in one line (subject to changes).
- Signal has a fully featured stl style list interface.
- A convinient accumulator API (replacing the old marshaller API).
- Removal of unnecessary memory management functionality.
- Lightweight class "trackable" for use as base class of your
class hierarchy replaces class "Object".
data loss in libgobject, causing all kinds of annoying glitches in almost
all gtk2-applications (in words of macallan@, who requested this change).
Bump PKGREVISION to 1.
Mon Aug 8 23:23:53 PDT 2005
0.22 release. new crypto library, bug fixes, ui improvements
- switch from crypto++ to botan as underlying crypto library.
this should not cause any user-visible changes; let us know
if it does. special thanks to Matt Johnston
<matt@ucc.asn.au>, Kaushik Veeraraghavan
<kaushikv@gmail.com>, Matthew Gregan
<kinetik@orcon.net.nz>.
- incompatible change to netsync permission hooks: the
get_netsync_anonymous_read_permitted hook has been removed;
instead, get_netsync_read_permitted will be called with a
key name of nil. server administrators should update/review
their configuration
- new option for merge and propagate: --lca. Until we get a
long-term solution to the various 3-way merge problems, this
should be more convenient than using explicit_merge.
- many small improvements to error messages, fixes of minor
annoyances, netsync tickers more accurate, etc.
Sun Jul 17 16:48:26 PDT 2005
0.21 release. bug fixes, performance improvements, and ui
improvements.
- fixes a number of major performance bugs in 0.20's netsync
implementation. special thanks to Matt Johnston
<matt@ucc.asn.au>.
- fixes a number of major bugs in 0.20's (rewritten)
cvs_import command.
- configury kluges to work around g++ 4.0/boost 1.32
incompatibilities. special thanks to Christof Petig
<christof@petig-baender.de>, Matthew Gregan
<kinetik@orcon.net.nz>, Jordan Breeding
<jordan.breeding@mac.com>.
- ui enhancements:
- new netsync option "--exclude": branches are included if
they match any of the given globs, unless they match any
of the given --exclude globs. special thanks to Timothy
Brownawell <tbrownaw@gmail.com>.
- new netsync option client "--set-default": makes it easy
to change default server/branches.
- "diff" now takes options "--context" and "--external", to
output context diffs and to invoke an external diff
program for full control over output formatting. new
option "--diff-args" pass arguments to external diff
program; new hook "external_diff" allows further
configuration. special thanks to Vladimir Vukicevic
<vladimirv@gmail.com>.
- b: and t: selectors now match exactly, instead of matching
as substrings. globbing is supported for inexact
matching. special thanks to Brian Downing
<bdowning@lavos.net>, Jordan Breeding
<jordan.breeding@mac.com>.
- new command 'db kill_tag_locally'. special thanks to Jordan
Breeding <jordan.breeding@mac.com>.
- now uses sqlite3 prepared statements. special thanks to
Derek Scherger <derek@echologic.com>.
- 'db migrate' is now a complete no-op if there is no
migration to do; automated scripts can now call it
optimistically and cheaply to guarantee up-to-dateness.
- new hash correctness tests. special thanks to Kaushik
Veeraraghavan <kaushikv@gmail.com>.
- upgrading from 0.20: you must run 'monotone db
migrate' once against each of your databases, to add
new sql indexes.
Tue Jul 5 23:57:10 PDT 2005
0.20 release. features, ui improvements, performance
improvements, and bug fixes.
- major changes in netsync UI: serve/sync/push/pull now take a
list of globs; clients can request arbitrary sets of
branches, not just predefined "collections". write
permissions are now granted on a per-db level (they were
before anyway).
- where you used to say, e.g., "monotone pull
net.venge.monotone", you should instead say
"monotone pull net.venge.monotone*". This may
require shell-quoting.
- 'get_netsync_write_permitted' hooks must be changed
to take only one argument, the 'identity'.
'get_netsync_{read,anonymous_read}_permitted' hooks
now take a branch argument instead of a collection,
and will be called for each branch that a client
requests.
- 0.19 clients cannot talk to 0.20 servers, and vice-versa.
- special thanks to Timothy Brownawell
<tbrownaw@gmail.com>, Richard Levitte
<richard@levitte.org>.
- other major changes:
- cvs_import re-written; many bugs fixed. now
supports tags.
- many minor netsync changes:
- netsync traffic is now cryptographically authenticated
against corruption and man-in-the-middle attacks.
special thanks to Ethan Blanton <elb@elitists.net>,
Matt Johnston <matt@ucc.asn.au>.
- new hooks that are called when server receives data:
note_netsync_*_received. special thanks to Timothy
Brownawell <tbrownaw@gmail.com>.
- ancestry graphs that pass outside the given branch
are now synchronized correctly. special thanks to
Timothy Brownawell <tbrownaw@gmail.com>.
- UI improvements:
- 'log' options changed: --depth has become --last;
new options --no-merges, --diffs, --brief.
- 'status' has new option --brief. special thanks to
Derek Scherger <derek@echologic.com>.
- 'serve' has new option --pid-file. special thanks
to Matthew Gregan <kinetik@orcon.net.nz>.
- all commands taking restrictions now take option
--depth, to limit recursion through subdirectories.
special thanks to Joel Reed <joelwreed@comcast.com>.
- merge command all take --author, --date now.
- 'checkout', 'update' take --revision, instead of
using positional arguments. special thanks to Derek
Scherger <derek@echologic.com>, Richard Levitte
<richard@levitte.org>.
