OSDN Git Service
Dan Winship [Sun, 22 Apr 2012 19:20:14 +0000 (15:20 -0400)]
gio: fix error handling in async case of GProxyAddressEnumerator
In the async case, a failed DNS lookup was causing the proxy
resolution to bail out immediately, rather than just moving on to the
next potential proxy (which might not need us to do the DNS lookup
beforehand). Fix that.
Sasi Bhushan [Mon, 14 May 2012 08:02:06 +0000 (13:32 +0530)]
Updated Telugu Translations
Jiro Matsuzawa [Sun, 13 May 2012 10:22:24 +0000 (19:22 +0900)]
[l10n] Update Japanese translation
Christian Kirbach [Tue, 8 May 2012 17:15:09 +0000 (19:15 +0200)]
[l10n] Updated German translation
Jiro Matsuzawa [Tue, 8 May 2012 11:24:53 +0000 (20:24 +0900)]
[l10n] Update Japanese translation
Alexander Shopov [Sat, 28 Apr 2012 04:55:20 +0000 (07:55 +0300)]
Updated Bulgarian translation
Gil Forcada [Sun, 6 May 2012 14:20:22 +0000 (16:20 +0200)]
[l10n] Updated Catalan translation
Matej Urbančič [Fri, 4 May 2012 11:42:36 +0000 (13:42 +0200)]
Updated Slovenian translation
Marek Černocký [Fri, 4 May 2012 07:32:11 +0000 (09:32 +0200)]
Updated Czech translation
Chun-wei Fan [Wed, 2 May 2012 03:06:14 +0000 (11:06 +0800)]
Fix GIO/GObject Visual C++ projects
-Make up for the missed DLL_EXPORT-it's actually needed for all GLib DLL
builds, omitting this caused problems to surface due to recent works to
make GDBus work on Windows.
-Also use the FFI_BULIDING macro for GObject builds as the suggessted
workaround for using static LibFFI builds (as we do now)-please see
ffi.h.in. This will fix the build of GObject against LibFFI 3.0.11, but
it is probable that this will change at some point for LibFFI.
Michael Biebl [Tue, 1 May 2012 16:42:12 +0000 (18:42 +0200)]
configure: Reset LIBS after ELF check
If action-if-found is not specified, AC_CHECK_LIB will append the library
to LIBS. As we don't want to link everything against libelf, reset LIBS
after doing the checks.
Rafał Mużyło [Sun, 15 Apr 2012 15:17:07 +0000 (11:17 -0400)]
configure: Fix typo in ELF check
Signed-off-by: Colin Walters <walters@verbum.org>
Ryan Lortie [Tue, 1 May 2012 05:18:04 +0000 (01:18 -0400)]
bump version
Ryan Lortie [Tue, 1 May 2012 05:04:49 +0000 (01:04 -0400)]
GLib 2.32.2
Alexander Larsson [Thu, 26 Apr 2012 08:05:41 +0000 (10:05 +0200)]
Dist gio/dbus-daemon.xml and fix dependency for builddir != srcdir
Cosimo Cecchi [Mon, 16 Apr 2012 14:45:12 +0000 (10:45 -0400)]
tests/date: force English locale running the GDateTime tests
The parsing test needs to make some assumption about the locale
representation of the string to be parsed, so we need to explicitly
override the locale here.
Ryan Lortie [Sat, 3 Mar 2012 16:31:05 +0000 (11:31 -0500)]
GApplication: allow null application_id
GApplication application ID is now permitted to be NULL, in which case
G_APPLICATION_NON_UNIQUE will be implicitly enabled.
https://bugzilla.gnome.org/show_bug.cgi?id=671249
Bruno Brouard [Mon, 30 Apr 2012 19:54:22 +0000 (21:54 +0200)]
Updated French translation
Xavier Claessens [Wed, 25 Apr 2012 13:02:25 +0000 (15:02 +0200)]
Fix g_clear_object macro with C++ compilers
In C++ it is invalid to cast a void* to void**.
Also add a static check to ensure
sizeof(*object_ptr) == sizeof (gpointer).
