platform/upstream/glib.git
13 years agostore some extra info in GSettingsKeyInfo
Ryan Lortie [Sun, 13 Nov 2011 22:31:39 +0000 (23:31 +0100)]
store some extra info in GSettingsKeyInfo

13 years agoGSettingsKeyInfo: drop unused variable
Ryan Lortie [Sun, 13 Nov 2011 22:24:24 +0000 (23:24 +0100)]
GSettingsKeyInfo: drop unused variable

13 years agorename gsettingsschema.h to -internal.h
Ryan Lortie [Sun, 13 Nov 2011 20:41:01 +0000 (21:41 +0100)]
rename gsettingsschema.h to -internal.h

13 years agounGObjectify GSettingsSchema
Ryan Lortie [Sun, 13 Nov 2011 20:38:31 +0000 (21:38 +0100)]
unGObjectify GSettingsSchema

13 years agoFix the fix for G*_TO_POINTER casts on 32 bits
Christian Persch [Thu, 17 Nov 2011 13:21:24 +0000 (14:21 +0100)]
Fix the fix for G*_TO_POINTER casts on 32 bits

The 'fix' from commit 16292dd753cb63f8ccb2267644aefbd2632dd52c broke the
build on 32 bit because it was missing the parentheses around "gint" / "guint"
in glib_gp[u]i_cast.

https://bugzilla.gnome.org/show_bug.cgi?id=661546

13 years agoUpdated Spanish translation
Jorge González [Thu, 17 Nov 2011 13:21:12 +0000 (14:21 +0100)]
Updated Spanish translation

13 years agoUpdated Spanish translation
Daniel Mustieles [Thu, 17 Nov 2011 11:36:54 +0000 (12:36 +0100)]
Updated Spanish translation

13 years agoFix doc formatting
Matthias Clasen [Thu, 17 Nov 2011 05:01:18 +0000 (00:01 -0500)]
Fix doc formatting

13 years agoDeal gracefully with NULL log domains
Matthias Clasen [Thu, 17 Nov 2011 02:42:28 +0000 (21:42 -0500)]
Deal gracefully with NULL log domains

13 years ago[notify] remove some rather bogus 'inline' use
Ryan Lortie [Wed, 16 Nov 2011 15:46:19 +0000 (15:46 +0000)]
[notify] remove some rather bogus 'inline' use

13 years ago[notify] add warning to gobjectnotifyqueue.c
Ryan Lortie [Wed, 16 Nov 2011 15:45:31 +0000 (15:45 +0000)]
[notify] add warning to gobjectnotifyqueue.c

Mention that this file gets installed, so people should probably not
modify it.

13 years ago[notify] dispatch 'notify' directly if not frozen
Ryan Lortie [Wed, 16 Nov 2011 13:02:23 +0000 (13:02 +0000)]
[notify] dispatch 'notify' directly if not frozen

Avoid the notify queue in the case that we're not already frozen.

13 years ago[notify] add 'conditional' to _notify_queue_freeze
Ryan Lortie [Wed, 16 Nov 2011 15:42:36 +0000 (15:42 +0000)]
[notify] add 'conditional' to _notify_queue_freeze

When the 'conditional' parameter is TRUE, the queue will only be frozen
(ie: have its freeze count increase by one) if it is already frozen.

This will allow us to avoid a freeze-notify-thaw in the case that we
just want to notify on a single property.

Another approach may have been to add an is_frozen() type call and avoid
even increasing the freeze count at all in this case.  Unfortunately,
I'm not totally sure what is the exact expected semantics of
simultaneous notifications in multiple threads and this may interact
badly with someone freezing or thawing in between our check and
emission.

13 years ago[notify] Remove GObjectNotifyContext indirection
Ryan Lortie [Wed, 16 Nov 2011 15:41:06 +0000 (15:41 +0000)]
[notify] Remove GObjectNotifyContext indirection

This silly abstraction is causing more trouble than it's worth -- just
use the relevant bits directly.

13 years ago[notify] lift some logic out of _notify_queue_add
Ryan Lortie [Wed, 16 Nov 2011 15:38:25 +0000 (15:38 +0000)]
[notify] lift some logic out of _notify_queue_add

Lift the check-if-READABLE and redirect-target logic from out of
g_object_notify_queue_add() into its own function, get_notify_pspec().

