platform/upstream/glib.git
13 years agotiny fix to g_io_modules_scan_all_in_directory() docs
Dan Winship [Mon, 29 Nov 2010 14:36:07 +0000 (09:36 -0500)]
tiny fix to g_io_modules_scan_all_in_directory() docs

13 years agogsettingsschema: plug leak
Xan Lopez [Mon, 29 Nov 2010 14:29:12 +0000 (15:29 +0100)]
gsettingsschema: plug leak

13 years agoAdd another bug
Matthias Clasen [Mon, 29 Nov 2010 13:10:11 +0000 (08:10 -0500)]
Add another bug

13 years agoFixes GB#635187: Always unbox GVariant parameter received via dbus for an action
Aleksander Morgado [Thu, 18 Nov 2010 17:00:57 +0000 (18:00 +0100)]
Fixes GB#635187: Always unbox GVariant parameter received via dbus for an action

13 years agoAdd new extension point to the list
Matthias Clasen [Mon, 29 Nov 2010 06:39:06 +0000 (01:39 -0500)]
Add new extension point to the list

13 years agoDocumentation tweak
Matthias Clasen [Mon, 29 Nov 2010 06:20:58 +0000 (01:20 -0500)]
Documentation tweak

13 years agoUpdate NEWS
Matthias Clasen [Mon, 29 Nov 2010 06:16:02 +0000 (01:16 -0500)]
Update NEWS

13 years agoVarious doc tweaks
Matthias Clasen [Mon, 29 Nov 2010 04:55:43 +0000 (23:55 -0500)]
Various doc tweaks

13 years agoRemove nonexisting type
Matthias Clasen [Mon, 29 Nov 2010 03:11:57 +0000 (22:11 -0500)]
Remove nonexisting type

13 years agoRemoved mention of inexistent GNOME 2.0 porting guide
Matt Rajca [Sat, 13 Nov 2010 16:01:34 +0000 (10:01 -0600)]
Removed mention of inexistent GNOME 2.0 porting guide

13 years agoFixed typo in public string in gsocks (630559)
Matt Rajca [Sat, 13 Nov 2010 19:01:11 +0000 (13:01 -0600)]
Fixed typo in public string in gsocks (630559)

13 years agogobject: added property test
Jonas Holmberg [Wed, 22 Sep 2010 21:02:05 +0000 (23:02 +0200)]
gobject: added property test

Added test for setting properties with g_object_new.

13 years agogobject: initialize memory in g_object_new_valist
Jonas Holmberg [Thu, 23 Sep 2010 23:20:50 +0000 (01:20 +0200)]
gobject: initialize memory in g_object_new_valist

memset parameters array in g_object_new_valist to zeroes when expanding
the array to avoid acces to uninitialized memory.

13 years agoFixes GB#530786: GFileMonitor 'changed' signal underdocumented
Aleksander Morgado [Thu, 25 Nov 2010 17:38:33 +0000 (18:38 +0100)]
Fixes GB#530786: GFileMonitor 'changed' signal underdocumented

13 years agoFix the wrong-category schema test
Christian Persch [Fri, 26 Nov 2010 21:46:51 +0000 (22:46 +0100)]
Fix the wrong-category schema test

It's supposed to test a <default l10n="..."> with a non-existent
category, so make it actually check this!

Bug #635882.

13 years agoGFileMonitor: Don't accept negative values for rate-limit
Matthias Clasen [Sun, 28 Nov 2010 22:14:49 +0000 (17:14 -0500)]
GFileMonitor: Don't accept negative values for rate-limit

Patch by Alksander Morgado,
https://bugzilla.gnome.org/show_bug.cgi?id=635768

13 years agoInherit gettext-domain from <schemalist>
Christian Persch [Fri, 26 Nov 2010 20:05:20 +0000 (21:05 +0100)]
Inherit gettext-domain from <schemalist>

When the <schema> doesn't have a 'gettext-domain' attribute, but
the <schemalist> does, use that one.

Bug #635640.

