platform/upstream/glib.git
12 years agoe G_SOURCE_CONTINUE/REMOVE internally
Matthias Clasen [Wed, 25 Jan 2012 19:18:34 +0000 (14:18 -0500)]
e G_SOURCE_CONTINUE/REMOVE internally

Now that we have these macros, we should use them.
This commit covers everything in gio/.

12 years agoUse G_SOURCE_CONTINUE/REMOVE internally
Matthias Clasen [Wed, 25 Jan 2012 13:53:05 +0000 (08:53 -0500)]
Use G_SOURCE_CONTINUE/REMOVE internally

Now that we have these macros, we should use them.
This commit covers everything in glib/.

12 years agoGHash: add note about randomness in hashing
Ryan Lortie [Wed, 25 Jan 2012 15:09:09 +0000 (10:09 -0500)]
GHash: add note about randomness in hashing

To clarify our statements about being robust against outside attackers
degrading the performance of your hashtable.

12 years agoresources: tool: Fix a compiler warning
Christian Persch [Tue, 24 Jan 2012 13:18:12 +0000 (14:18 +0100)]
resources: tool: Fix a compiler warning

Use G_GSIZE_FORMAT.

12 years agoresources: compiler: Add autoselected output format
Christian Persch [Mon, 23 Jan 2012 21:51:44 +0000 (22:51 +0100)]
resources: compiler: Add autoselected output format

This allows simplifying the make rules.

Bug #668539.

12 years agoresources: compiler: Add dependency generator
Christian Persch [Mon, 23 Jan 2012 19:42:20 +0000 (20:42 +0100)]
resources: compiler: Add dependency generator

Bug #668532.

12 years agoresources: compiler: Allow stripping blanks from xml data
Christian Persch [Sat, 14 Jan 2012 21:34:15 +0000 (22:34 +0100)]
resources: compiler: Allow stripping blanks from xml data

It's hardly useful to bloat the resource data with blanks intended only
for human readability, so add a preprocessing option that uses xmllint --noblanks
to strip these.

Bug #667929.

12 years agoGKeyFile: make list separators effective again
Matthias Clasen [Wed, 25 Jan 2012 12:32:39 +0000 (07:32 -0500)]
GKeyFile: make list separators effective again

We must preserve the list_separator across clear;init
sequences.

12 years agoFix the new test
Matthias Clasen [Wed, 25 Jan 2012 12:32:20 +0000 (07:32 -0500)]
Fix the new test

12 years agoAdd a test for g_key_file_set_list_separator
Matthias Clasen [Wed, 25 Jan 2012 12:26:58 +0000 (07:26 -0500)]
Add a test for g_key_file_set_list_separator

A recent 'harmless' fix broke it...

12 years agoGArray: initialize the clear_func pointer
Guillaume Desmottes [Wed, 25 Jan 2012 11:03:14 +0000 (12:03 +0100)]
GArray: initialize the clear_func pointer

https://bugzilla.gnome.org/show_bug.cgi?id=668650

12 years agogresource-tool: fix the build
Xan Lopez [Wed, 25 Jan 2012 09:28:27 +0000 (10:28 +0100)]
gresource-tool: fix the build

12 years agoMacro to insert a new GNode "after" the given sibling.
Ravi Sankar Guntur [Sat, 21 Jan 2012 16:47:18 +0000 (22:17 +0530)]
Macro to insert a new GNode "after" the given sibling.

https://bugzilla.gnome.org/show_bug.cgi?id=626258

Signed-off-by: Ravi Sankar Guntur <ravi.g@samsung.com>
12 years agoHonor the glib_cv_g_atomic_lock_free env var in configure
Enrique Ocaña González [Tue, 24 Jan 2012 12:24:51 +0000 (13:24 +0100)]
Honor the glib_cv_g_atomic_lock_free env var in configure

This fixes bug: https://bugzilla.gnome.org/show_bug.cgi?id=668572

12 years agoDeprecate GValueArray
Emmanuele Bassi [Wed, 4 Jan 2012 09:31:12 +0000 (09:31 +0000)]
Deprecate GValueArray

The GValueArray type was added in a time, during the Jurassic era or so,
when GArray did not have a representable GType. The GValueArray API has
various issues as well:

  - it doesn't match the other GLib array types;
  - it is not reference counted;
  - the structure is fully exposed on the stack, so it cannot be
    extended to add reference counting;
  - it cannot be forcibly resized.