This avoid common mistake of missing '&'
in g_clear_object(&obj);
https://bugzilla.gnome.org/show_bug.cgi?id=674634
Fran Diéguez [Mon, 30 Apr 2012 14:22:09 +0000 (16:22 +0200)]
Updated Galician translations
Luca Ferretti [Mon, 30 Apr 2012 13:41:08 +0000 (15:41 +0200)]
l10n: Updated Italian translation
Мирослав Николић [Sun, 29 Apr 2012 10:08:04 +0000 (12:08 +0200)]
Updated Serbian translation
Piotr Drąg [Sat, 28 Apr 2012 16:03:13 +0000 (18:03 +0200)]
Updated Polish translation
Jonh Wendell [Sat, 28 Apr 2012 13:11:25 +0000 (10:11 -0300)]
Updated Brazilian Portuguese translation
Daniel Mustieles [Sat, 28 Apr 2012 09:30:41 +0000 (11:30 +0200)]
Updated Spanish translation
Aurimas Černius [Fri, 27 Apr 2012 19:17:10 +0000 (22:17 +0300)]
Updated Lithuanian translation
Debarshi Ray [Mon, 16 Apr 2012 21:00:01 +0000 (23:00 +0200)]
GAppInfo: overwrite the DISPLAY only if it is set in the launch context
If the launch context is a GAppLaunchContext, and not a
GdkAppLaunchContext, then g_app_launch_context_get_display will return
NULL because the get_display virtual method is undefined. The DISPLAY
might still be inherited from the parent process, in which case
overwriting it with NULL breaks the launch.
This is a regression introduced in:
de834bed306565c0652050665eafff4dfcdf0d8b
Fixes: https://bugzilla.gnome.org/672786
Michael Olbrich [Wed, 25 Apr 2012 07:46:35 +0000 (09:46 +0200)]
make sure configure fails if AC_CHECK_ALIGNOF cannot detect the alignment
When cross-compiling with gcc >= 4.5 AC_CHECK_ALIGNOF fails to detect the
correct alignment. Without a previous AC_CHECK_TYPE for the same type, the
alignment is silently set to '0'.
This makes sure that configure fails and reports the problem.
Signed-off-by: Michael Olbrich <m.olbrich@pengutronix.de>
https://bugzilla.gnome.org/show_bug.cgi?id=674483
Alexander Larsson [Tue, 24 Apr 2012 12:53:49 +0000 (14:53 +0200)]
win32: Fix build with srcdir != builddir
Missed this part in the last commit
(cherry picked from commit
62905cda0b42163bbd582278260f659f94e4d2bf)
Alexander Larsson [Tue, 24 Apr 2012 12:03:38 +0000 (14:03 +0200)]
win32: Fix build of gdbus-daemon-generated.[ch]
gdbus-daemon-generated.[ch] failed to build because it depended
on gdbus-2.0/codegen/gdbus-codegen which was build during the SUBDIRS part
of the build, however SUBDIRS are done *after* processing BUILT_SOURCES,
and these files are in BUILT_SOURCES.
The fix is simple, instead of running the gdbus-codegen code we
run the gdbus-codegen.in code, which works fine for uninstalled execution.
I also removed Makefile from the dependencies to avoid rebuilding the file
in tarballs, as Makefiles are written at configure time. We should be able to
ship the prebuilt files in the tarballs.
When running uninstalled
(cherry picked from commit
88bfc9b289e2b40bd7bff94dcdb06eb0b7b53f67)
Chun-wei Fan [Mon, 23 Apr 2012 16:03:33 +0000 (00:03 +0800)]
Visual C++ projects: Clean/fix up
Clean/fix up the Preprocessor Definitions for the various projects, where
we purge out the unneeded macros and add _DEBUG to the Debug builds of
various projects that somehow lacked this.
This will also fix the GIO build under Visual C++ 2008, as the _DEBUG macro
in the release builds will cause a debug entry to appear in its manifest
file during the build, which will cause GIO-using applications to fail
to run on systems not running Visual C++/Studio 2008 due to its embedding
of a badly-generated manifest file.