Use that function at the site of our two calls to
g_object_notify_queue_add().

13 years ago[notify] remove an obviously false comment
Ryan Lortie [Wed, 16 Nov 2011 15:37:17 +0000 (15:37 +0000)]
[notify] remove an obviously false comment

13 years ago[notify] lift some code outside of critical region
Ryan Lortie [Wed, 16 Nov 2011 15:36:57 +0000 (15:36 +0000)]
[notify] lift some code outside of critical region

13 years ago[notify] drop some unused code
Ryan Lortie [Wed, 16 Nov 2011 15:35:58 +0000 (15:35 +0000)]
[notify] drop some unused code

13 years ago[notify] merge gobjectnotifyqueue.c into gobject.c
Ryan Lortie [Wed, 16 Nov 2011 15:31:58 +0000 (15:31 +0000)]
[notify] merge gobjectnotifyqueue.c into gobject.c

This was done as a separate file before, and #include brought it into
gobject.c.  That's a bit mad, so stop doing that.

Unfortunately, the insanity steps up a level: gobjectnotifyqueue.c is
installed in the public include dir, so we can't just get rid of it
entirely.

13 years ago[notify] remove some useless forward declarations
Ryan Lortie [Wed, 16 Nov 2011 12:47:40 +0000 (12:47 +0000)]
[notify] remove some useless forward declarations

These cause trouble in the next commit.

13 years agolocalfileenumerator: Improve non-stat'ing code
Benjamin Otte [Tue, 1 Nov 2011 23:31:50 +0000 (00:31 +0100)]
localfileenumerator: Improve non-stat'ing code

We now avoid the per-enumerated-file stat for type and names. We could
improve this further by moving things to the no_stat function, but this
is what the file chooser needs for autocomplete, so I am happy.

13 years agolocalfileinfo: Split out attributes that don't need a stat
Benjamin Otte [Tue, 1 Nov 2011 22:37:35 +0000 (23:37 +0100)]
localfileinfo: Split out attributes that don't need a stat

This will be used in the next commit.

13 years agolocalfileenumerator: Take the type from the readdir() call
Benjamin Otte [Tue, 1 Nov 2011 22:21:42 +0000 (23:21 +0100)]
localfileenumerator: Take the type from the readdir() call

That way, we can avoid stat() calls for all enumerated files in various
cases. In particular in the autocompletion code in the GTK filechooser.

13 years agogio: Add tests for g_file_attribute_matcher_subtract()
Benjamin Otte [Tue, 1 Nov 2011 19:12:27 +0000 (20:12 +0100)]
gio: Add tests for g_file_attribute_matcher_subtract()

13 years agofileinfo: Add g_file_attribute_matcher_subtract()
Benjamin Otte [Tue, 1 Nov 2011 19:11:47 +0000 (20:11 +0100)]
fileinfo: Add g_file_attribute_matcher_subtract()

Added as public API so I can write tests, the use case is local.

13 years agotests: Add tests for fileattributematchers
Benjamin Otte [Tue, 1 Nov 2011 18:12:17 +0000 (19:12 +0100)]
tests: Add tests for fileattributematchers

13 years agofileinfo: Add g_file_attribute_matcher_to_string()
Benjamin Otte [Tue, 1 Nov 2011 18:06:39 +0000 (19:06 +0100)]
fileinfo: Add g_file_attribute_matcher_to_string()

This is to be mainly used for debugging and tests.

13 years agofileinfo: Store namespace::* as 0th attribute
Benjamin Otte [Tue, 1 Nov 2011 16:50:22 +0000 (17:50 +0100)]
fileinfo: Store namespace::* as 0th attribute

This way, we can get_attribute() namespaces. This will be important in
the next commit.

13 years agofileinfo: Change the way attribute matchers are created
Benjamin Otte [Tue, 1 Nov 2011 12:58:09 +0000 (13:58 +0100)]
fileinfo: Change the way attribute matchers are created

We now sort the matchers and remove unnecessary duplicates (like
removing standard:type when we already match standard:*), so that we can
do more complex operations on them easily in later commits.