The nice thing is that now we have a GArray type that can replace in
full GValueArray, so we can deprecate the latter, and reduce the
complexity in GLib, application code, and bindings.

https://bugzilla.gnome.org/show_bug.cgi?id=667228

12 years agoarray: Add a clear function
Emmanuele Bassi [Wed, 4 Jan 2012 07:41:47 +0000 (07:41 +0000)]
array: Add a clear function

Like GPtrArray has a "free function" that can be used to free memory
associated to each pointer in the array, GArray would benefit from
having a "clear function" that can be used to clear the content of
each element of the array when it's removed, or when the entire array
is freed.

https://bugzilla.gnome.org/show_bug.cgi?id=667243

12 years agoGHashTable: Add a note about hash collisions
Matthias Clasen [Wed, 25 Jan 2012 02:11:13 +0000 (21:11 -0500)]
GHashTable: Add a note about hash collisions

12 years agogresource: Initialize i18n
Matthias Clasen [Tue, 24 Jan 2012 22:37:09 +0000 (17:37 -0500)]
gresource: Initialize i18n

...otherwise translations don't work. Pointed out by Kjartan Maraas.

https://bugzilla.gnome.org/show_bug.cgi?id=668561

12 years agokeyfile tests: Add testcase for bug #634232
Ryan Lortie [Tue, 24 Jan 2012 21:59:59 +0000 (16:59 -0500)]
keyfile tests: Add testcase for bug #634232

Add a testcase based on one by Alfredo Dal'Ava Júnior (the original
reporter of the bug).

https://bugzilla.gnome.org/show_bug.cgi?id=634232

12 years agoGKeyFile: remove approximate_size optimisation
Ryan Lortie [Tue, 24 Jan 2012 21:45:49 +0000 (16:45 -0500)]
GKeyFile: remove approximate_size optimisation

This is micro-optimisation of the worst kind and it's causing obscure
bugs in edge cases.  Better just remove this entirely.

https://bugzilla.gnome.org/show_bug.cgi?id=634232

12 years agoGSocket: fix complile on platforms without source-specific multicast
Dan Winship [Tue, 24 Jan 2012 13:28:33 +0000 (08:28 -0500)]
GSocket: fix complile on platforms without source-specific multicast

Some platforms don't have the source-specific multicast sockopts, and
so would fail to compile. Fix that (and return an error if the caller
tries to use source-specific). Also clarify the docs a bit.

https://bugzilla.gnome.org/show_bug.cgi?id=668468

12 years agoWrite list-of value correctly
Christian Persch [Sun, 30 Jan 2011 19:45:52 +0000 (20:45 +0100)]
Write list-of value correctly

Fixes a c&p typo. Bug #639099.

12 years agoUpdated Norwegian bokmål translation
Kjartan Maraas [Tue, 24 Jan 2012 09:53:35 +0000 (10:53 +0100)]
Updated Norwegian bokmål translation

12 years agoGMenuMarkup: call g_variant_ref_sink()
Matthias Clasen [Tue, 24 Jan 2012 00:21:22 +0000 (19:21 -0500)]
GMenuMarkup: call g_variant_ref_sink()

Watch out for them floating variants !
They are easily consumed...

