Matthias Clasen [Thu, 28 Jul 2011 03:27:24 +0000 (23:27 -0400)]
Revert "Optimize g_[s]list_free_full a bit"
This reverts commit
98b667d052b1274f80b8898a32d0753e9e2e5c1a.
The commit was not actually an optimization, since g_list_free is
pretty smart.
Nicolas Dufresne [Mon, 25 Jul 2011 19:59:58 +0000 (15:59 -0400)]
Added TLS PEM parser unit test
Nicolas Dufresne [Mon, 20 Jun 2011 22:49:39 +0000 (18:49 -0400)]
Rework PEM parsing to not be order dependent
Some valid PEM file would not work because the private key was put
before the certificate.
Tomas Bzatek [Tue, 26 Jul 2011 15:32:53 +0000 (17:32 +0200)]
Docs: Correct the namespace for g_file_query_filesystem_info()
Chun-wei Fan [Tue, 26 Jul 2011 15:23:31 +0000 (23:23 +0800)]
Updated glib-zip.in a bit further
Missed the bin/gsettings.exe part (corrected from bin/gsettings.)
Cosimo Cecchi [Tue, 26 Jul 2011 14:44:18 +0000 (16:44 +0200)]
utf8: annotate the end pointer in g_utf8_validate as out + allow-none
David Zeuthen [Tue, 26 Jul 2011 14:05:57 +0000 (10:05 -0400)]
More gdbus-codegen fixed to build on non-Unix
https://bugzilla.gnome.org/show_bug.cgi?id=655148#c6
Signed-off-by: David Zeuthen <davidz@redhat.com>
Ryan Lortie [Tue, 26 Jul 2011 11:49:20 +0000 (13:49 +0200)]
gtester: small Windows fix
This code was unconditionally present in the gtester Makefile:
test-nonrecursive: ${TEST_PROGS}
On Windows, our testcases are compiled with a .exe suffix. That means
that if we had 'foo' in TEST_PROGS, running "make check" would depend on
'foo' (not foo.exe) being compiled.
We could bring the EXEEXT in here to fix that up, but gtester doesn't
work on Windows at all, so better to just disable it in that case.
Ryan Lortie [Tue, 26 Jul 2011 11:38:04 +0000 (13:38 +0200)]
GDateTime test: 1970 doesn't exist on Windows
It is not safe to call the system library mktime() function on Windows
with a date in the 70s. Use 1990 instead.
Ryan Lortie [Tue, 26 Jul 2011 11:29:01 +0000 (13:29 +0200)]
g_format_size: fix on Windows
The long format that displays the exact number of bytes with separators
(ie: "123,456,789 bytes") uses the ' format modifier, which is
unsupported on Windows. Disable that for now, until we come up with a
better solution.
Ryan Lortie [Tue, 26 Jul 2011 11:26:10 +0000 (13:26 +0200)]
g_format_size: avoid silly GString use
We were using a GString for the purpose of doing a single printf(). Do
g_strdup_printf() instead.
Dieter Verfaillie [Mon, 18 Jul 2011 12:20:18 +0000 (14:20 +0200)]
Update glib-zip.in
Ryan Lortie [Tue, 26 Jul 2011 09:20:12 +0000 (11:20 +0200)]
glocalfile: don't bother with fstype on win32
Fixes the broken build on Windows.
Ryan Lortie [Tue, 26 Jul 2011 08:50:25 +0000 (10:50 +0200)]
Merge branch 'gwakeup'
Ryan Lortie [Mon, 25 Jul 2011 16:59:27 +0000 (18:59 +0200)]
GWakeup: fix Windows build breakage
...from the attempt to make it private.
Ryan Lortie [Mon, 25 Jul 2011 16:52:36 +0000 (18:52 +0200)]
.gitignore the gwakeup-fallback testcase
Ryan Lortie [Mon, 25 Jul 2011 16:52:18 +0000 (18:52 +0200)]
Remove g_wakeup_* from glib.symbols
Ryan Lortie [Mon, 25 Jul 2011 16:50:45 +0000 (18:50 +0200)]
GWakeup: make it private API
Colin requests that we keep this one private for now.