13 years agofileinfo: Remove optimization for GFileAttributeMatcher
Benjamin Otte [Tue, 1 Nov 2011 11:47:33 +0000 (12:47 +0100)]
fileinfo: Remove optimization for GFileAttributeMatcher

It makes code more complicated, in particular the code I'm about to add.

13 years agoFix G*_TO_POINTER casts on 32 bits
Lucas De Marchi [Thu, 6 Oct 2011 14:18:03 +0000 (11:18 -0300)]
Fix G*_TO_POINTER casts on 32 bits

If we don't do the cast to the proper size in 32 bits, things like below
doesn't work:

uint8_t u = 20;
void *p;

p = GUINT_TO_POINTER(u);

Signed-off-by: Colin Walters <walters@verbum.org>
https://bugzilla.gnome.org/show_bug.cgi?id=661546

13 years agowin32: Make g_get_monotonic_clock lockless
Alexander Larsson [Fri, 4 Nov 2011 15:45:41 +0000 (16:45 +0100)]
win32: Make g_get_monotonic_clock lockless

13 years agowin32: Use timeGetTime as monotonic base
Alexander Larsson [Fri, 4 Nov 2011 14:42:38 +0000 (15:42 +0100)]
win32: Use timeGetTime as monotonic base

This allows apps that need it to increase timer accuracy
using timeBeginPeriod

13 years agowin32: Add a monotonic timer
Alexander Larsson [Fri, 4 Nov 2011 10:27:00 +0000 (11:27 +0100)]
win32: Add a monotonic timer

13 years agoUpdated POTFILES.in
Piotr Drąg [Tue, 15 Nov 2011 19:46:17 +0000 (20:46 +0100)]
Updated POTFILES.in

13 years agoUpdated Spanish translation
Daniel Mustieles [Tue, 15 Nov 2011 16:37:39 +0000 (17:37 +0100)]
Updated Spanish translation

13 years agodocs: Remove tmpl directory
Javier Jardón [Tue, 15 Nov 2011 12:11:55 +0000 (12:11 +0000)]
docs: Remove tmpl directory

We use inline comments now

13 years agoFix 1bit mutex test on platforms with 32 bit aligned pointers
Sjoerd Simons [Sun, 13 Nov 2011 16:49:57 +0000 (17:49 +0100)]
Fix 1bit mutex test on platforms with 32 bit aligned pointers

The 1 bit mutex tests asserts: ((gsize) ptrs) % 8, ==, 0), which fails
when the platform only aligns porters to 32 bits (e.g. S390 and
powerpc).

I'm not sure why this assertion was placed here, but given
that internally g_pointer_bit_trylock uses g_atomic_int_or internally
change the assertion so it only requires the alignment to be a multiple
of sizeof(int)

13 years agoAvoid unused variable warnings in g_ascii_strtod
Matthias Clasen [Tue, 15 Nov 2011 04:30:18 +0000 (23:30 -0500)]
Avoid unused variable warnings in g_ascii_strtod

13 years agoGThread: handle thread names safely
Matthias Clasen [Tue, 15 Nov 2011 03:18:13 +0000 (22:18 -0500)]
GThread: handle thread names safely

Avoid a race condition where the string may be freed before
the new thread got around to using it. Also add a test for
thread names.

https://bugzilla.gnome.org/show_bug.cgi?id=663381

13 years agoSwitch to using template-free docs build
Matthias Clasen [Tue, 15 Nov 2011 02:18:12 +0000 (21:18 -0500)]
Switch to using template-free docs build

All docs are finally inline, so we can do this switch.

13 years agoMove remaining docs inline
Matthias Clasen [Mon, 14 Nov 2011 12:44:52 +0000 (07:44 -0500)]
Move remaining docs inline

This introduces a fake source file just for holding
docs that have no good place elsewhere. Not great, but
better than templates.