12 years agoUpdated Galician translations
Fran Diéguez [Mon, 23 Jan 2012 23:50:50 +0000 (00:50 +0100)]
Updated Galician translations

12 years agoUpdated POTFILES.in
Piotr Drąg [Mon, 23 Jan 2012 19:15:30 +0000 (20:15 +0100)]
Updated POTFILES.in

12 years agoAdd docs for gresource
Matthias Clasen [Mon, 23 Jan 2012 05:23:18 +0000 (00:23 -0500)]
Add docs for gresource

12 years agoAdd a resource tool
Matthias Clasen [Sun, 22 Jan 2012 07:39:11 +0000 (02:39 -0500)]
Add a resource tool

This lets you poke at resources in elf files and
standalone resource bundles. So far, only listing
and extracting resources is supported. The support
for elf files requires libelf.

12 years agog_date_time_format: fix output in non-UTF-8 locales
Dan Winship [Thu, 19 Jan 2012 15:39:57 +0000 (10:39 -0500)]
g_date_time_format: fix output in non-UTF-8 locales

In non-UTF-8 locales, the translations and nl_langinfo() return values
must be converted to UTF-8 before being returned to the caller.
Likewise, when making a recursive call to expand a format like '%x',
the format string must first be converted to UTF-8.

https://bugzilla.gnome.org/show_bug.cgi?id=668250

12 years agobump rev
Matthias Clasen [Fri, 20 Jan 2012 20:42:44 +0000 (15:42 -0500)]
bump rev

12 years ago2.31.12 2.31.12
Matthias Clasen [Fri, 20 Jan 2012 20:10:29 +0000 (15:10 -0500)]
2.31.12

12 years agoUpdates
Matthias Clasen [Fri, 20 Jan 2012 20:10:00 +0000 (15:10 -0500)]
Updates

12 years agoSilence some compiler warnings
Matthias Clasen [Fri, 20 Jan 2012 13:08:06 +0000 (08:08 -0500)]
Silence some compiler warnings

12 years agoSilence a compiler warning
Matthias Clasen [Fri, 20 Jan 2012 13:07:52 +0000 (08:07 -0500)]
Silence a compiler warning

12 years agoFix a refcounting error
Matthias Clasen [Fri, 20 Jan 2012 13:02:26 +0000 (08:02 -0500)]
Fix a refcounting error

'new' is created floating, therefore it is consumed by
g_settings_set, and unreffing it after that call is not right.

12 years agoNote G_MESSAGES_DEBUG in README
Ryan Lortie [Fri, 20 Jan 2012 03:13:51 +0000 (22:13 -0500)]
Note G_MESSAGES_DEBUG in README

12 years agotests for GSettings action binding
Ryan Lortie [Thu, 19 Jan 2012 15:49:06 +0000 (10:49 -0500)]
tests for GSettings action binding

https://bugzilla.gnome.org/show_bug.cgi?id=668279

12 years agoadd a way to create a GAction from GSettings
Ryan Lortie [Thu, 19 Jan 2012 14:40:24 +0000 (09:40 -0500)]
add a way to create a GAction from GSettings

g_settings_create_action() will create a GAction for the named key,
allowing it to be added to the action group of the application (so that
the setting can be directly manipulated from menus, for example).

https://bugzilla.gnome.org/show_bug.cgi?id=668279

12 years agogsignal: add g_signal_handlers_disconnect_by_data
Ryan Lortie [Thu, 19 Jan 2012 14:01:08 +0000 (09:01 -0500)]
gsignal: add g_signal_handlers_disconnect_by_data

Similar to g_signal_handlers_disconnect_by_func() but disconnects all
functions that use the given user_data.

https://bugzilla.gnome.org/show_bug.cgi?id=668269

12 years agogioenums.h: Remove a trailing comma.
Murray Cumming [Thu, 19 Jan 2012 15:18:05 +0000 (16:18 +0100)]
gioenums.h: Remove a trailing comma.