Alexandre Rostovtsev [Mon, 16 Apr 2012 02:57:25 +0000 (22:57 -0400)]
Do not detect GNUstep as Cocoa
GNUstep also installs Foundation/Foundation.h
https://bugzilla.gnome.org/show_bug.cgi?id=674172
Signed-off-by: William Hua <william@attente.ca>
William Hua [Wed, 18 Apr 2012 03:49:47 +0000 (23:49 -0400)]
Use Cocoa header for Cocoa test instead of Foundation.
Alexander Larsson [Tue, 17 Apr 2012 13:50:01 +0000 (15:50 +0200)]
Fix build on win32
Some errnos don't exist on win32
(cherry picked from commit
2414d7d6a197e0c4b184607160d6cd8a378df177)
Alexander Larsson [Fri, 20 Apr 2012 08:23:34 +0000 (10:23 +0200)]
win32: Implement _g_dbus_get_machine_id using machine guid
This is what libdbus uses, so we're compatible.
(cherry picked from commit
5fb44f70dcac16ff6f13d3d36c330ddd617a5d97)
Alexander Larsson [Wed, 18 Apr 2012 15:09:37 +0000 (17:09 +0200)]
win32: Support autolaunching dbus daemon
(cherry picked from commit
dccce383670592f12510019e92ea91a618a1810c)
Alexander Larsson [Wed, 18 Apr 2012 14:25:29 +0000 (16:25 +0200)]
Add _g_io_win32_get_module to get the gio HMODULE
(cherry picked from commit
bd148e127a5863de4aef260cd879e0eda370c711)
Alexander Larsson [Mon, 16 Apr 2012 08:16:04 +0000 (10:16 +0200)]
Add GDBusDAaemon, an implementation of a message bus
This is mostly complete, sans support for activation. However, its
not as picky as the libdbus implementation in terms like validation
and limits checking, nor is it as tested.
Its can be useful to test gdbus if dbus-daemon is not availible, but
its main reason for existance is to implement a default session bus
on win32 so that e.g. GApplication is guaranteed to work.
(cherry picked from commit
25581738a8d5ee7db2a6d9ebd908d59b2837cd70)
Alexander Larsson [Tue, 17 Apr 2012 13:50:53 +0000 (15:50 +0200)]
gdbus: Escape nonce files in dbus addressess
Otherwise the colon in c:\blah made for trouble
(cherry picked from commit
82aecce30131e275dff73f23ee91e7c40a9b3f03)
Alexander Larsson [Mon, 16 Apr 2012 07:55:29 +0000 (09:55 +0200)]
Support initial underscores in dbus codegen namespace
Before these were considered lowercase and thus got duplicated.
(cherry picked from commit
ff92fe95935b38eaacedb287f6600df000ed1c1a)
Benjamin Otte [Fri, 20 Apr 2012 12:31:05 +0000 (14:31 +0200)]
resourcefile: Set display name
Fixes resource filenames not being printed in CSS warnings for GTK.
https://bugzilla.gnome.org/show_bug.cgi?id=674345
Hannes Mueller [Sat, 4 Feb 2012 09:06:35 +0000 (10:06 +0100)]
avoid warning in gutils.h when using gcc with -Wconversion
Old solution to avoid warning with gcc flag -Wconversion does not
work with gcc >=4.3 since the behaviour for that flag has been
changed, ref. http://gcc.gnu.org/wiki/NewWconversion. Now cast for
__builtin_clzl is required, which itself is declared to return int.
https://bugzilla.gnome.org/show_bug.cgi?id=619026
Christian Persch [Mon, 9 Apr 2012 13:34:10 +0000 (15:34 +0200)]
resources: compiler: Fix resources on big endian architectures
Resources are always little endian, so the gvdb is byteswapped. When looking
up the value, it would return a new byteswapped variant, making the data
returned from do_lookup() invalid once that variant is unref'd. Since
byteswapping doesn't matter for the "ay" data anyway, just use
gvdb_table_get_raw_value() instead and only byteswap the length and flag
values.