13 years agoMove GModule docs inline
Matthias Clasen [Tue, 15 Nov 2011 02:06:30 +0000 (21:06 -0500)]
Move GModule docs inline

13 years agoMove GDate docs inline
Matthias Clasen [Sun, 13 Nov 2011 20:37:56 +0000 (15:37 -0500)]
Move GDate docs inline

13 years agoMove i18n docs inline
Matthias Clasen [Sun, 13 Nov 2011 05:43:50 +0000 (00:43 -0500)]
Move i18n docs inline

13 years agoMove slice and hook docs inline
Matthias Clasen [Sun, 13 Nov 2011 05:26:57 +0000 (00:26 -0500)]
Move slice and hook docs inline

13 years agoGSocketClient: improve error messages on connection failure
Dan Winship [Tue, 15 Nov 2011 00:46:02 +0000 (19:46 -0500)]
GSocketClient: improve error messages on connection failure

Include the hostname (or proxy hostname if it was the connection to
the proxy server that failed) in the GError message when
g_socket_client_connect* fail.

https://bugzilla.gnome.org/show_bug.cgi?id=661266

13 years agoGUnixInput/OutputStream: fix blocking methods to always block
Dan Winship [Mon, 14 Nov 2011 23:27:34 +0000 (18:27 -0500)]
GUnixInput/OutputStream: fix blocking methods to always block

Previously, if you created a GUnixInputStream or GUnixOutputStream
from a non-blocking file descriptor, it might sometimes return
G_IO_ERROR_WOULD_BLOCK from g_input_stream_read/g_output_stream_write,
which is wrong. Fix that. (Use the GPollableInput/OutputStream methods
if you want non-blocking I/O.)

Also, add a test for this to gio/tests/unix-streams.

Also, fix the GError messages to say "Error reading from file
descriptor", etc instead of "Error reading from unix" (which was
presumably from a bad search and replace job).

https://bugzilla.gnome.org/show_bug.cgi?id=626866

13 years agoFix glib/tests/utils for g_parse_debug_string() changes
Dan Winship [Mon, 14 Nov 2011 18:55:14 +0000 (13:55 -0500)]
Fix glib/tests/utils for g_parse_debug_string() changes

13 years agoadd GNetworkMonitor, for... monitoring the network
Dan Winship [Sun, 12 Jun 2011 19:59:36 +0000 (15:59 -0400)]
add GNetworkMonitor, for... monitoring the network

Add GNetworkMonitor and its associated extension point, provide a base
implementation that always claims the network is available, and a
netlink-based implementation built on top of that that actually tracks
the network state.

https://bugzilla.gnome.org/show_bug.cgi?id=620932

13 years agoGInetAddressMask: new type for internet address range matching
Dan Winship [Sat, 1 Oct 2011 12:31:54 +0000 (08:31 -0400)]
GInetAddressMask: new type for internet address range matching

Eg, for matching a GInetAddress to a range like "10.0.0.0/8" or
"fe80::/10"

https://bugzilla.gnome.org/show_bug.cgi?id=620932

13 years agogunixinput/outputstream: fix docs
Dan Winship [Sun, 13 Nov 2011 21:39:45 +0000 (16:39 -0500)]
gunixinput/outputstream: fix docs

Don't try to link to the docs of undocumented private classes

13 years agoGUnixInputStream, GUnixOutputStream: support ordinary files better
Dan Winship [Tue, 27 Apr 2010 20:54:18 +0000 (16:54 -0400)]
GUnixInputStream, GUnixOutputStream: support ordinary files better

If the fd is not a pipe or socket, fall back to using threads to do
async I/O rather than poll, since poll doesn't work the way you want
for ordinary files.

https://bugzilla.gnome.org/show_bug.cgi?id=606913

13 years agoSome more docs reshuffling
Matthias Clasen [Sun, 13 Nov 2011 03:52:24 +0000 (22:52 -0500)]
Some more docs reshuffling

13 years agoMore consistent doc formatting
Matthias Clasen [Sun, 13 Nov 2011 02:54:42 +0000 (21:54 -0500)]
More consistent doc formatting

Move some things around, make capitalization of short descriptions
more consistent.

13 years agoCleanups to debug env var handling
Matthias Clasen [Sun, 13 Nov 2011 00:13:44 +0000 (19:13 -0500)]
Cleanups to debug env var handling

13 years agoParse G_DEBUG only once
Matthias Clasen [Sat, 12 Nov 2011 23:36:52 +0000 (18:36 -0500)]
Parse G_DEBUG only once

13 years agoglib-unix.h: Add G_BEGIN/END_DECLS
Matthias Clasen [Sat, 12 Nov 2011 15:26:31 +0000 (10:26 -0500)]
glib-unix.h: Add G_BEGIN/END_DECLS

The omission was pointed out in bug 663880.