Include it at each point of use (libglib, libgio, tests).
Ryan Lortie [Mon, 25 Jul 2011 15:43:28 +0000 (17:43 +0200)]
GWakeup: test fallback case
We need to test the case of eventfd in the libc but no kernel support.
In order to do that, we add a separate compile of the GWakeup testcase
that interposes an 'eventfd' symbol that always returns -1 with errno
set. That will trigger the fallback case.
Ryan Lortie [Mon, 25 Jul 2011 14:35:08 +0000 (16:35 +0200)]
GWakeup: add signal safety note
Note that g_wakeup_signal() is safe to call from a UNIX signal handler
(since this is a likely place to want to call it from).
David Zeuthen [Mon, 25 Jul 2011 14:31:17 +0000 (10:31 -0400)]
More fixes for non-Unix builds
Signed-off-by: David Zeuthen <davidz@redhat.com>
David Zeuthen [Mon, 25 Jul 2011 13:56:01 +0000 (09:56 -0400)]
Fix the build on non-Unix
https://bugzilla.gnome.org/show_bug.cgi?id=655148
Signed-off-by: David Zeuthen <davidz@redhat.com>
Ryan Lortie [Mon, 25 Jul 2011 13:08:47 +0000 (15:08 +0200)]
GWakeup .gitignore fixes
Ryan Lortie [Mon, 25 Jul 2011 12:37:18 +0000 (14:37 +0200)]
gtk-doc GWakeup
Ryan Lortie [Mon, 25 Jul 2011 13:02:28 +0000 (15:02 +0200)]
GCancellable: port to GWakeup
Ryan Lortie [Mon, 25 Jul 2011 13:07:33 +0000 (15:07 +0200)]
port GMainContext to GWakeup
Ryan Lortie [Mon, 25 Jul 2011 13:07:16 +0000 (15:07 +0200)]
add a couple of testcases for GWakeup
Ryan Lortie [Mon, 25 Jul 2011 13:05:03 +0000 (15:05 +0200)]
Add GWakeup
GWakeup is a utility class to handle the cross-thread signalling needs
of GMainContext and GCancellable. It may find some other users as well.
The desire here is to properly hide the implementation details in a
module which can be properly unit tested and used in GMainContext and
GCancellable without a rats nest of #ifdef.
Ryan Lortie [Mon, 25 Jul 2011 13:25:13 +0000 (15:25 +0200)]
g_unix_open_pipe: fix some bugs
Fix some bugs in the fallback case of g_unix_open_pipe:
- close both halves of the pipe on error (not just one)
- set the cloexec flag on both halves of the pipe (instead of settings
it twice on one half)
Ryan Lortie [Mon, 25 Jul 2011 13:09:57 +0000 (15:09 +0200)]
Windows (mingw32) .gitignore additions
Murray Cumming [Mon, 25 Jul 2011 07:23:51 +0000 (09:23 +0200)]
Docs: Fix tiny typo.
Chun-wei Fan [Mon, 25 Jul 2011 03:42:17 +0000 (11:42 +0800)]
Update Windows READMEs
-Tell people about that libFFI is now needed.
-Tell people not to compile GLib in paths containing spaces.
Dan Winship [Sun, 24 Jul 2011 18:55:05 +0000 (14:55 -0400)]
gio/tests/socket: add some basic IPv4 and IPv6 tests
Dan Winship [Sun, 24 Jul 2011 18:52:03 +0000 (14:52 -0400)]
GSocket: fix connected state after async connect
socket->priv->connected was only being set if g_socket_connect()
succeeded right away; in the case where it returns G_IO_ERROR_PENDING,
it never got set. Fix that by having g_socket_check_connect_result()
set it on success.
Dan Winship [Sun, 24 Jul 2011 16:41:11 +0000 (12:41 -0400)]
update .gitignore
Matthias Clasen [Sun, 24 Jul 2011 01:59:49 +0000 (21:59 -0400)]
Explicitly unset the session bus address for the peer tests
Otherwise, we may run into trouble as opening a peer-to-peer
connection uses a socket client, which uses a proxy resolver
which may end up using gsettings, whose dconf backend may end
up using the session bus to talk to dconfd...