13 years agoMake gio/tests/g-icon pass again
Dan Winship [Sat, 27 Nov 2010 21:56:44 +0000 (16:56 -0500)]
Make gio/tests/g-icon pass again

It got broken in two different ways by the g_str_hash() change
(354d655b)

13 years agoTypo fix
Christian Persch [Mon, 15 Nov 2010 18:55:57 +0000 (19:55 +0100)]
Typo fix

13 years agoUse same variable names in .h and .c
Christian Persch [Sat, 27 Nov 2010 11:45:45 +0000 (12:45 +0100)]
Use same variable names in .h and .c

13 years agoAdd initial TLS (SSL) support to gio
Dan Winship [Mon, 21 Dec 2009 19:50:32 +0000 (20:50 +0100)]
Add initial TLS (SSL) support to gio

This adds an extension point for TLS connections to gio, with a
gnutls-based implementation in glib-networking.

Full TLS support is still a work in progress; the current API is
missing some features, and parts of it may still be changed before
2.28.

https://bugzilla.gnome.org/show_bug.cgi?id=588189

13 years agomake GProxyConnection public, as GTcpWrapperConnection
Dan Winship [Tue, 2 Nov 2010 00:22:24 +0000 (20:22 -0400)]
make GProxyConnection public, as GTcpWrapperConnection

GProxyConnection is a class that was added for proxy support;
g_socket_client_connect() returns a GSocketConnection, but in some
cases (eg, encrypted SOCKS), GProxy might return a GIOStream that is
not a GSocketConnection. In that case, GSocketClient would wrap the
stream up in a GProxyConnection, which is a subclass of
GSocketConnection but uses the input/output streams of the wrapped
connection.

GTlsConnection is not a GSocketConnection, so it has the same problem,
so it will need the same treatment. Rename the class to
GTcpWrapperStream, and make it public, so people can extract the base
stream from it when necessary.

(This is not ideal and GSocketClient will need to be revisited as an
API at some point...)

https://bugzilla.gnome.org/show_bug.cgi?id=588189

13 years agoAdd pollable input/output streams
Dan Winship [Sat, 18 Sep 2010 17:05:25 +0000 (13:05 -0400)]
Add pollable input/output streams

When interfacing with APIs that expect unix-style async I/O, it is
useful to be able to tell in advance whether a read/write is going to
block. This adds new interfaces GPollableInputStream and
GPollableOutputStream that can be implemented by a GInputStream or
GOutputStream to add _is_readable/_is_writable, _create_source, and
_read_nonblocking/_write_nonblocking methods.

Also, implement for GUnixInput/OutputStream and
GSocketInput/OutputStream

https://bugzilla.gnome.org/show_bug.cgi?id=634241

13 years agoGCancellable: add g_cancellable_create_source()
Dan Winship [Sat, 6 Nov 2010 19:49:55 +0000 (15:49 -0400)]
GCancellable: add g_cancellable_create_source()

g_cancellable_create_source() returns a GSource that triggers when its
corresponding GCancellable is cancelled. This can be used with
g_source_add_child_source() to add cancellability to a source.

Port gasynchelper's FDSource to use this rather than doing its own
cancellable handling, and also fix up its callback argument order to
be more normal.

https://bugzilla.gnome.org/show_bug.cgi?id=634239

13 years agogmain: add g_source_add_child_source and g_source_remove_child_source
Dan Winship [Sat, 6 Nov 2010 14:11:15 +0000 (10:11 -0400)]
gmain: add g_source_add_child_source and g_source_remove_child_source

This adds "child source" support to GSource. A child source behaves
basically like a GPollFD; when you add a source to a context, all of
its child sources are added with the same priority; when you destroy a
source, all of its child sources are destroyed; and when a child
source triggers, its parent source's dispatch function is run.

Use cases include:

    - adding a GTimeoutSource to another source to cause the source to
      automatically trigger after a certain timeout.