13 years agogstrfuncs.c: Fix variable declaration
Chun-wei Fan [Sat, 12 Nov 2011 02:38:31 +0000 (10:38 +0800)]
gstrfuncs.c: Fix variable declaration

Declare variables at top of block to stop C89 compilers from complaining.

13 years agoGDBusWorker: tolerate read errors while closing
Simon McVittie [Fri, 11 Nov 2011 14:41:50 +0000 (14:41 +0000)]
GDBusWorker: tolerate read errors while closing

My previous fix for GNOME#662100 was incomplete: it seems that with some
timings, the stream can be closed with an async read in-flight. This
can make the read fail immediately with G_IO_ERROR_CLOSED instead of
becoming cancelled.

This happens reliably on an embedded device, and rarely on my laptop;
repeating the test 100 times in quick succession reliably reproduces
the bug on my laptop.

It seems as though what we really want is to ignore read errors, once
we've established that we want to close the connection anyway - this
means that after asking to close, you're immune to exit-on-close,
which seems like a good rule.

An additional subtlety is that continuing to read after we know we
want to close is still required, otherwise we'll never emit ::closed.

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=662100
Bug-NB: NB#287088
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Colin Walters <walters@verbum.org>
13 years agoGDBusWorker: debug on read errors if transport debugging is enabled
Simon McVittie [Thu, 10 Nov 2011 19:06:43 +0000 (19:06 +0000)]
GDBusWorker: debug on read errors if transport debugging is enabled

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=662100
Bug-NB: NB#287088
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Colin Walters <walters@verbum.org>
13 years agogdbus-exit-on-close test: don't leak a variant
Simon McVittie [Thu, 3 Nov 2011 18:33:08 +0000 (18:33 +0000)]
gdbus-exit-on-close test: don't leak a variant

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=662100
Bug-NB: NB#287088
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Colin Walters <walters@verbum.org>
13 years agogdbus-exit-on-close test: optionally be more verbose for debugging
Simon McVittie [Thu, 3 Nov 2011 18:32:53 +0000 (18:32 +0000)]
gdbus-exit-on-close test: optionally be more verbose for debugging

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=662100
Bug-NB: NB#287088
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Colin Walters <walters@verbum.org>
13 years agogio: more implementations of GFileDescriptorBased
Dan Winship [Sat, 3 Sep 2011 13:03:17 +0000 (09:03 -0400)]
gio: more implementations of GFileDescriptorBased

Implement GFileDescriptorBased in GSocket{Input,Output}Stream and
GUnix{Input,Output}Stream.

https://bugzilla.gnome.org/show_bug.cgi?id=616852

13 years agoNeed to include xlocale.h on Darwin for strtod_l
Kristian Rietveld [Wed, 9 Nov 2011 07:32:33 +0000 (08:32 +0100)]
Need to include xlocale.h on Darwin for strtod_l

This fixes the build of gstrfuncs.c on Darwin.

13 years agoAdd G_SIGNAL_DEPRECATED
Marc-André Lureau [Mon, 7 Nov 2011 19:04:15 +0000 (20:04 +0100)]
Add G_SIGNAL_DEPRECATED

Similar to G_PARAM_DEPRECATED. It will warn only for users of the
signals, so a signal can still be emited without warning, for
compatibility reasons.

Apparently, there is no way user flags could have been used before,
so that shouldn't break anyone.

https://bugzilla.gnome.org/show_bug.cgi?id=663581

13 years agogio/tests/proxy: don't use deprecated functions
Dan Winship [Mon, 7 Nov 2011 18:50:00 +0000 (13:50 -0500)]
gio/tests/proxy: don't use deprecated functions

13 years agoFix g_variant_get_gtype() deprecation message
Dan Winship [Mon, 7 Nov 2011 18:48:30 +0000 (13:48 -0500)]
Fix g_variant_get_gtype() deprecation message

