Colin Walters [Tue, 5 Jul 2011 13:42:39 +0000 (09:42 -0400)]
gio: Fix srcdir != builddir build for docs
Matthias Clasen [Tue, 5 Jul 2011 13:17:02 +0000 (09:17 -0400)]
Update NEWS for 2.29.10
Matthias Clasen [Mon, 4 Jul 2011 20:03:33 +0000 (16:03 -0400)]
Fix the build
When renaming a variable from _SCRIPTS to _DATA, it helps to
also rename all references.
Matthias Clasen [Mon, 4 Jul 2011 17:09:54 +0000 (13:09 -0400)]
Disable a failing test
Chun-wei Fan [Mon, 4 Jul 2011 14:21:05 +0000 (22:21 +0800)]
Update GIO VS 2010 project templates
Revert EOL of these files to Unix style as they are to be processed
during "make dist"
Chun-wei Fan [Mon, 4 Jul 2011 13:47:59 +0000 (21:47 +0800)]
Updated GIO Visual Studio Projects
-In gio/Makefile.am, the name for one of the filters for capturing the
sources for the GIO VS Project Files is corrected.
-Remove the GIO source file items in the VS project files templates as
a result for this change, and move the entry of the "new"
gregistrysettingsbackend.c into the filter in gio/Makefile.am
Murray Cumming [Mon, 4 Jul 2011 06:58:00 +0000 (08:58 +0200)]
G_STATIC_REC_MUTEX_INIT: Fix missing initialization.
Add the third struct field, zeroed, to avoid a compiler warning.
Ryan Lortie [Sun, 3 Jul 2011 15:37:35 +0000 (17:37 +0200)]
Install completion scripts as DATA, not SCRIPTS
They are sourced by the shell and don't include a #!/bin/sh, so they shouldn't
be marked as executable.
Dan Winship [Wed, 29 Jun 2011 12:29:16 +0000 (08:29 -0400)]
gasyncresult: document that NULL GAsyncReadyCallback is allowed
https://bugzilla.gnome.org/show_bug.cgi?id=653484
Dieter Verfaillie [Thu, 30 Jun 2011 13:32:32 +0000 (09:32 -0400)]
GCancellable: Fix build on Win32
https://bugzilla.gnome.org/show_bug.cgi?id=653522
Emmanuele Bassi [Wed, 29 Jun 2011 17:07:29 +0000 (18:07 +0100)]
gio: Some more build fixes
Emmanuele Bassi [Wed, 29 Jun 2011 17:00:53 +0000 (18:00 +0100)]
Fix build
The g_return_val_if_fail() macro takes a return value.
Ryan Lortie [Wed, 29 Jun 2011 16:19:38 +0000 (17:19 +0100)]
parameter guards for GSimpleAction constructors
Ryan Lortie [Wed, 29 Jun 2011 16:13:58 +0000 (17:13 +0100)]
GAction: make GObject properties read-only
GAction is a read-only interface (as is visible by the lack of _set() functions
on its API). The properties on the interface currently force implementors to
support writing of the properties at construct time, however.
Lift that restriction.
Take advantage of this from GSimpleAction by nuking the set_property
function and setting the fields directly in the constructor.
Ryan Lortie [Wed, 29 Jun 2011 15:10:38 +0000 (16:10 +0100)]
actions: merge testcases change_state and entries
Since we have support for change_state in GActionEntry now.
Ryan Lortie [Wed, 29 Jun 2011 15:04:09 +0000 (16:04 +0100)]
Add 'change_state' callback to GActionEntry
Ryan Lortie [Wed, 29 Jun 2011 13:53:24 +0000 (14:53 +0100)]
Add example for 'change-state' signal
Pulled from the testcase.
Ryan Lortie [Wed, 29 Jun 2011 13:44:58 +0000 (14:44 +0100)]
Add testcase for GSimpleAction::change-state
Ryan Lortie [Wed, 29 Jun 2011 13:35:12 +0000 (14:35 +0100)]
Add 'change-state' signal to GSimpleAction
If connected, allows the user to control the result of
g_action_change_state().
