Ryan Lortie [Thu, 10 Jun 2010 12:06:32 +0000 (08:06 -0400)]
Mention GVariant format string docs from iter docs
It's not entirely clear what @format_string in iter_next() and
iter_loop() should be. Include a link to the GVariant format string
docs as a hint.
Tor Lillqvist [Thu, 10 Jun 2010 08:52:25 +0000 (11:52 +0300)]
Plug memory leak on Windows
Intern the string returned from g_win32_getlocale() and then free it.
Fixes bug #621168.
Matthias Clasen [Thu, 10 Jun 2010 03:44:13 +0000 (23:44 -0400)]
Fix GApplication tests to run without a session bus
We reuse code from the GDBus tests here to launch a session bus.
David Zeuthen [Wed, 9 Jun 2010 21:57:04 +0000 (17:57 -0400)]
GDBusConnection: Do not dispatch calls to unregistered objects or subtrees
There was a slight race where we ended up calling into user code if
the user managed to unregister an object (or subtree) in the window
between
- processing the remote call on the worker thread; and
- continuing handling it on the user code thread (via an idle handler)
This patch fixes the problem.
Signed-off-by: David Zeuthen <davidz@redhat.com>
David Zeuthen [Wed, 9 Jun 2010 21:08:34 +0000 (17:08 -0400)]
GDBusProxy: Fix error handling in synchronous initialization codepath
Signed-off-by: David Zeuthen <davidz@redhat.com>
Ryan Lortie [Wed, 9 Jun 2010 16:43:13 +0000 (12:43 -0400)]
g_variant_builder_add_parsed: fix type error
Problem caught by Juan A. Suarez Romero
Richard Hughes [Wed, 9 Jun 2010 08:36:47 +0000 (09:36 +0100)]
Do not do update-po at distcheck time to avoid touching files checked into version control
David Zeuthen [Wed, 9 Jun 2010 14:56:35 +0000 (10:56 -0400)]
Bug 621119 – GDBusProxy and objects with no properties
Fix proxy construction for objects with no properties in the case
where G_DBUS_PROXY_FLAGS_DO_NOT_LOAD_PROPERTIES isn't set.
The unfortunate side-effect here is that GDBusProxy can no longer be
used to test for "object existence", e.g. creating a GDBusProxy for
any path and interface will not fail. But that's not really a big
deal, if apps rely on that they are doing something very wrong.
https://bugzilla.gnome.org/show_bug.cgi?id=621119
Signed-off-by: David Zeuthen <davidz@redhat.com>
Juan A. Suarez Romero [Wed, 9 Jun 2010 08:27:39 +0000 (10:27 +0200)]
Fix warning
Disable functions defined but not used.
Juan A. Suarez Romero [Wed, 9 Jun 2010 08:23:39 +0000 (10:23 +0200)]
Fix warnings
Do explicit casts to avoid warnings.
blue dark [Wed, 9 Jun 2010 07:13:14 +0000 (15:13 +0800)]
Updated zh_CN translation.
Christian Persch [Tue, 8 Jun 2010 17:40:58 +0000 (19:40 +0200)]
Fix thinko
When replacing strcmp() with g_variant_is_of_type(), remove the "== 0"
part! Found by Colin Walters.
Colin Walters [Tue, 8 Jun 2010 15:21:48 +0000 (11:21 -0400)]
[GDBusConnection] Use Gio's default async implementation again
The fix was committed in git.
https://bugzilla.gnome.org/show_bug.cgi?id=620990
Christian Persch [Tue, 8 Jun 2010 11:40:38 +0000 (13:40 +0200)]
g_dbus_method_invocation_return_value consumes the floating variant
Bug #620953.
Christian Persch [Tue, 8 Jun 2010 11:36:58 +0000 (13:36 +0200)]
Simplify variant builder
Build the full return value with one builder, and don't unref the
unowned return value!
Bug #620954.
Christian Persch [Tue, 8 Jun 2010 11:36:00 +0000 (13:36 +0200)]
Use g_variant_is_of_type()
... instead of strcmp()'ing the type strings.