Instead of:

warning: ‘g_variant_get_gtype’ is deprecated (declared at ../../gobject/glib-types.h:242): Use '((GType) ((21) << (2)))' instead [-Wdeprecated-declarations]

show:

warning: ‘g_variant_get_gtype’ is deprecated (declared at ../../gobject/glib-types.h:242): Use ''G_VARIANT_GET_TYPE'' instead [-Wdeprecated-declarations]

Also, document the macro-expansion problem in the
G_GNUC_DEPRECATED_FOR docs

13 years agodocs: Improve g_variant_loop() docs
Benjamin Otte [Sun, 16 Oct 2011 19:58:29 +0000 (12:58 -0700)]
docs: Improve g_variant_loop() docs

Typo fix and changing "is not recommended" to the nice version of "don't
do this", aka "causes undefined behavior".

13 years agodocs: Tell which function to use to avoid errors.
Benjamin Otte [Mon, 25 Jul 2011 11:08:59 +0000 (13:08 +0200)]
docs: Tell which function to use to avoid errors.

13 years agoUpdate VS property sheets
Chun-wei Fan [Mon, 7 Nov 2011 06:14:45 +0000 (14:14 +0800)]
Update VS property sheets

gmain.h was added into the list of deprecated headers, so "install"
that as weel.

13 years agoAdd a test for CDATA in gschema.xml files
Matthias Clasen [Mon, 7 Nov 2011 05:14:28 +0000 (00:14 -0500)]
Add a test for CDATA in gschema.xml files

13 years agoAccept CDATA in gschema.xml files
Matthias Clasen [Mon, 7 Nov 2011 05:13:56 +0000 (00:13 -0500)]
Accept CDATA in gschema.xml files

13 years agowin32: Remove some debug spew on startup
Alexander Larsson [Thu, 3 Nov 2011 16:27:08 +0000 (17:27 +0100)]
win32: Remove some debug spew on startup

This is not needed anymore and was causing problems for pkg-config.

13 years agoMake the default log handler more useful
Matthias Clasen [Thu, 3 Nov 2011 05:48:54 +0000 (01:48 -0400)]
Make the default log handler more useful

We make the default log handler only print default and informational
messages if the log domain is explicitly requested.

https://bugzilla.gnome.org/show_bug.cgi?id=661926

13 years agoMove old deprecated gmain api into deprecated/
Matthias Clasen [Thu, 3 Nov 2011 04:39:31 +0000 (00:39 -0400)]
Move old deprecated gmain api into deprecated/

13 years agoRemove single-include guards in deprecated/
Matthias Clasen [Thu, 3 Nov 2011 04:30:56 +0000 (00:30 -0400)]
Remove single-include guards in deprecated/

Not allowed to include these headers single anymore, either.

13 years agoGIO: Don't use G_DISABLE_DEPRECATED for functions
Matthias Clasen [Thu, 3 Nov 2011 04:16:41 +0000 (00:16 -0400)]
GIO: Don't use G_DISABLE_DEPRECATED for functions

13 years agoGObject: Don't use G_DISABLE_DEPRECATED for functions
Matthias Clasen [Thu, 3 Nov 2011 04:09:32 +0000 (00:09 -0400)]
GObject: Don't use G_DISABLE_DEPRECATED for functions

13 years agoUpdate deprecation docs
Matthias Clasen [Thu, 3 Nov 2011 04:05:29 +0000 (00:05 -0400)]
Update deprecation docs

13 years agoDon't use G_DISABLE_DEPRECATED guards around deprecated functions
Matthias Clasen [Thu, 3 Nov 2011 04:04:15 +0000 (00:04 -0400)]
Don't use G_DISABLE_DEPRECATED guards around deprecated functions

I'm leaving the old-style deprecation guards in place around
deprecated macros, enumeration values, etc, for now.