    - wrapping an existing source type with a new type that has
      a different callback signature

    - creating a source that triggers based on different conditions
      at different times.

https://bugzilla.gnome.org/show_bug.cgi?id=634239

13 years agoAdd g_source_set_dummy_callback()
Dan Winship [Sun, 7 Nov 2010 16:49:40 +0000 (11:49 -0500)]
Add g_source_set_dummy_callback()

Use g_source_set_closure() and g_close_set_meta_marshal() to allow
setting a do-nothing callback on any source.

https://bugzilla.gnome.org/show_bug.cgi?id=634239

13 years agoImplement closure-related methods for gio GSource types
Dan Winship [Sun, 7 Nov 2010 16:05:26 +0000 (11:05 -0500)]
Implement closure-related methods for gio GSource types

Also, fix up the argument ordering on GFDSourceFunc

https://bugzilla.gnome.org/show_bug.cgi?id=634239

13 years agogmain: fix some silly code in a programmer-error case
Dan Winship [Sat, 6 Nov 2010 13:45:20 +0000 (09:45 -0400)]
gmain: fix some silly code in a programmer-error case

Previously if a source got finalized while still attached to a
context, it would warn and re-ref the source. But then it just freed
it anyway... So keep the warning but drop the re-ref.

https://bugzilla.gnome.org/show_bug.cgi?id=634239

13 years agogmain: move finalization of GSource outside of context lock
Dan Winship [Sat, 6 Nov 2010 13:35:25 +0000 (09:35 -0400)]
gmain: move finalization of GSource outside of context lock

This avoids ugly deadlock situations such as in
https://bugzilla.gnome.org/show_bug.cgi?id=586432

https://bugzilla.gnome.org/show_bug.cgi?id=626702

https://bugzilla.gnome.org/show_bug.cgi?id=634239

13 years agoAdd indices for new symbols in 2.28
Matthias Clasen [Wed, 24 Nov 2010 21:00:32 +0000 (16:00 -0500)]
Add indices for new symbols in 2.28

13 years agoAdd Since: tags for new api
Matthias Clasen [Wed, 24 Nov 2010 20:57:40 +0000 (15:57 -0500)]
Add Since: tags for new api

Also, remove <!-- --> from things like #GAppInfo<!-- -->s.
gtk-doc learned to deal with this a while ago.

13 years agodocs: gappinfo: Fix typo <envvar> -> <envar>
Javier Jardón [Wed, 24 Nov 2010 17:06:17 +0000 (18:06 +0100)]
docs: gappinfo: Fix typo <envvar> -> <envar>

13 years agotests: add a test for recommended and fallback GAppInfos
Cosimo Cecchi [Fri, 19 Nov 2010 10:37:44 +0000 (11:37 +0100)]
tests: add a test for recommended and fallback GAppInfos

13 years agoappinfo: add win32 fallback implementation
Cosimo Cecchi [Fri, 19 Nov 2010 09:39:33 +0000 (10:39 +0100)]
appinfo: add win32 fallback implementation

13 years agoappinfo: update docs for API addition
Cosimo Cecchi [Fri, 19 Nov 2010 09:39:16 +0000 (10:39 +0100)]
appinfo: update docs for API addition

13 years agoappinfo: only rewrite recommended applications when adding non-default
Cosimo Cecchi [Thu, 18 Nov 2010 17:49:32 +0000 (18:49 +0100)]
appinfo: only rewrite recommended applications when adding non-default

Otherwise you break the fallback + recommended distinction for a content
type, as you end up adding support for a given content type to *all* of
the applications claiming to support the supertype.

13 years agoappinfo: add API to get fallback and recommended GAppInfos for a ctype
Cosimo Cecchi [Thu, 18 Nov 2010 17:45:56 +0000 (18:45 +0100)]
appinfo: add API to get fallback and recommended GAppInfos for a ctype

This ways UIs can differentiate between them, and show them in different
section.
- a recommended app is an application that claims to support a content
  type directly, not by a supertype derivation.
- a fallback app is an application that claims to support a supertype of
  a given content type.

13 years agoFix off-by-1000 for GTimer
Owen W. Taylor [Mon, 22 Nov 2010 02:59:57 +0000 (21:59 -0500)]
Fix off-by-1000 for GTimer

Divide monotonic time by 1e6 not 1e9 to get seconds.