Matthias Clasen [Sun, 24 Jul 2011 01:16:28 +0000 (21:16 -0400)]
Properly bring up a session bus for application tests
As pointed out in bug 644601, session_bus_up() requires
us to set up environment variables for things to work.
Matthias Clasen [Sun, 24 Jul 2011 01:07:39 +0000 (21:07 -0400)]
Fix a typo
Yuri Kozlov [Sat, 23 Jul 2011 14:54:03 +0000 (18:54 +0400)]
Updated Russian translation
Luca Ferretti [Sat, 23 Jul 2011 12:52:13 +0000 (14:52 +0200)]
[l10n] Updated Italian translation
Vincent Untz [Thu, 21 Jul 2011 13:23:00 +0000 (15:23 +0200)]
GDesktopAppInfo: Add g_desktop_app_info_get_show_in()
Necessary for rebasing gnome-menus on top of GDesktopAppInfo.
https://bugzilla.gnome.org/show_bug.cgi?id=655044
Colin Walters [Fri, 22 Jul 2011 17:39:21 +0000 (13:39 -0400)]
glib.symbols: Update for expected ABI changes
Colin Walters [Fri, 22 Jul 2011 16:41:41 +0000 (12:41 -0400)]
glib-2.0.pc: Add -lrt to private libraries to assist static linking
https://bugzilla.gnome.org/show_bug.cgi?id=654078
Colin Walters [Fri, 22 Jul 2011 14:31:27 +0000 (10:31 -0400)]
Revert addition of g_key_file_has_key_full
Per IRC discussion, we can just ask bindings to use
g_key_file_get_value() to test for the existence of a key.
I left the "fixed" code in the source tree as static because it makes
more sense to me.
Behdad Esfahbod [Fri, 22 Jul 2011 14:33:47 +0000 (10:33 -0400)]
Don't use deprecated G_UNICODE_COMBINING_MARK
Ryan Lortie [Fri, 22 Jul 2011 13:47:11 +0000 (15:47 +0200)]
gitignore tweaks
Ryan Lortie [Fri, 22 Jul 2011 11:44:31 +0000 (13:44 +0200)]
Rename a few arguments
To fix header/function/doc-string mismatches.
Ryan Lortie [Fri, 22 Jul 2011 11:41:13 +0000 (13:41 +0200)]
Have gtk-doc ignore some #defines
We prefer to have gtk-doc pick up the function rather than the macro.
Ryan Lortie [Fri, 22 Jul 2011 11:25:32 +0000 (13:25 +0200)]
Replace @Varargs with @...
to make gtk-doc happy.
Ryan Lortie [Fri, 22 Jul 2011 10:59:44 +0000 (12:59 +0200)]
Remove duplicate 'struct real_pcre' declaration
The redundant forward declaration of 'struct real_pcre' before the
typedef was tripping up gtk-doc. Remove it.
Ryan Lortie [Fri, 22 Jul 2011 10:59:11 +0000 (12:59 +0200)]
Add deprecation guard to G_UNICODE_COMBINING_MARK
...since it's deprecated now and documented as such.
Ryan Lortie [Fri, 22 Jul 2011 10:57:59 +0000 (12:57 +0200)]
Reword some docs comments to avoid 'Returns ...'
If gtk-doc sees 'Returns ...' written in the text of the documentation
for a macro, it thinks that you are talking about the return value.
Avoid doing that and use 'Returns:' explicitly if we mean to.
Ryan Lortie [Fri, 22 Jul 2011 10:57:06 +0000 (12:57 +0200)]
move 'Since:' tags down
gtk-doc wants the Since: tag to be the absolute last thing in the docs
comment.
Xavier Claessens [Thu, 14 Jul 2011 08:51:06 +0000 (10:51 +0200)]
Add g_ptr_array_new_full
Fixes bug #654450
Ryan Lortie [Fri, 22 Jul 2011 07:48:34 +0000 (09:48 +0200)]
Bump the version
Ryan Lortie [Fri, 22 Jul 2011 06:55:35 +0000 (08:55 +0200)]
Release 2.29.14
Behdad Esfahbod [Thu, 21 Jul 2011 20:05:43 +0000 (16:05 -0400)]
Bug 655076 - Our normalization code misses some Full_Composition_Exclusion=True.