Ryan Lortie [Wed, 29 Jun 2011 12:55:59 +0000 (13:55 +0100)]
Add example to docs for using GActionEntry
Ryan Lortie [Wed, 29 Jun 2011 12:02:30 +0000 (13:02 +0100)]
Add a test case for GActionEntry
Ryan Lortie [Wed, 29 Jun 2011 12:00:30 +0000 (13:00 +0100)]
Add g_simple_action_group_add_entries()
A convenience API for creating lots of actions quickly.
Ryan Lortie [Wed, 29 Jun 2011 11:28:15 +0000 (12:28 +0100)]
allow NULL state in g_simple_action_new_stateful
as an undocumented feature
Ryan Lortie [Wed, 29 Jun 2011 09:11:59 +0000 (10:11 +0100)]
Make 4 incompatible changes to the GAction API
This commit represents an API break to GAction in the following ways:
- the 'set_state' entry in the GActionInterface vtable has been
renamed to 'change_state'. The number and order of vtable items has
not otherwise changed.
- g_action_set_state() has been renamed to g_action_change_state() to
match the updated vtable entry.
- the "state" property of the GAction interface has been changed to
read-only to reflect the fact that g_action_set_state() no longer
exists.
- GSimpleActionClass has been hidden. GSimpleAction can no longer be
subclassed.
>> Rationale
g_action_set_state() has never been a true setter in the sense that
calling it will update the value of the "state" property. It has always
been closer to "request 'state' to be changed to this value" with
semantics defined by the implementor of the interface. This is why the
equivalent method in GActionGroup had its name changed from 'set' to
'change'. This change makes the two interfaces more consistent and
removes any implication about the effect that calling set_state() should
have on the 'state' property.
>> Impact
This incompatible API break was undertaken only because I strongly
suspect that it will go entirely unnoticed. If the break actually
affects anybody, then we will accommodate them (possibly going as far as
to revert this commit entirely).
The virtual table change only impacts implementors of GAction. I
strongly suspect that this is nobody (except for GSimpleAction).
The hiding of GSimpleActionClass only impacts impacts subclasses of
GSimpleAction. I strongly suspect that none of these exist.
The changing of the property to be read-only only affects people who
were trying to change the state by using GObject properties. I strongly
suspect that this is nobody at all.
The removal of the g_action_set_state() call is the most dangerous, but
I still suspect that it will impact nobody outside of GLib. If anybody
is impacted by this change then, at their request, I will reintroduce
the API as a deprecated alias for g_action_change_state().
Chun-wei Fan [Wed, 29 Jun 2011 06:07:38 +0000 (14:07 +0800)]
Update Visual Studio README.txt's
-The VS2010 README.txt should have Windows CRLF EOL.
-Tell people about the added dependency on LibFFI
Chun-wei Fan [Wed, 29 Jun 2011 04:36:24 +0000 (12:36 +0800)]
Re-attempt to correct EOL on VS2010 solution
This time I realized that I needed to set autocrlf=false on my Windows side
... ugh...
This is one of those files that must have CRLF line endings to work
orrectly :|
Colin Walters [Tue, 28 Jun 2011 16:52:58 +0000 (12:52 -0400)]
GCancellable: Fix build on Win32
https://bugzilla.gnome.org/show_bug.cgi?id=653522
Colin Walters [Tue, 28 Jun 2011 14:03:15 +0000 (10:03 -0400)]
gmain: Fall back to pipes if kernel doesn't support EFD_CLOEXEC for eventfd()
Also remove the caching of checking for eventfd; just try it every time, it's
cheap enough to do so.
https://bugzilla.gnome.org/show_bug.cgi?id=653570
Ihar Hrachyshka [Mon, 27 Jun 2011 23:02:34 +0000 (02:02 +0300)]
Updated Belarusian translation.
Ihar Hrachyshka [Mon, 27 Jun 2011 22:27:27 +0000 (01:27 +0300)]
Updated Belarusian translation.
Ryan Lortie [Mon, 27 Jun 2011 10:28:55 +0000 (11:28 +0100)]
Change to AM_MAINTAINER_MODE([enable])
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=653429
Ihar Hrachyshka [Sat, 25 Jun 2011 16:14:09 +0000 (19:14 +0300)]
Updated Belarusian translation.