13 years agoCheck for (GIConv)-1 instead of NULL.
Ignacio Casal Quinteiro [Sun, 21 Nov 2010 20:20:44 +0000 (21:20 +0100)]
Check for (GIConv)-1 instead of NULL.

13 years agoAdded UG translation
Gheyret T.Kenji [Sat, 20 Nov 2010 10:26:41 +0000 (11:26 +0100)]
Added UG translation

13 years agoUpdated Norwegian bokmål translation
Kjartan Maraas [Thu, 18 Nov 2010 13:08:34 +0000 (14:08 +0100)]
Updated Norwegian bokmål translation

13 years agointrospection: Add annotations for g_file_copy, g_file_move
Colin Walters [Thu, 18 Nov 2010 01:06:07 +0000 (20:06 -0500)]
introspection: Add annotations for g_file_copy, g_file_move

13 years agog_str_hash: switch to using DJB hash
Ryan Lortie [Wed, 17 Nov 2010 17:19:54 +0000 (12:19 -0500)]
g_str_hash: switch to using DJB hash

This is the same as what we were already doing with 2 changes:

  - use an initial value of 5381 instead of 0

  - multiply by 33 in each round instead of 31

13 years agog_str_hash: clean up code
Ryan Lortie [Wed, 17 Nov 2010 16:57:48 +0000 (11:57 -0500)]
g_str_hash: clean up code

Un-unroll the first iteration.

No functional changes here.

13 years agoGIO: Hide mounts having a path element starting with dot
Tomas Bzatek [Tue, 16 Nov 2010 13:20:07 +0000 (14:20 +0100)]
GIO: Hide mounts having a path element starting with dot

When a mount is mounted to a directory whose path contains an element
starting with dot, let's treat it as hidden, just like the usual Unix
approach.

13 years agodocs: Update the URL of some documentation links
Javier Jardón [Sun, 14 Nov 2010 01:08:51 +0000 (02:08 +0100)]
docs: Update the URL of some documentation links

http://developer.gnome.org -> http://library.gnome.org

13 years agoAdded UG translation
Gheyret T.Kenji [Sat, 13 Nov 2010 23:05:02 +0000 (00:05 +0100)]
Added UG translation

14 years agoAdded UG translation
Gheyret T.Kenji [Sat, 13 Nov 2010 21:10:03 +0000 (22:10 +0100)]
Added UG translation

14 years ago[l10n] Updated Italian translation
Luca Ferretti [Sat, 13 Nov 2010 13:01:29 +0000 (14:01 +0100)]
[l10n] Updated Italian translation

14 years agoUpdated Galician translations
Fran Diéguez [Fri, 12 Nov 2010 23:59:45 +0000 (00:59 +0100)]
Updated Galician translations

14 years agoG_STATIC_ASSERT: clarify when use is valid
Ryan Lortie [Fri, 12 Nov 2010 02:47:13 +0000 (21:47 -0500)]
G_STATIC_ASSERT: clarify when use is valid

Clarify when the use of G_STATIC_ASSERT is valid and fix up an invalid
use of it in GDBus.

14 years agoPartial revert of last commit
Ryan Lortie [Fri, 12 Nov 2010 02:30:03 +0000 (21:30 -0500)]
Partial revert of last commit

Revert incorrect changes to gdbusmessage.c

14 years agoemblemed-icon: allow passing a NULL emblem to g_emblemed_icon_new()
Cosimo Cecchi [Wed, 10 Nov 2010 11:03:11 +0000 (12:03 +0100)]
emblemed-icon: allow passing a NULL emblem to g_emblemed_icon_new()

So that if we already have a list of emblems we know we want to add to
the icon, we can use e.g. a for loop to add them all, without picking
the first and looping from the second.

https://bugzilla.gnome.org/show_bug.cgi?id=634504

14 years agoAPI: Reinstate "gio-desktop-app-info-lookup" extension point
Benjamin Otte [Tue, 9 Nov 2010 12:19:19 +0000 (07:19 -0500)]
API: Reinstate "gio-desktop-app-info-lookup" extension point

Removing an extension point is an API and ABI break. In particular, it
causes (older) gvfs versions to fail loading with a linkage error from
ld which in turn makes the desktop unusable.
So this reinstate the extension point and API provided by it, but
deprecates and does not use it. So no functionality is changed.