- 'commit' takes new --message-file option.
- new features:
- new commands: "db kill_branch_locally", "db
kill_revision_locally", useful for correcting some
mistakes. special thanks to Brian Campbell
<brian.p.campbell@dartmouth.edu>, Sebastian Spaeth
<Sebastian@sspaeth.de>.
- new file attribute 'manual_merge', to prevent invocation of
merger on binary files. hook added to guess correct
value at 'add' time. special thanks to Riccardo
Ghetta <birrachiara@tin.it>.
- new 'earlier than', 'later than' selectors. special
thanks to Riccardo Ghetta <birrachiara@tin.it>.
- new automate commands:
- 'stdio', for efficient use by
front-ends. special thanks to Timothy Brownawell
<tbrownaw@gmail.com>.
- 'certs', for fetching certs on a revision in a
parseable (basic io-based) format. special thanks
to Grahame Bowland <grahame@angrygoats.net>.
- 'inventory' output changed incompatibly; should be
much more usable now, and stable. special thanks to
Derek Scherger <derek@echologic.com>.
- better memory/performance when handling large files.
special thanks to Eric Anderson
<anderse-monotone@cello.hpl.hp.com>, Timothy Brownawell
<tbrownaw@gmail.com>, Matt Johnston <matt@ucc.asn.au>,
Matthew Gregan <kinetik@orcon.net.nz>.
- new text mode browser in contrib/mtbrowse.sh, by Henry
Nestler <Henry@BigFoot.de>.
- improved zsh completion in contrib/monotone.zsh_completion,
by Joel Reed <joelwreed@comcast.com>.
- upgrading from 0.19: database and working copies are
fully compatible. netsync clients and servers need
to be upgraded together, as described above. the
many ui changes may require script updates.
will install Perl modules into the "vendor" directories:
chat/vicq math/udunits
databases/rrdtool mbone/beacon
devel/p5-subversion
Bump their PKGREVISIONs.
* Improved implementation of canSubstituteForClass: the default
implementation of this method tests the classes to see if
initWithCoder: or encodeWithCoder: is implemented on a subclass to
determine automatically if that class has the same encoding
signature as the original class, if it does, it can be substituted.
* Improved handling of classes which use cell classes in the custom
class inspector. The inspector now autmatically replaces the cell
class with the appropriate one when the user selects a given
subclass.
* Browser based class editor in document panel. This interface is
more like the one on OSX. The user now has a choice in
preferences to determine which view they would like to use.
* Translation tools. The Document->Translate menu allows the user
to export string and import strings in the strings format, so that
someone can easily translate just the strings in the file and
doesn't need to directly edit anything in Gorm. The strings file
can then be loaded back into Gorm and all of the relevant strings
are updated.
* Alignment tools. In the new Layout menu there are options to
align views, center views, bring views to front or push them to
the back of the view layers.
* Implementation of IBViewResourceDraggingDelegate. This allows
updating of the pull down in the inspectors panel dynamically. It
requires the developer of a palette to implement some code to
enable this, as on OSX.
* Lots of bugfixes and usability changes are also included in this
release.
Bugfixes in exiv2 extractor fixing remaining issues.
Changed plugins to not use filename but always only
rely on mmapped memory. Extended API with function
that allows running getKeywords on data in memory
(instead of filename). Extended API with encode
and decode functions for binary metadata.
Changes 0.5.1:
Preliminary integration of exiv2 support (not enabled
by default due to bugs). Moved Python and Java
bindings into separate packages.
print keywords.
2005-07-18 0.8.4
Fixed:
- extra CRs in CSV export files on Windows platform (sf bug 1195742)
- activity RDBMS columns were being reported in changes
- fix name collision in roundup.cgi script (sf bug 1203795)
- fix handling of invalid interval input
- search locale files relative ro roundup installation path (sf bug 1219689)
- use translation for boolean property rendering (sf bug 1225152)
- enabled disabling of REMOTE_USER for when it's not a valid username (sf
bug 1190187)
- fix invocation of hasPermission from templating code (sf bug 1224172)
- have 'roundup-admin security' display property restrictions (sf bug
1222135)
- fixed templating menu() sort_on handling (sf bug 1221936)
- allow specification of pagesize, sorting and filtering in "classhelp"
popups (sf bug 1211800)
- handle dropped properies in rdbms/metakit journal export (sf bug 1203569)
- handle missing Subject lines better (sf bug 1198729)
- sort/group by missing values correctly (sf bugs 1198623, 1176897)
- discard, don't bounce messages to the mailgw when the messages's sender
is invalid (ie. when we try to bounce, we get a 550 "unknown user
account" response from the SMTP server) (sf bug 1190906)
- removed debugging code from cgi/actions.py
- refactored hyperdb.rawToHyperdb, allowing a number of improvements
(thanks Ralf Schlatterbeck)
- don't try to set a timeout for IMAPS (thanks Paul Jimenez)
- present Reject exception messages to web users (sf bug 1237685)
* New features and minor improvements
- A bug in DBX mode introduced in release 3.3.10 has been fixed.
- A bug which prevented DDD from properly displaying ISO-8859-1 characters
(and probably other character sets too) under Motif has been fixed (1)
Changes from 1.43:
Version 1.48
- Fixed incompatibility with perl version < 5.8x of the capture function;
the fix excludes the usage of other tied class :-(
Version 1.47
- Maintainance release. A few minor changes
Version 1.46
- Fixed bug in nested capture, introduced with v1.45
- Simplified the capture implementation
- Added a few tests
Version 1.45
- Full reimplemention of the capture() function, greatly improved
- Removed the Tie::Handle dependency
- Improvement of the slurp() function
- Added the 'markers' option in the load_mml function and tests.