13 years agoglib-mkenums: add --identifier-prefix and --symbol-prefix args
Dan Winship [Fri, 28 Oct 2011 19:38:52 +0000 (15:38 -0400)]
glib-mkenums: add --identifier-prefix and --symbol-prefix args

Allow passing --identifier-prefix and --symbol-prefix to glib-mkenums,
with the same meanings as in g-ir-scanner, to allow fixing up the enum
name parsing globally rather than needing to add a /<* *>/ override to
each enum.

https://bugzilla.gnome.org/show_bug.cgi?id=661797

13 years agodocs: don't refer to a deprecated glib-mkenums flag
Dan Winship [Fri, 28 Oct 2011 19:38:27 +0000 (15:38 -0400)]
docs: don't refer to a deprecated glib-mkenums flag

Refer to the new flag instead

https://bugzilla.gnome.org/show_bug.cgi?id=661797

13 years agoglib-mkenums.in: fix @ENUMPREFIX@ with /*< underscore_name=... >*/
Dan Winship [Fri, 14 Oct 2011 18:32:11 +0000 (14:32 -0400)]
glib-mkenums.in: fix @ENUMPREFIX@ with /*< underscore_name=... >*/

For enums defined with an underscore_name override, @ENUMPREFIX@
would include a trailing "_". Fix.

https://bugzilla.gnome.org/show_bug.cgi?id=661797

13 years agogdbus: avoid warning when finalizing a GDBusObjectManagerClient
Aleksander Morgado [Thu, 27 Oct 2011 14:54:02 +0000 (16:54 +0200)]
gdbus: avoid warning when finalizing a GDBusObjectManagerClient

If the GDBusObjectManagerClient doesn't get a name owner during its lifetime,
`on_control_proxy_g_signal' will never be connected to any signal, so we
shouldn't dump any warning in that case.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=662858

13 years agog_bus_own_name: fix race when unowning a name immediately after owning it
David Zeuthen [Thu, 27 Oct 2011 14:30:58 +0000 (10:30 -0400)]
g_bus_own_name: fix race when unowning a name immediately after owning it

... and also add a test to verify that the fix works.

https://bugzilla.gnome.org/show_bug.cgi?id=662808

Signed-off-by: David Zeuthen <davidz@redhat.com>
13 years agoRevert "gthread: G_STATIC_MUTEX_INIT: Fix this for the non-win32 case."
Ryan Lortie [Wed, 26 Oct 2011 19:10:33 +0000 (15:10 -0400)]
Revert "gthread: G_STATIC_MUTEX_INIT: Fix this for the non-win32 case."

This reverts commit 52fd106724aa79ad57ecaa7ad9d340f8c89da06d.

13 years agogthread: G_STATIC_MUTEX_INIT: Fix this for the non-win32 case.
Murray Cumming [Wed, 26 Oct 2011 09:38:13 +0000 (11:38 +0200)]
gthread: G_STATIC_MUTEX_INIT: Fix this for the non-win32 case.

Also initialize the unused member. This was correct before but
was broken when the ifndef was moved inside the GStaticMutex
struct:
http://git.gnome.org/browse/glib/commit/glib/gthread.h?id=24652730a9faaedb19b9e90024077eb7f75a6907
This avoids a compiler warning.

13 years agoAdd TAGS files to .gitignore
Damien Lespinau [Tue, 25 Oct 2011 02:25:59 +0000 (22:25 -0400)]
Add TAGS files to .gitignore

"make tags" is the best way to produce tags in an automake project and
yet very few people are using it so the tags files it generates are never
added to the .gitignore files.

Add TAGS files to .gitignore then.

https://bugzilla.gnome.org/show_bug.cgi?id=615424

13 years agoFix some build issues on OpenBSD
Robert Nagy [Tue, 25 Oct 2011 01:18:48 +0000 (21:18 -0400)]
Fix some build issues on OpenBSD

Adapt to some OpenBSD header quirks.
Patch by Robert Nagy.

https://bugzilla.gnome.org/show_bug.cgi?id=661386

13 years agoFix document generation in out of tree builds
Sjoerd Simons [Fri, 21 Oct 2011 13:40:56 +0000 (15:40 +0200)]
Fix document generation in out of tree builds

https://bugzilla.gnome.org/show_bug.cgi?id=662390

13 years agog_parse_debug_string: invert flags given besides "all"
Will Thompson [Thu, 20 Oct 2011 15:07:03 +0000 (16:07 +0100)]
g_parse_debug_string: invert flags given besides "all"

Any flags specified as well as "all" are subtracted from the result,
allowing the user to specify FOO_DEBUG="all,bar,baz" to mean "give me
debugging information for everything except bar and baz".