Claudio Saavedra [Fri, 24 Jun 2011 13:04:54 +0000 (16:04 +0300)]
GSimpleAction: fix typo
Krzesimir Nowak [Fri, 24 Jun 2011 11:20:06 +0000 (13:20 +0200)]
Typo fix.
Fixes jhbuilding gobject-introspection.
Chun-wei Fan [Fri, 24 Jun 2011 03:22:47 +0000 (11:22 +0800)]
Refine Visual Studio Projects
-Reinstate build/win32/vs10/glib.sln with the correct EOL (DOS/Windows) so
that the file can be correctly recognized by Windows, rather than having
the "Unrecognized Visual Studio Version".
-Update the main GLib projects to output the DLL/LIB file into <Release or
Debug>\<Win32 or x64>\bin for all configurations.
-Update/simplify the property sheets to copy all DLL and LIB files from
<Release or Debug>\<Win32 or x64>\bin for all configurations.
-Update the VS 2008 property sheet to seperate the intermediate directories
for all projects as well to avoid errors/warnings of being unable to
write/access the PDB files as they are in use.
Chun-wei Fan [Fri, 24 Jun 2011 03:13:33 +0000 (11:13 +0800)]
Temporarily remove build/win32/vs10/glib.sln
To fix up for the line endings (this needs to be in the DOS/Windows
format)
Matthias Clasen [Fri, 24 Jun 2011 01:31:40 +0000 (21:31 -0400)]
Unicode: add a g_utf8_substring convenience api
This function is useful in the GTK+ accessibility implementations,
and seems like a nice thing to have around in general.
Christian Persch [Thu, 23 Jun 2011 23:07:18 +0000 (01:07 +0200)]
Typo fix
Christian Persch [Thu, 23 Jun 2011 16:35:40 +0000 (18:35 +0200)]
Plug a mem leak in the regex test
Christian Persch [Thu, 23 Jun 2011 16:31:29 +0000 (18:31 +0200)]
Add boxed GType for GMatchInfo
Christian Persch [Thu, 23 Jun 2011 16:27:29 +0000 (18:27 +0200)]
Make GMatchInfo refcounted
Patrick Welche [Thu, 23 Jun 2011 14:08:46 +0000 (15:08 +0100)]
build: fix gtk-doc when srcdir != builddir
https://bugzilla.gnome.org/show_bug.cgi?id=653250
Sebastian Dröge [Thu, 23 Jun 2011 07:34:59 +0000 (09:34 +0200)]
Fix typo in GTime docs
deprected -> deprecated
Matthias Clasen [Tue, 21 Jun 2011 04:54:00 +0000 (00:54 -0400)]
Add tests for GValueArray
Matthias Clasen [Tue, 21 Jun 2011 04:52:43 +0000 (00:52 -0400)]
Improve test coverage for GValue transformations
Matthias Clasen [Tue, 21 Jun 2011 04:51:57 +0000 (00:51 -0400)]
Run some tests with errorcheck mutexes too
All in the name of better test coverage.
Yuri Kozlov [Wed, 22 Jun 2011 16:10:43 +0000 (20:10 +0400)]
Updated Russian translation
Chun-wei Fan [Wed, 22 Jun 2011 07:22:55 +0000 (15:22 +0800)]
Update config.h.win32(.in)
Make file contents more like the config.h(.in) contents
Colin Walters [Wed, 22 Jun 2011 03:24:07 +0000 (23:24 -0400)]
build: Cache check for futex() and eventfd()
Colin Walters [Wed, 22 Jun 2011 01:43:19 +0000 (21:43 -0400)]
GCancellable: Use Linux eventfd() instead of pipe
See commit
f626dd2b4311bd82137c5b208ab2de288c3e6fae for rationale;
basically it's cheaper than a pipe.
https://bugzilla.gnome.org/show_bug.cgi?id=653140
Colin Walters [Tue, 21 Jun 2011 23:01:10 +0000 (19:01 -0400)]
gmain: use Linux eventfd() for main context wake up
The Linux eventfd() call is basically tailor made for the main loop
wake up pipe - all we want is a threadsafe way to write to a file
descriptor, and wake up the context on the other end; we don't care
about the content at all.