Bug #620954.
Christian Persch [Tue, 8 Jun 2010 11:35:22 +0000 (13:35 +0200)]
Plug a mem leak
Bug #620954.
Javier Jardón [Tue, 8 Jun 2010 14:25:38 +0000 (16:25 +0200)]
[docs] Fix typos in some g_file_* functions
Reported by Alexander Saprykin in bug
https://bugzilla.gnome.org/show_bug.cgi?id=620947
Matthias Clasen [Tue, 8 Jun 2010 05:10:17 +0000 (01:10 -0400)]
Bump version
Matthias Clasen [Tue, 8 Jun 2010 05:09:06 +0000 (01:09 -0400)]
2.25.8
Matthias Clasen [Tue, 8 Jun 2010 03:41:06 +0000 (23:41 -0400)]
Add missing marshaler
Matthias Clasen [Tue, 8 Jun 2010 03:28:30 +0000 (23:28 -0400)]
Fix a missing parameter in a doc comment
Pointed out by David Zeuthen.
Matthias Clasen [Tue, 8 Jun 2010 03:27:21 +0000 (23:27 -0400)]
Fix a signal signature
GApplication::action was erroneously declaring the timestamp parameter
as int instead of uint.
Matthias Clasen [Tue, 8 Jun 2010 02:34:07 +0000 (22:34 -0400)]
Fix a misspelt doc comment
Matthias Clasen [Tue, 8 Jun 2010 02:22:39 +0000 (22:22 -0400)]
Add a lost <SECTION>
Matthias Clasen [Tue, 8 Jun 2010 02:21:47 +0000 (22:21 -0400)]
Rename GApplication::appid to GApplication::application-id
Matthias Clasen [Tue, 8 Jun 2010 01:53:32 +0000 (21:53 -0400)]
Fix !srcdir checks
Matthias Clasen [Tue, 8 Jun 2010 01:23:37 +0000 (21:23 -0400)]
Updates
Matthias Clasen [Mon, 7 Jun 2010 22:27:33 +0000 (18:27 -0400)]
Document signals
Javier Jardón [Mon, 7 Jun 2010 21:55:34 +0000 (23:55 +0200)]
[docs] GApplication is available since Gio 2.26
Matthias Clasen [Mon, 7 Jun 2010 21:48:09 +0000 (17:48 -0400)]
Use g types for consistency
Matthias Clasen [Mon, 7 Jun 2010 17:25:39 +0000 (13:25 -0400)]
Merge the wip/gapplication branch
This adds a GApplication object to GIO, which is the core of
an application support class, supporting
- uniqueness
- exporting actions (simple scripting)
- standard actions (quit, activate)
The implementation for Linux uses D-Bus, takes a name on the
session bus, and exports a org.gtk.Application interface.
Implementations for Win32 and OS X are still missing.
Ryan Lortie [Mon, 7 Jun 2010 10:58:57 +0000 (12:58 +0200)]
Bug 620767 - Typo in GSettings documentation
Use the correct variable name and work around the escaping of '@'.
Expand/clarify the section on how translation of <default> is handled.
Matthias Clasen [Mon, 7 Jun 2010 10:23:24 +0000 (06:23 -0400)]
Include the right header
Ryan Lortie [Mon, 7 Jun 2010 08:18:43 +0000 (10:18 +0200)]
Bug 620496 - schema compiler: reject invalid paths
The GSettings schema compiler was accepting any string as a path. It is
probably quite a common mistake to suspect that '/apps/foo' is a valid
path name when this will cause all sorts of trouble later. Check for
this case and report the error.
Matthias Clasen [Sun, 6 Jun 2010 20:42:06 +0000 (16:42 -0400)]
Make g_assertion_message_error take a const GError*
This was requested in bug 620265.