12 years agoGSettingsSchema: ignore empty schema caches
Ryan Lortie [Thu, 19 Jan 2012 03:15:30 +0000 (22:15 -0500)]
GSettingsSchema: ignore empty schema caches

glib-compile-schemas used to generate these.  They're harmless and they
mean that no schemas are installed in a particular directory, so just
ignore them.

https://bugzilla.gnome.org/show_bug.cgi?id=656301

12 years agoresources: Plug mem leaks in resource compiler
Christian Persch [Sat, 14 Jan 2012 20:28:29 +0000 (21:28 +0100)]
resources: Plug mem leaks in resource compiler

==13007== 173 bytes in 1 blocks are definitely lost in loss record 90 of 106
==13007==    at 0x402AD89: malloc (vg_replace_malloc.c:236)
==13007==    by 0x407DDBA: standard_malloc (gmem.c:85)
==13007==    by 0x407E318: g_try_malloc (gmem.c:271)
==13007==    by 0x40654DE: g_file_get_contents (gfileutils.c:756)
==13007==    by 0x804A531: main (glib-compile-resources.c:580)

==13007== 521 (56 direct, 465 indirect) bytes in 1 blocks are definitely lost in loss record 100 of 106
==13007==    at 0x402AD89: malloc (vg_replace_malloc.c:236)
==13007==    by 0x407DDBA: standard_malloc (gmem.c:85)
==13007==    by 0x407E160: g_malloc (gmem.c:159)
==13007==    by 0x4091D8D: g_slice_alloc (gslice.c:1003)
==13007==    by 0x40674A1: g_hash_table_new_full (ghash.c:676)
==13007==    by 0x804B252: gvdb_hash_table_new (gvdb-builder.c:76)
==13007==    by 0x43C66B2: (below main) (libc-start.c:226)

12 years agoGApplication: drop support for appmenu/menubars
Ryan Lortie [Mon, 16 Jan 2012 19:47:08 +0000 (14:47 -0500)]
GApplication: drop support for appmenu/menubars

This has been moved over to GtkApplication now.

https://bugzilla.gnome.org/show_bug.cgi?id=668118

12 years agoupdate .gitignores
Dan Winship [Wed, 18 Jan 2012 18:28:00 +0000 (13:28 -0500)]
update .gitignores

12 years agoFix glib-compile-resources usage when cross-compiling
Dan Winship [Wed, 18 Jan 2012 17:05:07 +0000 (12:05 -0500)]
Fix glib-compile-resources usage when cross-compiling

Copy the behavior of glib-genmarshal: require an installed copy of it,
and use that rather than the built copy.

12 years agoGDBusConnection: note that exit-on-close is sometimes TRUE
Will Thompson [Wed, 18 Jan 2012 10:14:37 +0000 (10:14 +0000)]
GDBusConnection: note that exit-on-close is sometimes TRUE

g_bus_get_finish() and g_bus_get_sync() both document that the returned
object will usually have exit-on-close set to TRUE, but the property's
documentation specified that its default is FALSE. While that's
technically true from a GObject perspective, it's not accurate from the
API user's perspective.

https://bugzilla.gnome.org/show_bug.cgi?id=668163

Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: David Zeuthen <zeuthen@gmail.com>
12 years agoGUnixConnection: add async variant for send/receive_credentials()
Xavier Claessens [Mon, 16 Jan 2012 13:51:19 +0000 (14:51 +0100)]
GUnixConnection: add async variant for send/receive_credentials()

https://bugzilla.gnome.org/show_bug.cgi?id=629503

12 years agog_base64_encode_step: clarify break_lines behavior a bit
Dan Winship [Wed, 18 Jan 2012 14:02:21 +0000 (09:02 -0500)]
g_base64_encode_step: clarify break_lines behavior a bit

break_lines uses LFs, not CRLFs like you might expect (since it's
designed for email-related use), but we can't change that now since
the caller has to allocate the output buffer and so the
number-of-bytes-output is part of the ABI. So, just document that.