Version 1.44
- Fixed missing quotes in the slurp and _path_mtime function
- Added the 'optional' option in the load_mml function
New features
Added support for parsing Python code. Special comment blocks in Python
are of the ## # Some comment #
or you can rely on Python docstrings. This code uses the new parser
interface and was based on the work done by a group of students as part
of a compiler design project. Added support for C# style XML tags in the
comments. See the manual or the C# language specification for more
details. Thanks to Talin for doing most of the work. Added support for
the JavaDoc command {@inheritDoc}.
And some bug fixes. For details see:
http://www.stack.nl/~dimitri/doxygen/changelog.html
Version 6.2 01-Aug-05
---------------------
1. There was no test for integer overflow of quantifier values. A construction
such as {1111111111111111} would give undefined results. What is worse, if
a minimum quantifier for a parenthesized subpattern overflowed and became
negative, the calculation of the memory size went wrong. This could have
led to memory overwriting.
2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
operating environments where this matters.
4. Applied Giuseppe Maxia's patch to add additional features for controlling
PCRE options from within the C++ wrapper.
5. Named capturing subpatterns were not being correctly counted when a pattern
was compiled. This caused two problems: (a) If there were more than 100
such subpatterns, the calculation of the memory needed for the whole
compiled pattern went wrong, leading to an overflow error. (b) Numerical
back references of the form \12, where the number was greater than 9, were
not recognized as back references, even though there were sufficient
previous subpatterns.
6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
versions of gcc, e.g. 2.95.4.
Version 6.1 21-Jun-05
---------------------
1. There was one reference to the variable "posix" in pcretest.c that was not
surrounded by "#if !defined NOPOSIX".
2. Make it possible to compile pcretest without DFA support, UTF8 support, or
the cross-check on the old pcre_info() function, for the benefit of the
cut-down version of PCRE that is currently imported into Exim.
3. A (silly) pattern starting with (?i)(?-i) caused an internal space
allocation error. I've done the easy fix, which wastes 2 bytes for sensible
patterns that start (?i) but I don't think that matters. The use of (?i) is
just an example; this all applies to the other options as well.
4. Since libtool seems to echo the compile commands it is issuing, the output
from "make" can be reduced a bit by putting "@" in front of each libtool
compile command.
5. Patch from the folks at Google for configure.in to be a bit more thorough
in checking for a suitable C++ installation before trying to compile the
C++ stuff. This should fix a reported problem when a compiler was present,
but no suitable headers.
6. The man pages all had just "PCRE" as their title. I have changed them to
be the relevant file name. I have also arranged that these names are
retained in the file doc/pcre.txt, which is a concatenation in text format
of all the man pages except the little individual ones for each function.
7. The NON-UNIX-USE file had not been updated for the different set of source
files that come with release 6. I also added a few comments about the C++
wrapper.
Version 6.0 07-Jun-05
---------------------
1. Some minor internal re-organization to help with my DFA experiments.
2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
didn't matter for the library itself when fully configured, but did matter
when compiling without UCP support, or within Exim, where the ucp files are
not imported.
3. Refactoring of the library code to split up the various functions into
different source modules. The addition of the new DFA matching code (see
below) to a single monolithic source would have made it really too
unwieldy, quite apart from causing all the code to be include in a
statically linked application, when only some functions are used. This is
relevant even without the DFA addition now that patterns can be compiled in
one application and matched in another.
The downside of splitting up is that there have to be some external
functions and data tables that are used internally in different modules of
the library but which are not part of the API. These have all had their
names changed to start with "_pcre_" so that they are unlikely to clash
with other external names.
4. Added an alternate matching function, pcre_dfa_exec(), which matches using
a different (DFA) algorithm. Although it is slower than the original
function, it does have some advantages for certain types of matching
problem.
5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
including restarting after a partial match.
6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
defined when compiling for Windows was sent to me. I have put it into the
code, though I have no means of testing or verifying it.
7. Added the pcre_refcount() auxiliary function.
8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
match before or at the first newline in the subject string. In pcretest,
the /f option on a pattern can be used to set this.
9. A repeated \w when used in UTF-8 mode with characters greater than 256
would behave wrongly. This has been present in PCRE since release 4.0.
10. A number of changes to the pcregrep command:
(a) Refactored how -x works; insert ^(...)$ instead of setting
PCRE_ANCHORED and checking the length, in preparation for adding
something similar for -w.
(b) Added the -w (match as a word) option.
(c) Refactored the way lines are read and buffered so as to have more
than one at a time available.
(d) Implemented a pcregrep test script.
(e) Added the -M (multiline match) option. This allows patterns to match
over several lines of the subject. The buffering ensures that at least
8K, or the rest of the document (whichever is the shorter) is available
for matching (and similarly the previous 8K for lookbehind assertions).
(f) Changed the --help output so that it now says
-w, --word-regex(p)
instead of two lines, one with "regex" and the other with "regexp"
because that confused at least one person since the short forms are the
same. (This required a bit of code, as the output is generated
automatically from a table. It wasn't just a text change.)
(g) -- can be used to terminate pcregrep options if the next thing isn't an
option but starts with a hyphen. Could be a pattern or a path name
starting with a hyphen, for instance.
(h) "-" can be given as a file name to represent stdin.