https://bugzilla.gnome.org/show_bug.cgi?id=673409
(cherry picked from commit
5ef34e5f11b4f71a66615b6506108ee540cc3420)
Colin Walters [Mon, 16 Apr 2012 15:50:05 +0000 (11:50 -0400)]
tests/mainloop: Add a test for recursive child sources
https://bugzilla.gnome.org/show_bug.cgi?id=669260
Dan Winship [Fri, 13 Apr 2012 15:44:34 +0000 (11:44 -0400)]
tests/mainloop: add a test for child sources
https://bugzilla.gnome.org/show_bug.cgi?id=669260
Dan Winship [Wed, 11 Apr 2012 19:21:17 +0000 (15:21 -0400)]
gmain: block child sources when blocking the parent
When blocking a source that has child sources, we need to consider the
children blocked as well. Otherwise they will still trigger repeatedly
in an inner loop started from the parent source's callback.
https://bugzilla.gnome.org/show_bug.cgi?id=669260
Piotr Drąg [Mon, 16 Apr 2012 19:19:23 +0000 (21:19 +0200)]
Updated Polish translation
Kalev Lember [Sat, 14 Apr 2012 12:48:13 +0000 (15:48 +0300)]
Only build gmenumodel test on unix
It depends on gdbus-sessionbus.c which only builds on unix.
Kalev Lember [Sat, 14 Apr 2012 12:40:17 +0000 (15:40 +0300)]
tests/fileutils: Fix Windows build
Yinghua Wang [Mon, 16 Apr 2012 06:06:56 +0000 (14:06 +0800)]
update Simplified Chinese (zh_CN) translation
David Zeuthen [Sat, 14 Apr 2012 16:29:30 +0000 (12:29 -0400)]
GSocketControlMessage: Don't warn about unknown messages
If we do this, the tests added in bug 673943 will cause warnings.
Signed-off-by: David Zeuthen <davidz@redhat.com>
Matthias Clasen [Fri, 13 Apr 2012 23:54:42 +0000 (19:54 -0400)]
Bump version
Matthias Clasen [Fri, 13 Apr 2012 23:53:42 +0000 (19:53 -0400)]
2.32.1
Matthias Clasen [Fri, 13 Apr 2012 22:51:23 +0000 (18:51 -0400)]
Fix distcheck
I forgot to dist a test script, and forgot to look for it
in srcdir instead of builddir.
Christian Kirbach [Fri, 13 Apr 2012 17:39:19 +0000 (19:39 +0200)]
[l10n] Updated German translation
Dan Winship [Fri, 13 Apr 2012 15:43:09 +0000 (11:43 -0400)]
tests/mainloop: fix a race condition
Rather than depending on the machine's speed/load, just interlock
between the two threads properly.
Dan Winship [Fri, 13 Apr 2012 15:42:39 +0000 (11:42 -0400)]
tests/mainloop: use g_assert_cmpint(), for better error messages
Matthias Clasen [Fri, 13 Apr 2012 13:31:08 +0000 (09:31 -0400)]
Make the markup-parse test independent of the locale
I added a setlocale call, because we need it for Unicode to
come out right; but I forgot to fix the locale, so we now
fail when comparing error messages to the expected (English)
result. Correct this by setting LANG explicitly to en_US.utf-8.
https://bugzilla.gnome.org/show_bug.cgi?id=669285
Bruce Cowan [Fri, 13 Apr 2012 12:27:39 +0000 (13:27 +0100)]
Updated British English translation
Kalev Lember [Fri, 13 Apr 2012 10:22:46 +0000 (13:22 +0300)]
GResource docs: fix typo
Ryan Lortie [Fri, 13 Apr 2012 00:04:32 +0000 (20:04 -0400)]
GSettings docs: clarify what is a good path
Add an explicit note to the docs about choosing paths based on domain
names, not ones like "/apps/", "/desktop/" or "/system/".
Ryan Lortie [Thu, 12 Apr 2012 23:55:34 +0000 (19:55 -0400)]
glib-compile-schemas: warn about bad dconf paths
For quite some time the recommended usage of GSettings and dconf has
been to use paths like /org/gnome/example/. Use of /apps/ has spilled
over from GConf and is continuing to make its way into a number of
applications as they port.
glib-compile-schemas will now warn about these types of paths being
used. This generates a lot of noise, but hopefully it will reduce the
number of ported applications making this mistake.
Alexander Larsson [Thu, 12 Apr 2012 14:50:20 +0000 (16:50 +0200)]
Fall back to SO_PEERCRED if credentials passing fails
Turns out libdbus doesn't send struct ucred credentials on linux, but
just relies on the SO_PEERCRED support. However, gdbus does send, and
expect to recieve a ucred credential. So, when libdbus talks to a
gdbus server the authentication fails to send the credentials.