https://bugzilla.gnome.org/show_bug.cgi?id=668158

12 years agogsocket: make this compile on Windows again
Dan Winship [Wed, 18 Jan 2012 00:46:34 +0000 (19:46 -0500)]
gsocket: make this compile on Windows again

https://bugzilla.gnome.org/show_bug.cgi?id=668071

12 years agoUpdated Norwegian bokmål translation
Kjartan Maraas [Tue, 17 Jan 2012 16:11:43 +0000 (17:11 +0100)]
Updated Norwegian bokmål translation

12 years agoFix broken build with -Wundefined
Kjartan Maraas [Tue, 17 Jan 2012 16:10:19 +0000 (17:10 +0100)]
Fix broken build with -Wundefined

12 years agoresources: Skip initial underscores in resource section names
Alexander Larsson [Tue, 17 Jan 2012 11:32:37 +0000 (12:32 +0100)]
resources: Skip initial underscores in resource section names

Initial underscores are used in c identifier to make them private,
for instance in Gtk+. However, we don't want to have this in the
resource section name, that just looks ugly.

12 years agogsocket: Use alternate IPv6 multicast group sockopt names
Dan Winship [Mon, 16 Jan 2012 22:15:06 +0000 (17:15 -0500)]
gsocket: Use alternate IPv6 multicast group sockopt names

Apparently IPV6_JOIN_GROUP and IPV6_LEAVE_GROUP are more portable than
IPV6_ADD_MEMBERSHIP and IPV6_DROP_MEMBERSHIP. (Windows and Linux have
both, but OS X only has the latter.)

12 years agoBack to the odd
Matthias Clasen [Mon, 16 Jan 2012 19:46:26 +0000 (14:46 -0500)]
Back to the odd

12 years ago2.31.10 2.31.10
Matthias Clasen [Mon, 16 Jan 2012 19:00:18 +0000 (14:00 -0500)]
2.31.10

12 years agoUpdates
Matthias Clasen [Mon, 16 Jan 2012 16:50:07 +0000 (11:50 -0500)]
Updates

12 years agoGInetSocketAddress: add IPv6 flowinfo and scope_id fields
Dan Winship [Thu, 29 Dec 2011 14:42:25 +0000 (09:42 -0500)]
GInetSocketAddress: add IPv6 flowinfo and scope_id fields

struct sin6_addr has two additional fields that struct sin_addr
doesn't. Add support for those to GInetSocketAddress, and make sure
they don't get lost when converting between glib and native types.

https://bugzilla.gnome.org/show_bug.cgi?id=635554

12 years agoginetaddress.c: fix an incorrect enum name in the docs
Dan Winship [Thu, 29 Dec 2011 14:18:53 +0000 (09:18 -0500)]
ginetaddress.c: fix an incorrect enum name in the docs

12 years agoGSocket: Use correct type for SO_BROADCAST parameter
Sebastian Dröge [Mon, 16 Jan 2012 18:22:28 +0000 (19:22 +0100)]
GSocket: Use correct type for SO_BROADCAST parameter

12 years agoGSocket: Reset the timeout in the GSocket GSource after it was triggered
Sebastian Dröge [Mon, 16 Jan 2012 14:49:50 +0000 (15:49 +0100)]
GSocket: Reset the timeout in the GSocket GSource after it was triggered

https://bugzilla.gnome.org/show_bug.cgi?id=667989

12 years agoGSocket: Add function to get the currently available bytes for reading
Ole André Vadla Ravnås [Mon, 16 Jan 2012 13:00:49 +0000 (14:00 +0100)]
GSocket: Add function to get the currently available bytes for reading

https://bugzilla.gnome.org/show_bug.cgi?id=668009

12 years agoGSocket: Add function to set/get the broadcast setting on a socket
Sebastian Dröge [Fri, 13 Jan 2012 12:01:35 +0000 (13:01 +0100)]
GSocket: Add function to set/get the broadcast setting on a socket

https://bugzilla.gnome.org/show_bug.cgi?id=623187

12 years agoGSocket: Add function for setting unicast TTL
Sebastian Dröge [Fri, 13 Jan 2012 11:48:02 +0000 (12:48 +0100)]
GSocket: Add function for setting unicast TTL