(i) When file names are being printed, "(standard input)" is used for
the standard input, for compatibility with GNU grep. Previously
"<stdin>" was used.
(j) The option --label=xxx can be used to supply a name to be used for
stdin when file names are being printed. There is no short form.
(k) Re-factored the options decoding logic because we are going to add
two more options that take data. Such options can now be given in four
different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
(l) Added the -A, -B, and -C options for requesting that lines of context
around matches be printed.
(m) Added the -L option to print the names of files that do not contain
any matching lines, that is, the complement of -l.
(n) The return code is 2 if any file cannot be opened, but pcregrep does
continue to scan other files.
(o) The -s option was incorrectly implemented. For compatibility with other
greps, it now suppresses the error message for a non-existent or non-
accessible file (but not the return code). There is a new option called
-q that suppresses the output of matching lines, which was what -s was
previously doing.
(p) Added --include and --exclude options to specify files for inclusion
and exclusion when recursing.
11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
Hopefully, it now does.
12. Missing cast in pcre_study().
13. Added an "uninstall" target to the makefile.
14. Replaced "extern" in the function prototypes in Makefile.in with
"PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
world, but is set differently for Windows.
15. Added a second compiling function called pcre_compile2(). The only
difference is that it has an extra argument, which is a pointer to an
integer error code. When there is a compile-time failure, this is set
non-zero, in addition to the error test pointer being set to point to an
error message. The new argument may be NULL if no error number is required
(but then you may as well call pcre_compile(), which is now just a
wrapper). This facility is provided because some applications need a
numeric error indication, but it has also enabled me to tidy up the way
compile-time errors are handled in the POSIX wrapper.
16. Added VPATH=.libs to the makefile; this should help when building with one
prefix path and installing with another. (Or so I'm told by someone who
knows more about this stuff than I do.)
17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
passes PCRE_DOTALL to the pcre_compile() function, making the "." character
match everything, including newlines. This is not POSIX-compatible, but
somebody wanted the feature. From pcretest it can be activated by using
both the P and the s flags.
18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
19. libpcre.pc was being incorrectly installed as executable.
20. A couple of places in pcretest check for end-of-line by looking for '\n';
it now also looks for '\r' so that it will work unmodified on Windows.
21. Added Google's contributed C++ wrapper to the distribution.
22. Added some untidy missing memory free() calls in pcretest, to keep
Electric Fence happy when testing.
Overview of Changes from GLib 2.6.5 to GLib 2.6.6
=================================================
* Fix several bugs with the handling
of comments in key files [Mikael Magnusson]
* Fix g_atomic_pointer_compare_and_exchange
on Sparc64 [Gert Doering]
* Fix a deadlock with threadpools [Hong Jen Yee]
* Make GLib work on BeOS again [Kian Duffy]
* Win32 bug fixes [Kazuki Iwamoto]
* Other bug fixes [Roger Leigh, Kjartan Maraas,
Morten Welinder]
* New and updated translations (bg,cs,cy,de,
en_CA,es,et,fr,gu,hu,ja,ko,nl,sq,sr,sr@Latn,sv,
te,zh_CN,zh_TW)
This includes security fix as zlib 1.2.2.
CHANGES
-------
1.35 - 16 July 2005
* Updated zlib source to 1.2.3
* Fixed problem with where two calls to gzclose would hang the debugger.
See https://rt.cpan.org/Ticket/Display.html?id=13789
* Added code from Alexey Tourbin to use XSLoader when available,
and DynaLoader otherwise.
* Documented that the compress & uncompress functions were not
the same as the Unix utilities of the same name.
* Fixed 05gzsetp -- it left a temp file hanging around.
* Integrate core change 24787 - SvUPGRADE returns void in blead
* Integrate core change 24788 - Makefile.PL adjustments for the core
-restore dependencies on libtool and pkg-config to be "full",
(this is an IDE, using all that gnu stuff)
-fix the baudrate patch to give it a chance to compile on NetBSD<current
(B460800 went in just recently)
-don't try to remove $PREFIX/share/mimelnk
After a long time I made another release to bring some patches from the community to all users:
New: Better symbol handling of structs/classes/etc.
New: New realtime func/macros/var/struct define exporer in message windows.
Bug fix: IA64 and Solaris problems
Bug fix: Terminal Ctrl-C and Shift-Insert
Bug fix: Debugger improvements
Bug fix: Project generation
Bug fix: double tags loading
Bug fix: Patches from the debian tree (Thanks!)
Hopefully this will make anjuta more usable for now. Of course there are
many many other bugs but development is concentrating on anjuta2.
Changes:
* umbrello: Crash when deleting an attribute that is represented as an
association
* umbrello: Inline functions are generated in cpp file
* umbrello: Crash when closing a tab
* umbrello: Initial value of a new paramenter of a method cannot be set
the first time
* umbrello: Line Vertex hidden in class diagrams
* umbrello: Relationships between interfaces and classes change over file
reload
* umbrello: Associations turn into generalisation over file reopen
* umbrello: Code generator "could not find active language"
* umbrello: Cannot drag members from one classifier to another
* umbrello: Generating Javascript code from a class diagram hangs
* Window's path handling has changed to use native paths throughout.
* Startup/configuration of the libraries no longer requires
environement variables.
* Character sets are included in the library rather than being read
from data files.
* NSRunLoop and related classes use natvie win32 event handling on
Windows machines.
* New rules for building and using DLLs were implemented. Now gcc
and other tricks are used to properly export symbols, so no .def
file should be needed.