Christian Persch [Sun, 6 Jun 2010 20:32:04 +0000 (16:32 -0400)]
Don't do an extra strlen when g_variant_get_string() returns it already
Matthias Clasen [Sun, 6 Jun 2010 20:20:21 +0000 (16:20 -0400)]
Add single-include guards to new headers
As pointed out by Christian Persch in bug 620173, all the new
gdbus and gsettings headers were missing these.
Matthias Clasen [Sun, 6 Jun 2010 20:15:44 +0000 (16:15 -0400)]
Sort gio.h includes alphabetically
Michael Natterer [Sun, 6 Jun 2010 19:15:42 +0000 (21:15 +0200)]
gio: fix the build
Matthias Clasen [Sun, 6 Jun 2010 18:22:48 +0000 (14:22 -0400)]
Some cleanups
Always include config.h, make property strings for translation,
add since tags.
Will Thompson [Thu, 3 Jun 2010 14:09:09 +0000 (15:09 +0100)]
Add some symbols to glib-sections.txt
Most of these are private; the few that aren't were already documented
but not included in the gtkdoc output.
Will Thompson [Thu, 3 Jun 2010 14:08:58 +0000 (15:08 +0100)]
Document G_GNUC_DEPRECATED_FOR
Will Thompson [Thu, 3 Jun 2010 13:50:19 +0000 (14:50 +0100)]
Add examples for G_STRINGIFY and G_PASTE
Will Thompson [Thu, 3 Jun 2010 13:43:30 +0000 (14:43 +0100)]
Hide G_PASTE_ARGS in gtkdocs.
Javier Jardón [Sun, 6 Jun 2010 03:25:59 +0000 (05:25 +0200)]
[docs] Improve the g_get_system_config_dirs() docs.
The retrieved list of directories is XDG_CONFIG_DIRS,
Also mention the retrieved directory on Windows: CSIDL_COMMON_APPDATA
Javier Jardón [Sun, 6 Jun 2010 03:13:06 +0000 (05:13 +0200)]
[docs] Improve the g_get_system_data_dirs() docs.
The retrieved list of directories is XDG_DATA_DIRS
Javier Jardón [Sun, 6 Jun 2010 03:05:15 +0000 (05:05 +0200)]
[docs] Improve the g_get_user_cache_dir() docs
The retrieved directory is XDG_CACHE_HOME on UNIX platforms.
Also mention the retrieved directory on Windows: CSIDL_INTERNET_CACHE.
Javier Jardón [Sun, 6 Jun 2010 02:57:46 +0000 (04:57 +0200)]
[docs] Improve the g_get_user_data_dir() docs
The retrieved directory is XDG_DATA_HOME on UNIX platforms.
Also mention the retrieved directory on Windows: CSIDL_PERSONAL.
Javier Jardón [Sat, 5 Jun 2010 16:34:38 +0000 (18:34 +0200)]
[docs] Improve the g_get_user_config_dir() docs a bit
The retrieved directory is XDG_CONFIG_HOME on UNIX platforms.
Also mention the retrieved directory on Windows: CSIDL_APPDATA.
Ryan Lortie [Fri, 4 Jun 2010 21:07:40 +0000 (23:07 +0200)]
add get_permission API to GSettingsBackend
implement it in the various in-tree backends
also, lots of whitespace changes to realign the vtable members
Ryan Lortie [Fri, 4 Jun 2010 21:02:44 +0000 (23:02 +0200)]
gsettingsbackend.h: pretend to be gio.h
Since #include <gsettingsbackend.h> is a perfectly valid thing for
applications to do, and since we want to include gio headers from
gsettingsbackend.h, we need to effectively disable the #error we would
get from those headers (because we're not coming via gio.h).
We don't want to #include <gio/gio.h> here because this would cause
needless rebuilding of GSettingsBackend, GSettings,
GDelayedSettingsBackend, etc... every time someone changed anything in
any public header.
Ryan Lortie [Fri, 4 Jun 2010 20:32:01 +0000 (22:32 +0200)]
Bug 620582 - a simple GPermission implementation
add GSimplePermission, a trivial const implementation of GPermission
can-request and can-release are always false for this implementation and
the value of 'allowed' is decided at construction.