12 years agoGSocket: Add support for source-specific multicast (RFC 4604)
Sebastian Dröge [Fri, 13 Jan 2012 11:53:50 +0000 (12:53 +0100)]
GSocket: Add support for source-specific multicast (RFC 4604)

12 years agoGSocket: Add possibility to join a multicast group only on a specific interface
Sebastian Dröge [Fri, 13 Jan 2012 11:37:31 +0000 (12:37 +0100)]
GSocket: Add possibility to join a multicast group only on a specific interface

12 years agoGSocket: Add multicast-related functions
Dan Winship [Thu, 29 Dec 2011 16:01:23 +0000 (11:01 -0500)]
GSocket: Add multicast-related functions

Add APIs for sending and receiving multicast datagrams with GSocket.

Based on an earlier patch from Olivier Chalouhi.

https://bugzilla.gnome.org/show_bug.cgi?id=626589

12 years agoUpdated Norwegian bokmål translation
Kjartan Maraas [Sun, 15 Jan 2012 14:52:24 +0000 (15:52 +0100)]
Updated Norwegian bokmål translation

12 years agoGAsyncInitable: fix subclassibility
Dan Winship [Thu, 5 Jan 2012 18:04:37 +0000 (13:04 -0500)]
GAsyncInitable: fix subclassibility

If a class implements GAsyncInitable, and its parent also implements
it, then the subclass needs to call its parent's init_async() before
running its own. This was made more complicated by the fact that the
default init_finish() behavior was handled by the wrapper method
(which can't be used when making the super call) rather than the
default implementation itself. Fix that.

https://bugzilla.gnome.org/show_bug.cgi?id=667375

12 years agoFix typo
Alexander Larsson [Mon, 16 Jan 2012 09:38:35 +0000 (10:38 +0100)]
Fix typo

registred_resources -> registered_resources

12 years agoProtect call to pthread_condattr_setclock with define.
Patrick Welche [Thu, 12 Jan 2012 15:38:15 +0000 (15:38 +0000)]
Protect call to pthread_condattr_setclock with define.

While here update pthread_attr_setstacksize test to use AC_LINK_IFELSE
and avoid an unused variable in glib/tests/thread.c.

https://bugzilla.gnome.org/show_bug.cgi?id=667790

12 years agog_error_new_valist, g_error_copy: warn if domain is 0 or message is NULL
Simon McVittie [Fri, 30 Sep 2011 13:02:29 +0000 (14:02 +0100)]
g_error_new_valist, g_error_copy: warn if domain is 0 or message is NULL

Neither of those usages is valid, but there's a lot of use of 0 as a
domain "in the wild", so we can't g_return_if_fail yet.

Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Bug: https://bugzilla.gnome.org/show_bug.cgi?id=660371

12 years agoGError: small documentation addition
Matthias Clasen [Mon, 16 Jan 2012 04:18:09 +0000 (23:18 -0500)]
GError: small documentation addition

Document that out variables are not guaranteed to be set to
defined values if an error is thrown. Inspired by bug 658315

12 years agoBeef up Libs.private in glib-2.0.pc
Matthias Clasen [Mon, 16 Jan 2012 03:15:10 +0000 (22:15 -0500)]
Beef up Libs.private in glib-2.0.pc

This should help getting static builds working on mingw.
Based on a patch by Volker Grabsch, bug 619126.

At the same time, drop the unnecessary GLIB_RT_LIBS variable;
we are already adding -lrt to G_THREAD_LIBS.

12 years agoTrivial docs fix
Matthias Clasen [Mon, 16 Jan 2012 02:09:31 +0000 (21:09 -0500)]
Trivial docs fix

I like GLib to be consistently capitalized in the docs.