This reverts parts of commit 9b262f1c5fe5a6fd879f17cd7b80d8c54e33d80c.

Complaints-Also-To: Ryan Lortie <desrt@desrt.ca>

14 years agogio: Fix C99 style variable declarations (bug #633075)
Chun-wei Fan [Tue, 9 Nov 2010 12:36:53 +0000 (20:36 +0800)]
gio: Fix C99 style variable declarations (bug #633075)

-gdbusmessage.c and gregistrysettingsbackend.c is updated to fix C99-style declarations
-also fixed typo for displaying registry entry in gregistrysettingsbackend.c (\% -> \\%)

14 years agogtimer.c: Revert accidental overwrite
Chun-wei Fan [Tue, 9 Nov 2010 12:32:39 +0000 (20:32 +0800)]
gtimer.c: Revert accidental overwrite

Revert accidental overwrite as timing API was changed from 2.27.2 to 2.27.3

14 years agoBump version
Matthias Clasen [Tue, 9 Nov 2010 04:06:53 +0000 (23:06 -0500)]
Bump version

14 years agoUpdate NEWS 2.27.3
Matthias Clasen [Tue, 9 Nov 2010 02:40:17 +0000 (21:40 -0500)]
Update NEWS

14 years agoAdded and Moved checks for includes
Chun-wei Fan [Tue, 9 Nov 2010 01:53:12 +0000 (09:53 +0800)]
Added and Moved checks for includes

Moved checks for G_OS_WIN32 after GLib header includes and added other checks
required for Windows/MSVC builds

14 years agoUpdated README.txt for VS9
Chun-wei Fan [Tue, 9 Nov 2010 01:50:11 +0000 (09:50 +0800)]
Updated README.txt for VS9

14 years agoUpdated .def generation for x64
Chun-wei Fan [Tue, 9 Nov 2010 01:27:27 +0000 (09:27 +0800)]
Updated .def generation for x64

Some symbols need to be excluded from the .def file as they are set to be so on Win64

14 years agoVarious updates...
Chun-wei Fan [Tue, 9 Nov 2010 01:26:05 +0000 (09:26 +0800)]
Various updates...

-Made up for missed header files in glib and gio during "install"
-Added macro necessary for GLib/GModule .def generation under Win64
-updated location of getting glibconfig.h.win32 for building

14 years agoSome enhancements for Win64 and using existing PCRE installtions
Chun-wei Fan [Tue, 9 Nov 2010 01:23:47 +0000 (09:23 +0800)]
Some enhancements for Win64 and using existing PCRE installtions

Added option for people to use an existing PCRE build and updated .def generation for x64 systems (some symbols are set to be excluded from Win64 builds)
Also fixed the filter "PCRE" for the bundled PCRE as file layout changed

14 years agoAdded option to use existing installation of PCRE
Chun-wei Fan [Tue, 9 Nov 2010 01:11:48 +0000 (09:11 +0800)]
Added option to use existing installation of PCRE

Allow the person compiling GLib to use his/her own copy of PCRE, bug#633075

14 years agoMade up for missed source files
Chun-wei Fan [Tue, 9 Nov 2010 01:08:45 +0000 (09:08 +0800)]
Made up for missed source files

Made up for missing required source files in  gio.vcproj(in) for Windows builds

14 years agoGPeriodic: enable subclassing
Ryan Lortie [Mon, 8 Nov 2010 23:15:32 +0000 (18:15 -0500)]
GPeriodic: enable subclassing

Move GPeriodic struct and class struct to the header.  Do ->priv.

14 years agoGPeriodic: remove repair functions
Ryan Lortie [Mon, 8 Nov 2010 21:59:29 +0000 (16:59 -0500)]
GPeriodic: remove repair functions

Replace it with a single repair signal that either fires or does not.