* The GNUsteprc file in $GNUSTEP_SYSTEM_ROOT is deprecated. See the
new GNUstep.conf file (which is not installed by default).
* Psuedo-framwork support was implemented on MingW. This essentially
involves copying the framework to several places so it can be used
without links, which Windows does not really support. There's no
other clean way to do it.
* Support for xxx_WINDRES_FILES on MinGW was implemented.
* Add Khmer support [Jens Herden, Javier Sola]
* Fix bugs and crashes in the operation of PangoLayoutIter [Amit Aronovitch]
* Increase sharing of data between different copies of Pango
[Tommi Komulainen, Ross Burton]
* Remove dependence of opentype code on FreeType internals [David Turner,
Behdad Esfahbod]
* Cache metrics for the Win32 backend [Tor Lillqvist]
* Improve handling of RTL text when passing it to Uniscribe [Tor]
* Support for TrueType fonts on Win32 with characters outside the BMP [Tor]
* Handle alpha in XftColors again. [Mikael Magnusson]
* Better handling of MacThai encoded fonts [Theppitak Karoonboonyanan]
* Handle zero width chars properly in Hangul backend
[Young-Ho Cha, Changwoo Ryu]
* Fix Oriya character classification [Gora Mohanty, Gautam Sengupta]
* PangoAttrIterator bug fixes [Morten Welinder]
* Clean up copyright/licensing information [William N. Ray]
* Misc bug and build fixes [Sebastien Bacher, Brandon Bergren,
Chris Blizzard, Paolo Borelli, Behdad, Harshula, Alex Jones, Aivars Kalvans,
Stepan Kasal, Karel Kulhavy, Ryan Lortie, Ben Maurer, Sukhjinder Sidhu,
Manish Singh, Emil Soleyman-Zomalan, Morten]
07232005 - Fixed bug in zip archiver (thanks, Jörg Walter!). Updated zlib to
1.2.3, which properly includes the security fix. Fixed "make dist"
to handle .svn dirs and other file changes. Removed "debian"
directory. Fixed warning in physfs.c. Updated CWProjects.sit.
Upped version to 1.0.1.
07132005 - Moved to zlib122, and security fix discussed here:
http://www.cve.mitre.org/cgi-bin/cvename.cgi?name=CAN-2005-2096
06122005 - Added support for mingw to Unix build process (thanks, Matze!).
03162005 - Readded Spanish translation. Added Brazillian Portguese translation
and fixed unlocalized string in wad.c (thanks Danny!). Some MacOS
Classic fixes and MPW support (thanks, Chris!). Changed CWProjects
from SITX to SIT format, so MacOS Classic users can unpack it.
02152005 - Minor comment fix in platform/pocketpc.c
01052004 - Fixed HOG archiver sorting/file lookup (thanks, Chris!)
12162004 - Fixed some documentation/header comment typos (thanks, Gaetan!)
10302004 - Fixed a strcpy that should have been a strcat. (thanks, Tolga!)
Removed .cvsignore files (we're in a Subversion repository, now).
Build system respects external CFLAGS now. (thanks, Adam!)
09212003 - Happy September. Fixes to bootstrap on MacOS X. Seeking now tries
to respect pre-buffered data.
"annotate" has been implemented and "log" has been enhanced.
The patch queue has been updated and tests modified to work with ArX.
"fork", "replay", and "merge" no longer supports out-of-place forks.
"arx help" now works.
The autoconf scripts for Python configuration have been revamped to
use a standard autoconf macro. See INSTALL.GENERIC for details.
Revision libraries and "library" have been removed.
Some limitations in "merge", "diff" and "replay" have been removed.
** A RPM *.spec file is included.
** Fix compiler warnings, contributed by Frediano Ziglio.
** Gnulib is used, currently only to provide more robust ntlm_check_version.
** Address in license was updated to reflect the new FSF postal address.
Set TEST_TARGET.
** The macro AX_CREATE_STDINT_H that is used to create idn-int.h
** has been updated.
** Fix use of 'head -1' in configure script (should be 'head -n -1'),
** thanks to Carsten Lohrke.
** Announce the help-libidn mailing list in documentation and README.
** Translation updates.
** API and ABI is backwards compatible with the previous version.
* Increased version for GNOME 2.10 release version.
Changes 2.8.0:
* Increased version for GNOME 2.8 release version.
Changes 2.7.2:
* Now builds against latest gnome-vfs 2.7 API.
Changes 2.7.1:
This is the first version for GNOME 2.7/2.8 with additional API:
* Added Address.
* Added ResolveHandle.
* Added DnsSd functions.
* Drive: Added get_hal_udi().
* Volume: Added get_hal_udi().
libglademm 2.6 is API/ABI-compatible with libglademm 2.4.
Changes since libglademm 2.4:
* Glade::Xml: Added connect_clicked() for easy connection of
buttons and menuitems signals.
* Glade::VariablesMap: now handles Entry, ComboTextEntry,
SpinButton, *Scale, Calendar and CheckBox.
popular GUI library GTK+. Highlights include typesafe callbacks, widgets
extensible via inheritance and a comprehensive set of widget classes that
can be freely combined to quickly create complex user interfaces.
this fixes (at least) another security problem (DoS, CAN-2005-1849)
changes:
-Eliminate a potential security vulnerability when decoding invalid
compressed data
-Eliminate a potential security vulnerability when decoding specially
crafted compressed data
-Fix a bug when decompressing dynamic blocks with no distance codes
-Fix crc check bug in gzread() after gzungetc()
-Do not return an error when using gzread() on an empty file
* g++ 3.2 (and Mac OS X g++ 3.3) build fix.