12 years agoUpdated Spanish translation
Daniel Mustieles [Sun, 15 Jan 2012 18:28:05 +0000 (19:28 +0100)]
Updated Spanish translation

12 years agogmain: fix adding a child source to an already-attached source
Dan Winship [Sun, 15 Jan 2012 14:35:11 +0000 (09:35 -0500)]
gmain: fix adding a child source to an already-attached source

Adding a child source to an already-attached parent source would
crash, because we were passing the parent's context when setting the
child's priority.

12 years agoEnsure to generate a flags GType for empty Flags enums
Christian Persch [Sat, 14 Jan 2012 23:34:56 +0000 (00:34 +0100)]
Ensure to generate a flags GType for empty Flags enums

When the flags enum only has the default NONE = 0 entry, glib-mkenums
creates an enum type for it, not a flags type. Add an annotation to the
enum to ensure the correct GType is created.

Bug #667938.

12 years agoresources: Use g_set_error_literal where appropriate
Christian Persch [Sat, 14 Jan 2012 20:13:03 +0000 (21:13 +0100)]
resources: Use g_set_error_literal where appropriate

12 years agoAdd missing Since: docs
Christian Persch [Sat, 14 Jan 2012 23:40:25 +0000 (00:40 +0100)]
Add missing Since: docs

12 years agoUpdated Galician translations
Fran Diéguez [Sat, 14 Jan 2012 23:42:25 +0000 (00:42 +0100)]
Updated Galician translations

12 years agogio/tests: Fix out-of-source build and use uninstalled libraries
Rico Tzschichholz [Sat, 14 Jan 2012 10:18:07 +0000 (11:18 +0100)]
gio/tests: Fix out-of-source build and use uninstalled libraries

12 years agoarray: return_if_fail() if element size is 0
Benjamin Otte [Sat, 14 Jan 2012 00:13:42 +0000 (01:13 +0100)]
array: return_if_fail() if element size is 0

This is particular useful for:
  g_array_new (sizeof (MyStruct), FALSE, FALSE);
because the correct incantation is
  g_array_new (FALSE, FALSE, sizeof (MyStruct));
and these warnings will trigger in the first situation.

12 years agovalues: Use v_pointer for g_value_set_gtype()
Benjamin Otte [Sun, 1 Jan 2012 20:51:40 +0000 (21:51 +0100)]
values: Use v_pointer for g_value_set_gtype()

... and g_value_get_gtype(). G_TYPE_GTYPE is a pointer type, so it's
values should use the v_pointer member. This is especially true, because
the value collectors from varargs in gvaluecollector.h use that, too.

This should only cause issues when sizeof(glong) != sizeof(gpointer),
and I'm not aware of any such platform. Maybe win64?

12 years agogio/tests: add some missing files to EXTRA_DIST
Rico Tzschichholz [Fri, 13 Jan 2012 22:48:12 +0000 (23:48 +0100)]
gio/tests: add some missing files to EXTRA_DIST

12 years agogio: Fix for -Wformat-security
Rico Tzschichholz [Fri, 13 Jan 2012 22:18:01 +0000 (23:18 +0100)]
gio: Fix for -Wformat-security

12 years agotests: Fix build failure of resources by linking against libgmodule-2.0.la
Rico Tzschichholz [Fri, 13 Jan 2012 21:35:44 +0000 (22:35 +0100)]
tests: Fix build failure of resources by linking against libgmodule-2.0.la

12 years agoUpdated POTFILES.in
Piotr Drąg [Fri, 13 Jan 2012 20:11:03 +0000 (21:11 +0100)]
Updated POTFILES.in

12 years agoAdd GResource functions to gio.symbols
Alexander Larsson [Thu, 12 Jan 2012 13:50:56 +0000 (14:50 +0100)]
Add GResource functions to gio.symbols