The eventfd manual page basically explains the benefits:
Applications can use an eventfd file descriptor instead of a
pipe (see pipe(2)) in all cases where a pipe is used simply to
signal events. The kernel overhead of an eventfd file
descriptor is much lower than that of a pipe, and only one file
descriptor is required (versus the two required for a pipe).
When writing my multithreaded spawn test case I actually hit the 1024
file descriptor limit quickly, because we used 2 fds per main context.
This brings that down to 1.
https://bugzilla.gnome.org/show_bug.cgi?id=653140
Dan Winship [Tue, 21 Jun 2011 20:14:50 +0000 (16:14 -0400)]
gio/Makefile.am: Remove stray references to $(marshal_sources)
Colin Walters [Tue, 21 Jun 2011 14:26:58 +0000 (10:26 -0400)]
Fix a srcdir!=builddir issue, and export that we support it
Future jhbuild versions will consume the BUILD API and automatically
use a builddir, so mark us as supporting it.
Fran Diéguez [Tue, 21 Jun 2011 11:55:31 +0000 (13:55 +0200)]
Updated Galician translations
Philip Van Hoof [Sat, 18 Jun 2011 17:40:34 +0000 (19:40 +0200)]
Add iter_replace API to GHashTableIter
https://bugzilla.gnome.org/show_bug.cgi?id=652822
Matthias Clasen [Tue, 21 Jun 2011 02:49:34 +0000 (22:49 -0400)]
Make gmarshal.c actually compile standalone
It was lacking the necessary includes; in the past it has been
included towards the bottom of gsignal.c, and gained its includes
that way.
Colin Walters [Mon, 20 Jun 2011 13:06:07 +0000 (09:06 -0400)]
Stop using glib-genmarshal at build time
To help cross compilation, don't use glib-genmarshal in our
build. This is easy now that we have g_cclosure_marshal_generic().
In gobject/, add gmarshal.[ch] to git (making the existing entry
points stubs).
In gio/, simply switch to using g_cclosure_marshal_generic().
https://bugzilla.gnome.org/show_bug.cgi?id=652168
David Zeuthen [Mon, 20 Jun 2011 20:32:03 +0000 (16:32 -0400)]
GDBus: Unref worker from worker-thread to avoid race
... otherwise we might end up using the worker after it has been
freed. Reported by Dan Winship and Colin Walters.
This fix uncovered a bug in the /gdbus/nonce-tcp test case so "fix"
that as well to use a better way of having one thread wait for another
(using quotes for the word "fix" since it's pretty hackish to
busy-wait in one thread to wait for another).
Signed-off-by: David Zeuthen <davidz@redhat.com>
Colin Walters [Mon, 20 Jun 2011 18:09:31 +0000 (14:09 -0400)]
gthread tests: Remove spurious return;
Colin Walters [Mon, 20 Jun 2011 15:15:27 +0000 (11:15 -0400)]
gclosure: Add G_TYPE_ENUM and G_TYPE_PARAM for g_cclosure_marshal_generic()
These are fundamental types missed in the original commit.
Colin Walters [Mon, 20 Jun 2011 17:41:52 +0000 (13:41 -0400)]
gdbus-test-codegen: Allocate full integer space for uint16 properties
Because there is no G_TYPE_[U]INT16, we need to use integer types,
otherwise we'll end up overwriting memory (in this case on the stack).
Javier Jardón [Mon, 20 Jun 2011 10:48:25 +0000 (11:48 +0100)]
configure.ac: Generate xz tarballs by default
Chun-wei Fan [Mon, 20 Jun 2011 03:50:57 +0000 (11:50 +0800)]
Visual C++ 2010 project files update
Consolidate my previous commit into property file... should have realized
this sooner...
Chun-wei Fan [Mon, 20 Jun 2011 03:34:15 +0000 (11:34 +0800)]
Update Visual C++ 2010 Projects
-Seperate intermediate directories for each project to avoid intermittent
MSBuild errors that a build log cannot be written while in use, and
update the property sheet as necessary.