David Zeuthen [Thu, 21 Jul 2011 20:03:27 +0000 (16:03 -0400)]
gdbus-codegen: Add support for new org.gtk.GDBus.C.UnixFD annotation
Also add convenience _with_unix_fd_list variants to GDBusConnection,
GDBusProxy and GDBusMethodInvocation types to easily support this.
Signed-off-by: David Zeuthen <davidz@redhat.com>
Dan Williams [Wed, 20 Jul 2011 22:28:20 +0000 (17:28 -0500)]
Document how to free return values from g_key_file_get_*_list
Makes the docs consistent with the string list getters. Add
GOI annotations while we're at it.
Behdad Esfahbod [Thu, 21 Jul 2011 14:30:36 +0000 (10:30 -0400)]
Bug 655025 - #define G_UNICODE_SPACING_MARK G_UNICODE_COMBINING_MARK
David Zeuthen [Thu, 21 Jul 2011 13:43:59 +0000 (09:43 -0400)]
GDBus: Don't convert GVariant type 'h' to gint
The code-generator already uses GVariant* so generated code didn't
really work at all. We want that instead of gint to avoid confusion
because a 'h' instance is an _index_ into a GUnixFDList, not a file
descriptor.
Signed-off-by: David Zeuthen <davidz@redhat.com>
David Zeuthen [Thu, 21 Jul 2011 13:32:38 +0000 (09:32 -0400)]
gdbus-codegen: Use G_TYPE_STRV/gchar** for GVariant type 'ao'
This is possible now that we have better support for object path
arrays, see
http://git.gnome.org/browse/glib/commit/?id=
19878998bc386db78614f1c92ff8524a81479c7b
Note that this breaks the ABI of generated code but since
gdbus-codegen(1) has never yet been in a stable GLib release, this is
fine.
Signed-off-by: David Zeuthen <davidz@redhat.com>
Simon McVittie [Thu, 21 Jul 2011 09:22:01 +0000 (10:22 +0100)]
g_atomic_int_add: document that the return value is new
Bug: https://bugzilla.gnome.org/show_bug.cgi?id=654988
Reviewed-by: Matthias Clasen <mclasen@redhat.com>
Ryan Lortie [Thu, 21 Jul 2011 07:21:18 +0000 (09:21 +0200)]
Advise against using GPrivate
GStaticPrivate is really a lot better in almost every way.
Ryan Lortie [Thu, 21 Jul 2011 06:33:50 +0000 (08:33 +0200)]
GParam: try to avoid further invalid uses
In an attempt to avoid some potential future abuses of the GParamSpec
API, qualify the 'name' field of the structure as 'const' and add a
comment noting that it is an interned string.
This is a theoretical API break, but it will only ever result in
warnings -- and even then, only if you were already doing something
questionable.
Clean up some of the warnings that were caused internally in gparam.c
from these changes.
Behdad Esfahbod [Thu, 21 Jul 2011 02:11:08 +0000 (22:11 -0400)]
Add g_unicode_script_from_iso15924()
And adjust g_unicode_script_to_iso1592().
Benjamin Otte [Thu, 21 Jul 2011 00:56:05 +0000 (02:56 +0200)]
glocalfile: Fix compilation without USE_STATFS and USE_STATVFS
Behdad Esfahbod [Wed, 20 Jul 2011 23:04:29 +0000 (19:04 -0400)]
Bug 648271 - Add g_unicode_script_to_iso15924()
Add g_unicode_script_to_iso15924() and tests.
Daniel Mustieles [Wed, 20 Jul 2011 20:32:38 +0000 (22:32 +0200)]
Updated Spanish translation
Daniel Mustieles [Wed, 20 Jul 2011 20:26:59 +0000 (22:26 +0200)]
Updated Spanish translation
Ryan Lortie [Wed, 20 Jul 2011 19:49:16 +0000 (21:49 +0200)]
Change order of GFormatSizeFlags
David requested that I change the order of the flags.
Also, assign numerical values to the flags in the usual way. This
wasn't a bug yet, but only by chance.