Matej Urbančič [Fri, 4 Jun 2010 18:03:41 +0000 (20:03 +0200)]
Updated Slovenian translation
Ryan Lortie [Thu, 3 Jun 2010 20:58:52 +0000 (22:58 +0200)]
Bug 620519 - GPermission
Add an abstract interface representing the permission to perform an
action.
Murray Cumming [Fri, 4 Jun 2010 15:07:05 +0000 (17:07 +0200)]
Gio: gioenums.h: Remove trailing commas to avoid C++ warnings.
Javier Jardón [Sun, 30 May 2010 12:56:55 +0000 (14:56 +0200)]
gdbusaddress: Fix typo
Kristjan Schmidt [Thu, 3 Jun 2010 16:19:31 +0000 (18:19 +0200)]
Updated Esperanto translation
Matthias Clasen [Thu, 3 Jun 2010 15:40:02 +0000 (11:40 -0400)]
Don't spew a g_warning if inotify setup fails
We use is_supported when we are trying to find a local file monitor
implementation that works, and having the g_warning in there trips
the test suite.
Ryan Lortie [Thu, 3 Jun 2010 09:24:31 +0000 (11:24 +0200)]
Bug 620350 - add g_variant_builder_add_parsed()
A delicious blend of g_variant_new_parsed() and
g_variant_builder_add_value(). Now available in a GLib near you.
Ryan Lortie [Thu, 3 Jun 2010 07:41:33 +0000 (09:41 +0200)]
Bug 620349 – utf8ify GVariant printer
Take advantage of our knowledge that GVariant strings are always valid
utf8 when printing and parsing:
- allow valid printing unicode characters to pass through unescaped
- escape non-printing characters using \uxxxx or \Uxxxxxxxx format
- do the same in the parser
- update existing test cases to use utf8, add a new test case
Milan Bouchet-Valat [Wed, 2 Jun 2010 14:05:13 +0000 (16:05 +0200)]
Annotate GVariant and GSettings _strv() functions
Add GObject introspection annotations so that the length parameter is
correctly detected for g_variant_new_strv(), g_variant_get_strv() and
g_variant_dup_strv(). Also specify that it can be a NULL pointer in
g_variant_get_strv() and g_variant_dup_strv().
For g_settings_set_strv(), detect that a NULL value is allowed, meaning
empty array.
Closes bug #620384.
Signed-off-by: Ryan Lortie <desrt@desrt.ca>
Ryan Lortie [Wed, 2 Jun 2010 17:34:45 +0000 (19:34 +0200)]
improve gitignore (systemtap)
Kjartan Maraas [Wed, 2 Jun 2010 13:41:01 +0000 (15:41 +0200)]
Updated Norwegian bokmål translation
Ryan Lortie [Wed, 2 Jun 2010 02:00:58 +0000 (04:00 +0200)]
Support NULL value for g_settings_set_strv()
Allow easy setting of the empty array that we lost with the last patch.
Milan Bouchet-Valat [Tue, 1 Jun 2010 21:16:19 +0000 (23:16 +0200)]
Remove length parameter for g_settings_[gs]et_strv
Length of the array is redundant since it's NULL-terminated. This is not
consistent with many GLib and GTK+ functions, and adds complexity with
no real gain, while these convenience functions should be kept simple.
Closes bug #620312
Jorge González [Sun, 30 May 2010 16:23:21 +0000 (18:23 +0200)]
Updated Spanish translation
Fran Diéguez [Sun, 30 May 2010 15:43:57 +0000 (17:43 +0200)]
Updated Galician translations
Andika Triwidada [Sun, 30 May 2010 10:10:51 +0000 (17:10 +0700)]
Updated Indonesian translation
Jorge González [Sat, 29 May 2010 10:06:16 +0000 (12:06 +0200)]
Updated Spanish translation
Jorge González [Fri, 28 May 2010 18:21:02 +0000 (20:21 +0200)]
Updated Spanish translation
Jorge González [Fri, 28 May 2010 17:45:51 +0000 (19:45 +0200)]
Updated Spanish translation
Yaron Shahrabani [Fri, 28 May 2010 13:57:56 +0000 (16:57 +0300)]
Updated Hebrew translation.