-Minor cleanups of uneeded tags in the projects/properties
Christian Persch [Sat, 18 Jun 2011 14:22:53 +0000 (16:22 +0200)]
Clarify nul-termination of g_utf8_to_ucs4[_fast] result
The docs for g_utf8_to_ucs4_fast didn't mention that the resulting
string is terminated by a 0 character.
Bug #652897.
Matthias Clasen [Sun, 19 Jun 2011 05:42:48 +0000 (01:42 -0400)]
Add tests for GCache
Matthias Clasen [Sun, 19 Jun 2011 05:05:26 +0000 (01:05 -0400)]
asyncqueue: improve test coverage
Matthias Clasen [Sun, 19 Jun 2011 05:05:05 +0000 (01:05 -0400)]
utils: silence tests unless verbose
Matthias Clasen [Sun, 19 Jun 2011 05:04:48 +0000 (01:04 -0400)]
logging: improve test coverage
Matthias Clasen [Sun, 19 Jun 2011 05:04:30 +0000 (01:04 -0400)]
hash: Improve test coverage
Chun-wei Fan [Sat, 18 Jun 2011 09:55:58 +0000 (17:55 +0800)]
Update gsettings/glib-compile-schemas VS Projects
These tools require the use of GModule headers also, so update the include
directories so that the correct gmodule.h will be included instead of the
system-installed version.
Colin Walters [Fri, 17 Jun 2011 13:39:48 +0000 (09:39 -0400)]
gvariant: Mark g_variant_new_variant as constructor
https://bugzilla.gnome.org/show_bug.cgi?id=647796
Colin Walters [Fri, 17 Jun 2011 12:52:05 +0000 (08:52 -0400)]
gdatainputstream: Add Since: tags and update gio-sections.txt
https://bugzilla.gnome.org/show_bug.cgi?id=652758
Colin Walters [Thu, 16 Jun 2011 18:00:36 +0000 (14:00 -0400)]
GDataInputStream: Add _utf8() variants of _read_line
These will validate the resulting line, and throw a conversion error.
In practice these will likely be used by bindings, but it's good
for even C apps too that don't want to explode if that text file
they're reading into Pango actually has invalid UTF-8.
https://bugzilla.gnome.org/show_bug.cgi?id=652758
Colin Walters [Thu, 16 Jun 2011 17:14:44 +0000 (13:14 -0400)]
GDataInputStream: Clarify read_line() docs, mark as byte array
g_data_input_stream_read_line() and
g_data_input_stream_read_line_finish() don't do any encoding checks,
so we shouldn't call the returned value a "string" (which I'd like to
mean UTF-8). Annotate them as byte arrays and add encoding warnings
to the docstrings.
https://bugzilla.gnome.org/show_bug.cgi?id=652758
Patrick Welche [Tue, 14 Jun 2011 08:43:33 +0000 (09:43 +0100)]
Fix build due to trivial typos in
ab0e9dbf.
https://bugzilla.gnome.org/show_bug.cgi?id=652750
Murray Cumming [Wed, 15 Jun 2011 08:47:13 +0000 (10:47 +0200)]
G_STATIC_REC_MUTEX_INIT: Add a 0 to the initialization.
This avoids a compiler warning about an incomplete initialization
when using this.
Ryan Lortie [Wed, 15 Jun 2011 02:44:10 +0000 (22:44 -0400)]
Drop dead code in GVariant parser
There is no chance that an unsigned integer value will be negative after
we do the bounds check that enforces its non-negativity.
Caught by Matthias running Coverity.
Matthias Clasen [Wed, 15 Jun 2011 00:44:15 +0000 (20:44 -0400)]
Eliminate some dead code
Matthias Clasen [Wed, 15 Jun 2011 00:36:37 +0000 (20:36 -0400)]
Remove a redundant NULL check
Matthias Clasen [Wed, 15 Jun 2011 00:15:04 +0000 (20:15 -0400)]
Silence a compiler warning
Matthias Clasen [Tue, 14 Jun 2011 23:52:19 +0000 (19:52 -0400)]
Add some argument checks to filename conversion functions
Colin Walters [Tue, 14 Jun 2011 22:46:06 +0000 (18:46 -0400)]
gmain: Close race condition in _g_main_wake_up_all_contexts()
Running gthread/tests/spawn-multithreaded in a loop, I very easily hit:
GLib-CRITICAL **: g_main_context_wakeup: assertion `g_atomic_int_get (&context->ref_count) > 0' failed
Testing the refcount still left a window where we would fall into the
assertion. Fix this by just locking the context.