14 years agoNew function: g_clear_object()
Ryan Lortie [Mon, 8 Nov 2010 21:42:32 +0000 (16:42 -0500)]
New function: g_clear_object()

By analogy to g_clear_error, takes a pass-by-reference GObject reference
and, if non-%NULL, unrefs it and sets it equal to %NULL.

Bug #620263.

14 years agofix .gitignore
Dan Winship [Sun, 7 Nov 2010 17:56:44 +0000 (12:56 -0500)]
fix .gitignore

14 years agofix make check
Dan Winship [Sun, 7 Nov 2010 17:56:08 +0000 (12:56 -0500)]
fix make check

14 years agog_get_user_runtime_dir(): New function
Ryan Lortie [Sat, 6 Nov 2010 21:34:40 +0000 (17:34 -0400)]
g_get_user_runtime_dir(): New function

Get the value of the XDG_RUNTIME_DIR environment variable.

14 years agoAdd G_DBUS_SIGNAL_FLAGS_NO_MATCH_RULE
Ryan Lortie [Sat, 6 Nov 2010 02:07:07 +0000 (22:07 -0400)]
Add G_DBUS_SIGNAL_FLAGS_NO_MATCH_RULE

If specified, the signal subscription is setup client-side but the match
rule is not sent to the server.  This allows the caller to manually
register more detailed match rules.

14 years agoFix docs comment typo
Christian Persch [Sat, 6 Nov 2010 11:27:01 +0000 (12:27 +0100)]
Fix docs comment typo

14 years agoupdate Punjabi Translation by A S Alam
A S Alam [Sat, 6 Nov 2010 05:18:55 +0000 (10:48 +0530)]
update Punjabi Translation by A S Alam

14 years agoupdate Punjabi Translation by A S Alam
A S Alam [Sat, 6 Nov 2010 05:17:54 +0000 (10:47 +0530)]
update Punjabi Translation by A S Alam

14 years agoAdd g_variant_lookup() and tests
Ryan Lortie [Sat, 6 Nov 2010 01:33:06 +0000 (21:33 -0400)]
Add g_variant_lookup() and tests

Convenience API for doing lookups in dictionaries where the key is a
string or object path.

14 years agoFix markup
Matthias Clasen [Fri, 5 Nov 2010 18:50:01 +0000 (14:50 -0400)]
Fix markup

14 years agoFix up gapplication example
Matthias Clasen [Fri, 5 Nov 2010 18:29:15 +0000 (14:29 -0400)]
Fix up gapplication example

14 years agoDescribe enum and flags types a bit
Matthias Clasen [Fri, 5 Nov 2010 18:28:44 +0000 (14:28 -0400)]
Describe enum and flags types a bit

14 years agoUpdate the gschema dtd to include flags
Matthias Clasen [Fri, 5 Nov 2010 17:56:42 +0000 (13:56 -0400)]
Update the gschema dtd to include flags

14 years agoReport more useful errors from g_settings_set_value
Matthias Clasen [Fri, 5 Nov 2010 13:31:36 +0000 (09:31 -0400)]
Report more useful errors from g_settings_set_value

14 years agogsettings.m4: Fix rules to work when there are no schemas
Dan Winship [Thu, 4 Nov 2010 19:23:22 +0000 (15:23 -0400)]
gsettings.m4: Fix rules to work when there are no schemas

If there are no schemas, don't try to install "" at install time.
(In particular, automake conditionals don't work properly with
@-expanded rules, so if you conditionally build a schema, you'll
still unconditionally get the install rule.)

https://bugzilla.gnome.org/show_bug.cgi?id=633381

14 years agoClarify docs for g_path_is_absolute() semantics on Windows
Tor Lillqvist [Tue, 26 Oct 2010 09:20:22 +0000 (12:20 +0300)]
Clarify docs for g_path_is_absolute() semantics on Windows

14 years agoUpdated Belarusian translation
Ihar Hrachyshka [Sun, 31 Oct 2010 18:12:26 +0000 (20:12 +0200)]
Updated Belarusian translation

14 years agoAdd g_simple_async_report_take_gerror_in_idle
Christian Persch [Thu, 23 Sep 2010 11:47:13 +0000 (13:47 +0200)]
Add g_simple_async_report_take_gerror_in_idle

... and use it where appropriate. Saves an extra GError copy.