Ryan Lortie [Wed, 20 Jul 2011 19:47:00 +0000 (21:47 +0200)]
g_format_size: just use GString
Matthias wasn't too impressed by the homebrew stack-allocated string building I
was doing.
Switch to GString.
Ryan Lortie [Wed, 20 Jul 2011 19:30:03 +0000 (21:30 +0200)]
g_format_size: Return 'gchar' instead of 'char'
I meant to do that in the first place.
Ryan Lortie [Wed, 20 Jul 2011 17:58:43 +0000 (19:58 +0200)]
[ast, es, fr, nn] Update byte unit translations
The Asturian, French, Norwegian Nynorsk and Spanish translations
incorrectly translated "MB" and friends to "MiB" (french: "Mio"), etc.
This was in response to the incorrect use of "MB" in the (now
deprecated) g_format_size_for_display() function.
These strings are now used (correctly) in g_format_size(), so I have
updated the translations accordingly.
Additionally, the Norwegian Nynorsk translation was incorrectly
translating several larger units to "KiB", so that has been corrected as
well.
Ryan Lortie [Wed, 20 Jul 2011 17:44:39 +0000 (19:44 +0200)]
Change GLib size units policy
This commit changes GLib size units policy. We now prefer SI units and
allow for use of proper IEC units where desired.
g_format_size_for_display() which incorrectly mixed IEC units with SI
suffixes is left unmodified, but has been deprecated.
g_format_size() has been introduced which uses SI units and suffixes.
g_format_size_full() has also been added which takes a flags argument to
allow for use of IEC units (with correct suffixes). It also allows for
a "long format" output which includes the total number of bytes. For
example: "238.5 MB (238,472,938 bytes)".
Ryan Lortie [Wed, 20 Jul 2011 15:30:56 +0000 (17:30 +0200)]
g_format_size_for_display: some internal renames
Rename the size constants from KILOBYTE to KIBIBYTE (etc.) since that's
what they really are.
This is a strictly internal change with no externally-visible effect in
terms of API or functionality.
Giovanni Campagna [Thu, 30 Jun 2011 23:59:13 +0000 (01:59 +0200)]
gdbusintrospection: fix introspection annotations
g_dbus_interface_info_lookup_* were incorrectly considered
(transfer full) by introspected bindings, and this caused memory
corruptions.
https://bugzilla.gnome.org/show_bug.cgi?id=622921
Vincent Untz [Wed, 20 Jul 2011 07:22:12 +0000 (09:22 +0200)]
Stop using deprecated g_unicode_canonical_decomposition()
https://bugzilla.gnome.org/show_bug.cgi?id=654948
Ryan Lortie [Wed, 20 Jul 2011 14:34:55 +0000 (16:34 +0200)]
GVariant: document avoiding g_variant_iter_loop
The choice between g_variant_iter_next() and g_variant_iter_loop() is a
bit confusing for some people. Add a note to the documentation of
g_variant_iter_loop() to clarify that it should be avoided except in a
few specific cases.
Ryan Lortie [Wed, 20 Jul 2011 11:11:19 +0000 (13:11 +0200)]
GVariant: better support for object path arrays
Add G_VARIANT_TYPE_OBJECT_PATH_ARRAY along with accessor functions
g_variant_new_objv, g_variant_get_objv and g_variant_dup_objv. Also add
support for '^ao' and '^a&o' format strings for g_variant_new() and
g_variant_get().
https://bugzilla.gnome.org/show_bug.cgi?id=654955
Ryan Lortie [Wed, 20 Jul 2011 12:04:52 +0000 (14:04 +0200)]
GSettings: don't abort on missing schemas
Give a g_critical instead.
Ryan Lortie [Wed, 20 Jul 2011 08:52:07 +0000 (10:52 +0200)]
Add some blacklisted mount directories
Another long-carried Debian patch. No reason that we should not have
these few extra items in the (already long) upstream list.
Ryan Lortie [Wed, 20 Jul 2011 08:49:48 +0000 (10:49 +0200)]
wait longer in threadpool test case
If we are going to fail an assert, wait for a bit longer before doing so
(up to 5 seconds, if needed).