Yaron Shahrabani [Fri, 28 May 2010 13:54:58 +0000 (16:54 +0300)]
Updated Hebrew translation.
Lin Ma [Fri, 28 May 2010 08:57:10 +0000 (16:57 +0800)]
After talk with FEN dev, we dicide simply disable monitor function if
the current filesystem doesn't suport FEN.
Alexander Larsson [Fri, 1 Jan 2010 20:39:52 +0000 (21:39 +0100)]
Add dtrace and systemtap support for gobject
This adds static markers and systemtap tapsets for:
* type creation
* object lifetimes (creation, ref, unref, dispose, finalize)
* signal creation and emission
Signal emissions and finalization marker have a corresponding
*_end (or *-end in dtrace) version that is when the corresponding
operation is finished.
https://bugzilla.gnome.org/show_bug.cgi?id=606044
Alexander Larsson [Fri, 18 Dec 2009 20:25:47 +0000 (21:25 +0100)]
Initial support for dtrace and systemtap
This adds static markers for dtrace, which are also usable
by systemtap. Additionally it adds a tapset for systemtap
that makes it easier to use the static markers.
These are enabled by default.
This initial set of probes is rather limited:
* allocation and free using g_malloc & co
* allocation and free using g_slice
* gquark name tracking (useful for converting quarks to strings in probes)
Notes on naming:
Its traditional with dtrace to use probe names with dashes as
delimiter (slice-alloc). Since dashes are not usable in identifiers
the C code uses double underscores (slice__alloc) which is converted
to dashes in the UI. We follow this for the shared lowlevel probe
names.
Additionally dtrace supports putting a "provider" part in the probe
names which is essentially a namespacing thing. On systemtap this
field is currently ignored (but may be implemented in the future), but
this is not really a problem since in systemtap the probes are
specified by combining the solib file and the marker name, so there
can't really be name conflicts.
For the systemtap tapset highlevel probes we instead use names that
are systemtapish with single dashes as separators.
https://bugzilla.gnome.org/show_bug.cgi?id=606044
Mikhail Zabaluev [Thu, 27 May 2010 18:00:12 +0000 (14:00 -0400)]
Optimized the overlong sequence check in g_utf8_get_char_extended()
Rather make it branch to get the due sequence length for the resulting
character code, we can as well get the minimum code value in the initial
branching.
Ryan Lortie [Thu, 27 May 2010 17:11:49 +0000 (13:11 -0400)]
GSettings schema docs: mention need for EXTRA_DIST
Mention that you will need to use EXTRA_DIST for your schema file if it
is distributed with your project.
Ryan Lortie [Thu, 27 May 2010 17:07:54 +0000 (13:07 -0400)]
Remove duplicate copies of migration docs
Fixup for commit
133f66538dbf266be3c99b34f1eeee0a5e6068ac which
duplicated the contents of most of the migration documentation by
splitting it out into separate files but keeping the original file
intact (with a rename).
This removes the duplicated content from the renamed file.
Ryan Lortie [Thu, 27 May 2010 15:58:54 +0000 (11:58 -0400)]
gsettings m4: check for .xml in src/builddir
This checks for the .gschema.xml file in the srcdir and builddir and
runs the schema validation on which one it finds. This handles
non-srcdir builds in both cases: .gschema.xml is in the tarball and
.gschema.xml is generated.
Ryan Lortie [Thu, 27 May 2010 15:32:34 +0000 (11:32 -0400)]
GVariant: One more FreeBSD fix
FreeBSD's malloc() sometimes returns unaligned memory if you are
requesting small sizes. This can get GVariant into trouble. For
example, consider the type "mmi" containing the value "just nothing".