12 years agoAdd GLIB_COMPILE_RESOURCES to glib m4 macros
Alexander Larsson [Thu, 12 Jan 2012 20:09:15 +0000 (21:09 +0100)]
Add GLIB_COMPILE_RESOURCES to glib m4 macros

12 years agoAdd glib_compile_resources variable to the gio pc file
Alexander Larsson [Thu, 12 Jan 2012 19:51:56 +0000 (20:51 +0100)]
Add glib_compile_resources variable to the gio pc file

12 years agoAdd documentation for GResource
Alexander Larsson [Fri, 13 Jan 2012 15:39:28 +0000 (16:39 +0100)]
Add documentation for GResource

12 years agoSupport resource:/// uris
Alexander Larsson [Wed, 21 Dec 2011 22:37:56 +0000 (23:37 +0100)]
Support resource:/// uris

12 years agoAdd tests for GResource
Alexander Larsson [Wed, 21 Dec 2011 20:45:16 +0000 (21:45 +0100)]
Add tests for GResource

12 years agoInitial version of glib-compile-resources
Alexander Larsson [Wed, 21 Dec 2011 20:43:20 +0000 (21:43 +0100)]
Initial version of glib-compile-resources

This is the basic utility to create resource files, either binary
versions or source files that can be compiled into your app.

12 years agoInitial version of GResource
Alexander Larsson [Wed, 21 Dec 2011 20:38:23 +0000 (21:38 +0100)]
Initial version of GResource

GResource is a bundle of files combined into a single binary blog.
The API lets you access the files the resource contains by
using resource paths. You can also register resources with a
global list and access these globally in a merged resource namespace.

The normal way this is used is to link in the resources into your
application/library and have it be automatically registred.

Resources are compiled from an xml description using
glib-compile-resources.

12 years agoMerge remote-tracking branch 'gvdb/master'
Alexander Larsson [Fri, 13 Jan 2012 15:12:56 +0000 (16:12 +0100)]
Merge remote-tracking branch 'gvdb/master'

12 years agoAdd macros to handle constructor functions on the compilers that support it
Alexander Larsson [Wed, 21 Dec 2011 20:13:21 +0000 (21:13 +0100)]
Add macros to handle constructor functions on the compilers that support it

This is only supported on some compilers, so we define G_HAS_CONSTRUCTORS
when it is supported. However, when it is supported we guarantee that
both constructors and destructors work, in executables as well as shared
libraries (including runtime unloading of shared libraries).

Usage is a bit unorthodox, as some compilers need to use #pragma to
implement constructors, and #pragma can't be used in macros.

The canonical way to use this:

  #ifdef G_DEFINE_CONSTRUCTOR_NEEDS_PRAGMA
  #pragma G_DEFINE_CONSTRUCTOR_PRAGMA_ARGS(my_constructor)
  #endif
  G_DEFINE_CONSTRUCTOR(my_constructor)

  static void my_constructor (void)
  {
    ...

  #ifdef G_DEFINE_DESTRUCTOR_NEEDS_PRAGMA
  #pragma G_DEFINE_DESTRUCTOR_PRAGMA_ARGS(my_destructor)
  #endif
  G_DEFINE_DESTRUCTOR(my_destructor)

  static void my_destructor (void)
  {
    ...

12 years agoSupport creating a GvdbReader from data
Alexander Larsson [Fri, 13 Jan 2012 14:59:56 +0000 (15:59 +0100)]
Support creating a GvdbReader from data

This is needed for the glib resource work.

12 years agoDocs: Changed can not to cannot.
Murray Cumming [Fri, 13 Jan 2012 08:21:46 +0000 (09:21 +0100)]
Docs: Changed can not to cannot.

This is a backport from the glib commit 30fdc1a79938540af7bad0af68d0a3b8af14f117

12 years agoUpdated Hebrew translation.
Yaron Shahrabani [Fri, 13 Jan 2012 07:38:35 +0000 (09:38 +0200)]
Updated Hebrew translation.