https://bugzilla.gnome.org/show_bug.cgi?id=642452

13 years agoTest g_parse_debug_string ("all")
Will Thompson [Wed, 16 Feb 2011 10:48:22 +0000 (10:48 +0000)]
Test g_parse_debug_string ("all")

https://bugzilla.gnome.org/show_bug.cgi?id=642452

13 years agoFix 'make check' with clang
Matthias Clasen [Mon, 24 Oct 2011 17:43:11 +0000 (13:43 -0400)]
Fix 'make check' with clang

Patch by Elias Pipping

https://bugzilla.gnome.org/show_bug.cgi?id=662491

13 years agoFix g_hash_table_iter_replace
Matthias Clasen [Mon, 24 Oct 2011 17:25:15 +0000 (13:25 -0400)]
Fix g_hash_table_iter_replace

When reusing an existing key 'internally', we must avoid calling
the key_destroy function on the old key.

https://bugzilla.gnome.org/show_bug.cgi?id=662544

13 years agoGDBusConnection: document that this object is (meant to be) thread-safe
Simon McVittie [Fri, 21 Oct 2011 10:42:10 +0000 (11:42 +0100)]
GDBusConnection: document that this object is (meant to be) thread-safe

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=661992
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: David Zeuthen <davidz@redhat.com>
13 years agoGDBusConnection: make the closed flag atomic (but still lock to write)
Simon McVittie [Fri, 21 Oct 2011 14:46:00 +0000 (15:46 +0100)]
GDBusConnection: make the closed flag atomic (but still lock to write)

Strictly speaking, neither of the two uses that aren't under the lock
*needs* to be atomic, but it seems better to be obviously correct (and
we save another 4 bytes of struct).

One of these uses is in g_dbus_connection_is_closed(), any use of which
is inherently a race condition anyway.

The other is g_dbus_connection_flush_sync, which as far as I can tell
just needs a best-effort check, to not waste effort on a connection that
has been closed for a while (but I could be wrong).

I removed the check for the closed flag altogether in
g_dbus_connection_send_message_with_reply_unlocked, because it turns out
to be redundant with one in g_dbus_connection_send_message_unlocked,
which is called immediately after.

g_dbus_connection_close_sync held the lock to check the closed flag,
which is no longer needed.

As far as I can tell, the only reason why the lock is still desirable
when setting the closed flag is so that remove_match_rule can't fail
by racing with close notification from the worker thread - but
on_worker_closed needs to hold the lock anyway, to deal with other
data structures, so there's no point in trying to eliminate the
requirement to hold the lock.

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=661992
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: David Zeuthen <davidz@redhat.com>
13 years agoGDBusConnection: document which properties are protected by @lock
Simon McVittie [Fri, 21 Oct 2011 15:24:29 +0000 (16:24 +0100)]
GDBusConnection: document which properties are protected by @lock

Also, a few that don't need to be.

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=661992
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: David Zeuthen <davidz@redhat.com>
13 years agoGDBusConnection: access the exit-on-close flag atomically
Simon McVittie [Mon, 17 Oct 2011 18:27:23 +0000 (19:27 +0100)]
GDBusConnection: access the exit-on-close flag atomically

This isn't strictly necessary, because in every location where it's
checked, if the reading thread misses an update from another thread,
it's indistinguishable from the reading thread having been scheduled
before the writing thread, which is an unavoidable race condition that
callers need to cope with anyway. On the other hand, merging exit_on_close
into atomic_flags gives the least astonishing semantics to library users
and saves 4 bytes of struct, and if you're accessing exit-on-close often
enough for it to be a performance concern, you're probably doing it wrong.

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=661992
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: David Zeuthen <davidz@redhat.com>
13 years agoAnnotate GDBusConnection private functions with thread/lock status
Simon McVittie [Mon, 17 Oct 2011 12:27:03 +0000 (13:27 +0100)]
Annotate GDBusConnection private functions with thread/lock status

The thread shared between all GDBusWorker instances was variously called
the "worker thread" or "message handler thread", which I mostly changed to
"the GDBusWorker thread" to avoid ambiguity.

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=661992
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: David Zeuthen <davidz@redhat.com>