This is a long-standing Debian patch to fix build failures on really
slow machines.
Matthias Clasen [Wed, 20 Jul 2011 03:04:04 +0000 (23:04 -0400)]
post-release bump
Matthias Clasen [Wed, 20 Jul 2011 03:02:05 +0000 (23:02 -0400)]
2.29.12
Matthias Clasen [Wed, 20 Jul 2011 00:40:28 +0000 (20:40 -0400)]
Drop the warnings.sgml template
Holger Berndt [Thu, 3 Jun 2010 01:22:36 +0000 (03:22 +0200)]
Document how to unset attributes
Unsetting won't work on all attributes. Currently, only
metadata attributes are supported.
https://bugzilla.gnome.org/show_bug.cgi?id=620423
Johan Dahlin [Tue, 19 Jul 2011 17:18:10 +0000 (14:18 -0300)]
Pass in NULL instead of g_cclosure_marshal_generic
NULL is now a shortcut for g_cclosure_marshal_generic, so avoid
referencing it directly.
https://bugzilla.gnome.org/show_bug.cgi?id=654917
Johan Dahlin [Tue, 12 Jul 2011 16:39:45 +0000 (13:39 -0300)]
Use a generic marshaller if one is not specified
Since g_cclosure_marshal_generic is always enabled, it makes
sense to always use that instead of using generated ones.
https://bugzilla.gnome.org/show_bug.cgi?id=654917
Ryan Lortie [Tue, 19 Jul 2011 14:12:30 +0000 (16:12 +0200)]
GSettings: remove key length restrictions
The key length now stands effectively unlimited at 1024 characters.
https://bugzilla.gnome.org/show_bug.cgi?id=654536
Matthias Clasen [Tue, 19 Jul 2011 04:02:57 +0000 (00:02 -0400)]
Fix doc formatting errors
Matthias Clasen [Tue, 19 Jul 2011 03:58:32 +0000 (23:58 -0400)]
Move GError docs inline and ditch template
Matthias Clasen [Tue, 19 Jul 2011 03:23:17 +0000 (23:23 -0400)]
Move Unicode docs inline, ditch template
Matthias Clasen [Tue, 19 Jul 2011 03:00:09 +0000 (23:00 -0400)]
Remove queue.sgml
Behdad Esfahbod [Mon, 18 Jul 2011 22:00:40 +0000 (18:00 -0400)]
Simplify Hangul Jamo decomposition
The algorithm is not copy/paste from Unicode anymore, but it's easy
enough to follow the logic.
Behdad Esfahbod [Mon, 18 Jul 2011 21:52:40 +0000 (17:52 -0400)]
Bug 654651 - Better g_unicode_canonical_decomposition()
Add g_unichar_fully_decompose().
Deprecate g_unicode_canonical_decomposition().
Patrick Welche [Mon, 18 Jul 2011 16:58:01 +0000 (17:58 +0100)]
Avoid failing arguments to statfs() test on systems which use statvfs.
- move choice of statfs vs statvfs from gio/glocalfile.c to configure.ac
- if statvfs is the choice, then don't check number of arguments to statfs()
- use choice in gio/gunixmounts.c as well
https://bugzilla.gnome.org/show_bug.cgi?id=617949
Benjamin Otte [Mon, 18 Jul 2011 12:10:35 +0000 (14:10 +0200)]
gtester: Count inability to run a test in a binary as a failure
Previously, when a binary did not run a single test - for whatever
reason, from the binary not existing to the binary not using the gtester
framework - no failures were recorded. Now we record a non-successful
run of a binary that did not start any tests as a failure, too.
Note that this does not change the behavior of any binaries that exit
successfully or that report the start of any gtester test.
Benjamin Otte [Mon, 18 Jul 2011 11:39:36 +0000 (13:39 +0200)]
gtester: Return a failure exit code when a test fails
Previously, gtester used the testcase_fail_count as the number of tests
that failed in the latest run testcase, but then use that as the return
value of main().
Now we count the failed testcases of the whole run.
Kjartan Maraas [Mon, 18 Jul 2011 11:13:11 +0000 (13:13 +0200)]
Updated Norwegian bokmål translation