According to the type signature, the memory containing this should be
aligned to a boundary of 4 since it might contain an int. The
serialised size of this value is 1 byte, however, and when you ask
FreeBSD to allocate memory of that size, it knows you can't put an int
into it so it doesn't bother aligning it.
This patch modifies the GVariant serialiser to not assert the alignment
constraint in the case that the size of the serialised data is smaller
than its own alignment requirement.
Ryan Lortie [Thu, 27 May 2010 03:34:31 +0000 (23:34 -0400)]
More alignment-related fixes.
Partial-backout
8a21d8d23317ecebe46007f1fd5f7459bf182415. The
assertions should have remained relaxed since these functions are used
with non-posix_memalign()ed data.
Lin Ma [Thu, 27 May 2010 02:24:58 +0000 (10:24 +0800)]
Remove unused code.
Ryan Lortie [Wed, 26 May 2010 21:35:18 +0000 (17:35 -0400)]
include "config.h" for HAVE_POSIX_MEMALIGN
Ryan Lortie [Wed, 26 May 2010 21:31:52 +0000 (17:31 -0400)]
Use aligned memory in GVariant test cases.
The GVariant serialiser works well with non-8-aligned memory, but the
comparison serialiser in the test case depends on memory being
8-aligned. Use posix_memalign() to get the memory used by this
serialiser.
Colin Walters [Wed, 26 May 2010 20:21:15 +0000 (16:21 -0400)]
[mainloop-test] Fix compilation errors
Colin Walters [Wed, 26 May 2010 19:59:36 +0000 (15:59 -0400)]
[tests/gvariant] Handle flavored_free(NULL), since flavored_malloc can return NULL
Jorge González [Wed, 26 May 2010 16:12:22 +0000 (18:12 +0200)]
Updated Spanish translation
Jorge González [Wed, 26 May 2010 16:09:46 +0000 (18:09 +0200)]
Updated Spanish translation
Lin Ma [Wed, 26 May 2010 06:32:37 +0000 (14:32 +0800)]
bugster#
6955199, on hsfs portfs will fail, and FEN backend will fail
to run lstat and port_associate on root node.
Havoc Pennington [Tue, 20 Apr 2010 21:47:44 +0000 (17:47 -0400)]
GSource: add g_source_set_name, g_source_get_name, g_source_set_name_by_id
These allow applications to give meaningful names to their sources.
Source names can then be used for debugging and profiling, for
example with systemtap or gdb.
https://bugzilla.gnome.org/show_bug.cgi?id=606044
Havoc Pennington [Tue, 20 Apr 2010 21:47:44 +0000 (17:47 -0400)]
GSource: add g_source_set_name, g_source_get_name, g_source_set_name_by_id
These allow applications to give meaningful names to their sources.
Source names can then be used for debugging and profiling, for
example with systemtap or gdb.
https://bugzilla.gnome.org/show_bug.cgi?id=606044
David Zeuthen [Tue, 25 May 2010 15:59:57 +0000 (11:59 -0400)]
GDBus: Big-endian fixes
Tested this on my Powerbook G4 12", 867MHz PowerPC G4 running Fedora
11ish.
Signed-off-by: David Zeuthen <davidz@redhat.com>
Ryan Lortie [Tue, 25 May 2010 15:25:34 +0000 (11:25 -0400)]
GVariant: deal with non-8-aligned malloc()
Closes bug #619585.
Matthias Clasen [Tue, 25 May 2010 14:43:29 +0000 (10:43 -0400)]
Document that vtable is not copied
Fran Diéguez [Tue, 25 May 2010 07:56:33 +0000 (09:56 +0200)]
Updated Galician translations
Ryan Lortie [Tue, 25 May 2010 03:21:01 +0000 (23:21 -0400)]
.gitignore manpages (*.1)
Ryan Lortie [Tue, 25 May 2010 03:03:36 +0000 (23:03 -0400)]
post-release version bump.
Ryan Lortie [Tue, 25 May 2010 03:02:18 +0000 (23:02 -0400)]
Release 2.25.7.