* Compose: Fix slot lifetime regression introduced in 2.0.9.
* tests: Small ISO C++ correctness fix
* Don't specify unused function parameter names.
Valgrind is a suite of tools for debugging and profiling x86-Linux programs.
With the tools that come with Valgrind, you can automatically detect many
memory management and threading bugs, avoiding hours of frustrating
bug-hunting, making your programs more stable. You can also perform detailed
profiling, to speed up and reduce memory use of your programs.
The Valgrind distribution currently includes five tools: two memory error
detectors, a thread error detector, a cache profiler and a heap profiler.
News:
2005-07-16 Tollef Fog Heen
* configure.in: Release 0.19
2005-07-15 Tollef Fog Heen
* pkg.c (package_get_var): Make sure to g_strdup all the return
values and not return some values which should not be freed and
some which should. Yay valgrind. Freedesktop #3682
* configure.in: Fix default search path to be pkgconfig rather
than pkg-config again. Freedesktop #3662
* pkg.m4: Add a missing AC_MSG_RESULT. Thanks to Gary Kramlich
for noticing this and harassing me to fix it.
2005-06-29 Tollef Fog Heen
* configure.in: Release 0.18.1
* pkg.m4: Brown bag fix. pkg_failed was always set to “untried”.
Debian #316181.
2005-06-27 Tollef Fog Heen
* configure.in: Rename to pkg-config.
pkg-config 0.18
The inter-library dependencies check was too tight and caused
problems if one used the --no-undefined flag to libtool on Solaris
(since it there expands to -Wl,-z,defs which disallows undefined
symbols). Add a new name to .pc files: Libs.private which will
not be listed in the output of --libs unless --static is also
given.
Private libraries are libraries which are needed in the case of
static linking or on platforms not supporting inter-library
dependencies. They are not supposed to be used for libraries
which are exposed through the library in question. An example of
an exposed library is GTK+ exposing Glib. A common example of a
private library is libm.
Generally, if include another library's headers in your own, it's
a public dependency and not a private one.
Thanks a lot to James Henstridge for both the bug and the following
discussion.
Revision 0.18 2005/05/23 15:34:31 ray
moved declaration to top of function, M$ (and other) C compilers choke.
Revision 0.17 2005/05/05 22:26:01 ray
Changed PERL_MAGIC_backref to '<' for compatability with 5.6
Revision 0.16 2005/04/20 15:49:35 ray
Bug fix for id 11997, "Clone dies horribly when Scalar::Util::weaken is around"
see http://rt.cpan.org/Ticket/Display.html?id=11997 for details.
Revision 0.15.2.1 2005/05/05 21:55:30 ray
changed PERL_MAGIC_backref to '<' for backward compatibility with 5.6
around at either build-time or at run-time is:
USE_TOOLS+= perl # build-time
USE_TOOLS+= perl:run # run-time
Also remove some places where perl5/buildlink3.mk was being included
by a package Makefile, but all that the package wanted was the Perl
executable.
run-time dependency (DEPENDS) on a tool is to append a ":run" modifier
to the tool name, e.g.,
USE_TOOLS+= perl:run
Tools without modifiers or with an explicit ":build" modifier will
cause build dependencies (BUILD_DEPENDS) on those tools to be added.
This makes the notation a bit more compact.
- Fixed (again) glob patterns to allow both "!" (like modern glob) or
"^" (like existing shells) for negating character sets.
- Tweaked str_alloc to use realloc when appropriate.
- Fixed glob pattern matches to use "!" instead of "~" for negating
character sets like all other globs use.
- Fixed the installer libraries to include all the necessary object
files for producing the installer/instcheck/instshow programs.
- Added a generic installer program, similar to the "install" program
included with ezmlm.
- Fixed the instshow library to use and dispaly the install prefix just
like the other instlib programs.
- The installers now automatically create missing directories, and
installs into temporary files that get renamed.
- Added path_mkdirs function.
- Fixed typo bug in str_catfv and str_copyfv that caused failures on (at
least) PPC and S390 architectures. Thanks Gerrit Pape.
- Fixed 32-to-64 bit sign extension bug in fmt_multi on some 64-bit
platforms. Thanks Gerrit Pape.
- installer/instcheck program libraries now optionally install into a
prefix directory specified on the command line or with
$install_prefix (which can be set on the make command line).
- Installation is now done under seperate include/lib directories.
- Added sysdep for "extern int daylight" (replaced by a macro if not
present).
- Fixed a couple of length counting bugs in the fmt number code.
- Added multiple-item formatting functions fmt_multi (similar to printf,
but safer), fmt_multiv, obuf_putf, str_catf, and str_copyf, as well as
a set of formatted message functions.
> New in 1.12 (March 17, 2005)
>
> Build bug fixed: panel, forms, menu libs not found in build.
>
> Build bug fixed: Curses.pm treats version as number, so 1.10 == 1.1 .
>
> New in 1.11 (March 12, 2005)
>
> Various cleanup and improved diagnostics in the build.
>
> New in 1.10 (March 11, 2005)
>
> Build bug fixed: Makefile has undefined variable $libtyp.
>
> Build bug fixed: Makefile computes wrong guess hint file name.
>
> New in 1.09 (March 6, 2005)
>
> - Makefile.PL searches for curses header files instead of assuming
> based on $OSNAME.