Bug #633686.

14 years agoUse g_simple_async_result_{new_,}take_error
Christian Persch [Thu, 23 Sep 2010 11:02:50 +0000 (13:02 +0200)]
Use g_simple_async_result_{new_,}take_error

Bug #633685.

14 years agoAdd simple dynamic scheduling to GPeriodic
Ryan Lortie [Wed, 3 Nov 2010 04:16:16 +0000 (00:16 -0400)]
Add simple dynamic scheduling to GPeriodic

GPeriodic will now try not to use more than 50% of the CPU for its own
purposes unless there are no other tasks attempting to run.

14 years agoTweak GApplication docs
Matthias Clasen [Wed, 3 Nov 2010 02:39:58 +0000 (22:39 -0400)]
Tweak GApplication docs

14 years agodocs: Recommend against using GTimeVal
Ryan Lortie [Wed, 3 Nov 2010 02:12:55 +0000 (22:12 -0400)]
docs: Recommend against using GTimeVal

14 years agoGMainContext: store real time as int64
Ryan Lortie [Wed, 3 Nov 2010 02:03:08 +0000 (22:03 -0400)]
GMainContext: store real time as int64

14 years agoglocalfileinfo: remove non-use of GTimeVal
Ryan Lortie [Mon, 1 Nov 2010 21:04:31 +0000 (17:04 -0400)]
glocalfileinfo: remove non-use of GTimeVal

14 years agoAdd g_get_real_time() for wall-clock int64 micros
Ryan Lortie [Mon, 1 Nov 2010 20:40:36 +0000 (16:40 -0400)]
Add g_get_real_time() for wall-clock int64 micros

Similar in spirit to g_get_monotonic_time().

14 years agoClean up g_usleep()
Ryan Lortie [Mon, 1 Nov 2010 20:35:10 +0000 (16:35 -0400)]
Clean up g_usleep()

Remove some code that was written in 2000 to support OSes that do not
have nanosleep().  nanosleep() has been specified (in POSIX-1.2001) for
almost a decade now, so assume we have it (except on Windows).

Remove the checks for nanosleep and nsleep from configure.ac.

We're removing this code because we honestly believe that nobody will be
affected.  If this change negatively impacts you, please file a bug.

14 years agoGTimer: switch to monotonic time
Ryan Lortie [Mon, 1 Nov 2010 20:06:39 +0000 (16:06 -0400)]
GTimer: switch to monotonic time

and remove docs notes about threads having to be initialised.

14 years agoDrop GTimeSpec type
Ryan Lortie [Mon, 1 Nov 2010 19:47:29 +0000 (15:47 -0400)]
Drop GTimeSpec type

14 years agoSwitch GTimeSpec users to int64 microseconds
Ryan Lortie [Mon, 1 Nov 2010 19:46:35 +0000 (15:46 -0400)]
Switch GTimeSpec users to int64 microseconds

glib is trying to move toward using microseconds-in-gint64 as its
universal time format.

No real API breaks here since GTimeSpec is new this unstable release
series.

14 years agoGTimeout: simplify math for 'seconds' case
Ryan Lortie [Mon, 1 Nov 2010 17:31:37 +0000 (13:31 -0400)]
GTimeout: simplify math for 'seconds' case

The code was designed to deal with any granularity of timer and due to
the use of GTimeVal/GTimeSpec, the math for this gets extremely
confusing.

From a practical standpoint, we only ever have a granularity of seconds.

Take advantage of that fact in the code and vastly simplify the math.

14 years agoUpdated Spanish translation
Jorge González [Tue, 2 Nov 2010 09:46:10 +0000 (10:46 +0100)]
Updated Spanish translation

14 years agoUpdated Spanish translation
Jorge González [Mon, 1 Nov 2010 17:50:46 +0000 (18:50 +0100)]
Updated Spanish translation

14 years agoUpdated Galician translations
Fran Diéguez [Mon, 1 Nov 2010 16:15:38 +0000 (17:15 +0100)]
Updated Galician translations