Colin Walters [Wed, 8 Jun 2011 15:31:14 +0000 (11:31 -0400)]
gmain: Only run through signal delivery once per read()
This is what I intended to do before.
https://bugzilla.gnome.org/show_bug.cgi?id=652072
Colin Walters [Sat, 11 Jun 2011 16:10:57 +0000 (12:10 -0400)]
gmain: Use sigset_t for keeping track of installed signals
Minor code cleanup.
https://bugzilla.gnome.org/show_bug.cgi?id=652072
Colin Walters [Fri, 10 Jun 2011 14:48:07 +0000 (10:48 -0400)]
gspawn: Reset signal handlers for synchronous spawning
We should by default reset signal handlers; particularly with the
ability to install them via g_unix_signal_source_new(), we don't
want to call a user callback inside a fork()ed helper process.
https://bugzilla.gnome.org/show_bug.cgi?id=652072
Colin Walters [Sat, 11 Jun 2011 21:25:13 +0000 (17:25 -0400)]
glib/tests/unix.c: Also test SIGTERM
We expect this to not kill the process, so it'd be a good one to test.
https://bugzilla.gnome.org/show_bug.cgi?id=652072
Colin Walters [Fri, 10 Jun 2011 14:35:27 +0000 (10:35 -0400)]
spawn-multithreaded: Clean up IO channel code
I modeled the new bits after how gunixmount.c handles GIOChannel;
it's apparently easier not to look at the condition.
https://bugzilla.gnome.org/show_bug.cgi?id=652072
Colin Walters [Fri, 10 Jun 2011 14:14:25 +0000 (10:14 -0400)]
gspawn: Handle EINTR in a few more cases
I was debugging gthread/tests/spawn-multithreaded.c, and while I
don't think I actually hit EINTR in any of these cases, it'd be
good to fix them anyways.
https://bugzilla.gnome.org/show_bug.cgi?id=652072
Colin Walters [Fri, 10 Jun 2011 11:29:28 +0000 (07:29 -0400)]
spawn-singlethread.c: New test
https://bugzilla.gnome.org/show_bug.cgi?id=652072
Colin Walters [Fri, 10 Jun 2011 11:27:29 +0000 (07:27 -0400)]
spawn-multithreaded: Also look for lt-test-echo
I hate libtool =(
https://bugzilla.gnome.org/show_bug.cgi?id=652072
Matthias Clasen [Tue, 14 Jun 2011 20:32:55 +0000 (16:32 -0400)]
If dup fails don't call close()
Matthias Clasen [Tue, 14 Jun 2011 20:20:26 +0000 (16:20 -0400)]
Remove a redundant assertion
A gsize is never going below 0.
Matthias Clasen [Tue, 14 Jun 2011 20:18:45 +0000 (16:18 -0400)]
Don't compare unsigned numbers with 0
It does not work.
David Zeuthen [Tue, 14 Jun 2011 22:11:09 +0000 (18:11 -0400)]
GDBusMessage: Optimize serializer and deserializer
... by using a switch instead of if-then-else.
Signed-off-by: David Zeuthen <davidz@redhat.com>
Colin Walters [Tue, 14 Jun 2011 16:28:57 +0000 (12:28 -0400)]
gclosure: Add missing include
Matthias Clasen [Tue, 14 Jun 2011 15:40:32 +0000 (11:40 -0400)]
Don't do < 0 on an unsigned value
It won't work. read() returns a signed value, anyway.
Matthias Clasen [Tue, 14 Jun 2011 14:16:35 +0000 (10:16 -0400)]
Silence a compiler warning
Even though we are confident the filename will always
end in .gmarkup, the compiler doesn't know that...
Matthias Clasen [Tue, 14 Jun 2011 14:12:46 +0000 (10:12 -0400)]
Don't leak resources in error cases
Matthias Clasen [Tue, 14 Jun 2011 14:07:22 +0000 (10:07 -0400)]
Don't forget to free resources in an early return