>
> - Makefile.PL warns user to reconfigure environment when form.h is
> in /usr/include rather than just add -I/usr/include to the compile
> (this concerns the problem with the conflicting Perl form.h file).
> 2005-05-14 <joshua@always.joy.eth.net>
>
> * Release 1.06.
>
> 2005-05-09 <joshua@always.joy.eth.net>
>
> * Fix 64 bit issue. Encouraged by Eric Rybski
> <rybskej@yahoo.com>.
>
> 2005-04-23 <joshua@always.joy.eth.net>
>
> * Release 1.05.
>
> * Fix ancient memory leak in loop() noticed by Nikita Savin
> <nik@asdfgroup.com>.
>
> 2005-03-29 <joshua@always.joy.eth.net>
>
> * Release 1.04.
>
> * Preliminary /dev/poll support from Michael Pergament
> <ESeifert@t-online.de>. As of yet, there is no config test to
> turn on this code. Patches welcome.
>
> 2005-01-31 <joshua@always.joy.eth.net>
>
> * Release 1.03.
>
> * Apply Win32 patch from Graciliano M. P. <gmpowers@terra.com.br>.
>
> 2004-12-08 <joshua@always.joy.eth.net>
>
> * Release 1.02.
>
> 2004-12-08 Zefram <zefram@fysh.org>
>
> * Generic watchers: implementation, documentation, and tests.
>
> * Documentation refinements.
>
> 2004-11-29 <joshua@always.joy.eth.net>
>
> * Release 1.01.
>
> * Event.xs: Due to popular demand, queue_pending() is now a public
> API, for better or worse. Beware of race conditions.
>
> 2004-11-29 Zefram <zefram@fysh.org>
>
> * A var watcher can currently be created with no variable to
> watch, provided of course that it is parked. However, once a
> variable has been set, the watcher cannot be returned to the
> no-variable state. The variable to watch can be changed, but the
> watcher rejects any attempt to set the reference to undef.
>
> Applied a patch to allow the variable reference of a var watcher
> to be set to undef.
>
> * Applied patch to clarify the documentation (with tweaks from
> Joshua).
>
> 2004-05-14 <joshua@always.joy.eth.net>
>
> * Release 1.00 (with Marc Lehmann's encouragement).
>
> * c/watcher.c (pe_watcher_now): Check for a missing callback, not
> only a missing perl callback. Marc A. Lehmann <pcg@goof.com>
>
> 2004-04-26 <joshua@always.joy.eth.net>
>
> * lib/Event.pm (import): Make NO_TIME_HIRES actually work. Jerry
> D. Hedden <jerry@hedden.us>
>
> 2004-04-22 <joshua@always.joy.eth.net>
>
> * c/signal.c (pe_signal_stop): When a signal watcher stops, the
> signal counter is cleared. Otherwise the signal watcher can
> generate one last event after being stopped, which is a
> bug. (Pointed out by Zefram <zefram@fysh.org>)
>
> 2004-04-21 <joshua@always.joy.eth.net>
>
> * Release 0.88.
>
> * lib/Event.pm: Simplify hooking into Time::HiRes. Add a
> NO_HIRES_TIME export_ok symbol to turn off the probing.
>
> * Event.xs: Remove U2time. Remove install_time_api().
>
> * lib/Event.pod: More refinement from Zefram <zefram@fysh.org>.
>
> * lib/Event.pod: ($watcher->pending & signal watchers): More
> specific description by Zefram <zefram@fysh.org>.
>
> * c/watcher.c (pe_watcher_now): Fix SEGV reported by Zefram
> <zefram@fysh.org>. Add test.
>
> 2004-04-05 <joshua@always.joy.eth.net>
>
> * lib/Event.pod (timer): Correction by Randal L. Schwartz
> <merlyn@stonehenge.com>.
>
> 2004-04-03 <joshua@always.joy.eth.net>
>
> * lib/Event.pod (timer): Doc clarification
> (jdhedden@1979.usna.com).
>
> Tue Feb 18 21:57:04 2003 Joshua N Pritikin <vishnu@pobox.com>
>
> * Release 0.87.
>
> * c/var.c (tracevar_r, tracevar_w): Fix declaration for recent
> versions of perl (patch from Nick Eggleston <nick@dccinc.com>).
>
> Sat May 25 11:23:50 2002 Joshua N Pritikin <vishnu@pobox.com>
>
> * Release 0.86.
>
> * c/unix.c: Solaris works better without POLLWRBAND (Clemens
> Schrimpe <csch@Kiez.NET>).
>
> Fri Feb 1 12:20:33 2002 Joshua N Pritikin <vishnu@pobox.com>
>
> * Release 0.85.
>
> * Apply patch from Allen Smith <easmith@beatrice.rutgers.edu>
> to avoid infinite polling loop on some select implementations.
> Also reported by Marc Lehmann <pcg@goof.com>.
>
> Tue Jan 22 12:20:24 2002 Joshua N Pritikin <vishnu@pobox.com>
>
> * Doc StarvePrio (Allen Smith <easmith@beatrice.rutgers.edu>).
>
> * Remove IRIX special casing. Reports indicate that new versions
> of IRIX don't need any hacks. (Allen Smith
> <easmith@beatrice.rutgers.edu>)
>
- Update addresses two security issues
- From the ChangeLog:
> Version 2.18.2
> --------------
>
> + You can now create accounts with createaccount.cgi even
> when the "requirelogin" parameter is turned on. (Bug 294778)
>
> + Bugs that are in disabled groups may not show a padlock
> on the bug list, or may otherwise behave strangely. You
> can now fix this using sanitycheck.cgi. (Bug 277454)
>
> + If sendmail dies while you are marking a bug
> as a duplicate, the duplicates table will no longer become
> corrupted. (Bug 225042)
>
> + Any user can change a flag on any bug. This also allows the
> attacker to expose the summary of any bug, even a hidden bug.
>
> + Summaries of private bugs are sometimes exposed under a very rare
> condition if you use MySQL replication.
>
> Version 2.18.3
> --------------
>
> + The query.cgi page was broken in 2.18.2 by bug 300138.
> That is now fixed.
Bugs fixed in 1.9.6:
* Longstanding bugs:
- Correctly diagnose `#' comments following trailing backslash.
- Preserve backslashes preceding `##' lines.
- Preserve the order of items conditionally appended (+=) to variables.
- Fix support of installation directory names with spaces on systems
where mkinstalldirs cannot use `mkdir -p'.
- Avoid infinite loop in mdate-sh when TIME_STYLE is set.
- Do not output two definitions of SOURCES (an internal variable).
One was not formatted and could exceed the maximum line length of
some third-party tools (e.g., AIX 5.1 grep, breaking dependency
tracking).
- Do not empty info files when attempting to rebuild them without makeinfo.
- Be smarter when a Makefile.am references files in both "./dir" and
"dir": do not output two rules to create these directories.
(PR/461)
- Do not attempt to recover a missing *.elc file if it cannot be
created because emacs does not exist.
- Several aclocal fixes for issues occuring when configure.ac
includes some other m4 files explicitely with m4_include or m4_sinclude.
(PR/450)
- depcomp's cpp mode now understands preprocessors that output either
`#line 42 file' or `# 42 file'. (Only the latter was supported.)
* Other miscellaneous changes:
- Update the GPL, and the FSF postal address.
- Anticipate for python2.5 in AM_PATH_PYTHON.
- The manual should now compile without any warning from TeX.
2005-05-02 0.8.3
Feature:
- chinese translation by limodou
Fixed:
- fix reference to The Zope Book in Roundup FAQ
- disabled file logging in Roundup test suite (sf bug 1155649)
- return original string if message issue xref isn't valid
- fix nosyreaction.py to stop it setting the nosy list unnecessarily
(see doc/upgrading.txt for how to fix in your trackers)
- after logout, always display tracker home page
- web forms don't create new items if no item properties are set from UI
- item creation failed if multilink fields had invalid entries (sf bug
1177602)
- fix bdist_rpm (sf bug 1164328)
- fix checking of "Email Access" for Anonymous email registration (sf bug
1177057)
- disable "Email Access" for Anonymous by default to stop spam regsitering
users on public trackers
- send errors in the web interface to a logfile by default. Use the
"debug" multiprocess mode (roundup-server) or the DEBUG_TO_CLIENT var
(roundup.cgi) to have the errors appear in your browser
- fix setgid typo (sf bug 1171346)
- fix faulty find_template filename facility (sf bug 1163629)
- fix roundup-admin "export" so it creates the target dir if needed
- "fix" roundup-admin "import" to not use "universal newline support" since
the csv module appears to have its own ideas about such things (sf bug
1163890)
- fix installation docs referring to old-style configuration variables
- fix roundup-admin "find" for searching Multilinks (sf bug 1189465)
2005-03-03 0.8.2
Feature:
- roundup-server automatically redirects from trackers list
to the tracker page if there is only one tracker
Fixed:
- added content to ZRoundup refresh.txt file (sf bug 1147622)
- fix invalid reference to csv.colon_separated
- correct URL to What's New in setup.py meta-data
- change AUTOCOMMIT=OFF to AUTOCOMMIT=0 for MySQL (sf bug 1143707)
- compile message objects in 'setup.py build'
- use backend datatype for journal timestamps in RDBMSes
- fixes to the "Using an external password validation source"
customisation example (sf bugs 1153640 and 1155108)
- Added tests for UTF8 handling internally and in files via ContentObject.
- Added CGI tests.
- All CGI parameters are now decoded as UTF8. The -utf8 flag is now
deprecated, but will continue to work as a no-op for now.
- socialtext changes to allow formatter hooking
- Changes for themes
- -compress works better
- Fix the mixin problem with resolve_install_path
- Refactors to get rid of hub connections
- Kwid Formatter for Spork
- make sure the module_path comes from lib under cwd
- Make installer better
- Miyagawa's patch to bake cookies in redirect
- CGI patch from cdent
- Miyagawa added jar_opt support for different expires for each cookies
- Miyagawa added unit test for Spoon::Cookie
- fix deep-recursion bug on 'kwiki -update'
- Add a blank return to super
- Add the mixin method
- Refactored -Base to once and only once
- Remove some cruft
- Got export algorithm working better
- lots of fixes to get the thing working sufficienly for cvs2git
- main features
- optionally detect and display ancestor branch information for
new detected branches
- never allow more than one revision of a file to be in a single
patchset (this may disturb the patchset numbers for historical
data :-(
- use file revisions to help sort, so that initial imports
won't be listed as patchset 1, with patchset 2 as the actual
creation of the files.
- reports log times in localtime (use TZ to override)
- many downstream patches from Debian, special thanks to Marcus
Crafter for maintaining. From the Debian changelog:
- patch from Kim Hansen to fix time calculation bug
- documentation patch from J. Bruce Fields
- directory name handling patch from David Kilzer
- others I'm not sure of, I'd happily update the credit here