We fix this by falling back to g_socket_get_credentials() if we don't
get any credential messages.
Alexander Larsson [Thu, 12 Apr 2012 14:43:49 +0000 (16:43 +0200)]
Detect "empty" socket credentials on Linux
Linux uses struct ucred to pass over socket credentials. Historically
this has always worked in recievemsg, if SO_PASSCRED was set on the socket,
even if the remote side didn't pass any credits. But this change broke that:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=
16e5726269611b71c930054ffe9b858c1cea88eb;hp=
a9e9fd7182332d0cf5f3e601df3e71dd431b70d7
However, it doesn't actually fail getting the credentials, it just returns
an "empty" one, as initialized by cred_to_ucred() at:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=net/core/sock.c;h=
b29ab61b029cf7f76fda992ecfcb8dcaa06b0483;#l756
So, we detect this and fail the credentials reading.
This actually happened in real life with gdbus acting as a server, as
gdbus expected an ucred but libdbus didn't send one.
William Hua [Thu, 12 Apr 2012 03:25:09 +0000 (23:25 -0400)]
Be more precise about Carbon v. Cocoa.
OKANO Takayoshi [Thu, 12 Apr 2012 00:30:05 +0000 (09:30 +0900)]
[l10n] Update Japanese translation
Kalev Lember [Wed, 11 Apr 2012 13:41:48 +0000 (16:41 +0300)]
gio-2.0.pc: Avoid full path to executables
Instead of using full path in glib_compile_schemas and
glib_compile_resources variables, rely on having the executables in
PATH.
This fixes the cross-compiling case where we cannot execute the binaries
for target system. Instead of executing the target system's binaries, we
need to use the native versions installed on the build host. The easiest
way to find the native executables is to just pick them up from PATH.
In addition, this brings gio-2.0.pc in line with glib-2.0.pc -- the
latter has historically only listed the executable name and not the full
path.
https://bugzilla.gnome.org/show_bug.cgi?id=673911
Jeremy Huddleston [Wed, 11 Apr 2012 06:05:00 +0000 (02:05 -0400)]
Don't use fast enumeration in old version of Mac OS.
Signed-off-by: William Hua <william@attente.ca>
Jeremy Huddleston [Wed, 11 Apr 2012 05:44:29 +0000 (01:44 -0400)]
Link with Foundation framework on OSX.
Signed-off-by: William Hua <william@attente.ca>
Bruno Brouard [Wed, 11 Apr 2012 12:53:22 +0000 (14:53 +0200)]
Updated French translation
Chandan Kumar (ciypro) [Wed, 11 Apr 2012 10:55:22 +0000 (16:25 +0530)]
Updated HINDI translation
Kjartan Maraas [Wed, 11 Apr 2012 10:41:51 +0000 (12:41 +0200)]
Updated Norwegian bokmål translation
Matthias Clasen [Wed, 11 Apr 2012 03:52:30 +0000 (23:52 -0400)]
Updates
Carles Ferrando [Tue, 10 Apr 2012 21:12:29 +0000 (23:12 +0200)]
[l10n]Updated Catalan (Valencian) translation
Jordi Serratosa [Tue, 10 Apr 2012 21:12:23 +0000 (23:12 +0200)]
[l10n] Fixes on Catalan translation
Will Thompson [Fri, 6 Apr 2012 09:03:29 +0000 (10:03 +0100)]
gdbus: test case for 673612
When presented with an array of empty arrays of 8-byte-aligned types,
GDBus would incorrectly apply the 8-byte alignment when reading back.
https://bugzilla.gnome.org/show_bug.cgi?id=673612
Signed-off-by: David Zeuthen <davidz@redhat.com>
Will Thompson [Tue, 10 Apr 2012 12:20:09 +0000 (13:20 +0100)]
GDBusMessage: do not align for grandchildren of empty arrays.
D-Bus arrays are serialized as follows:
1. align to a 4-byte boundary (for the length)
2. uint32: the length of the serialized body in bytes
3. padding for the alignment of the body type (not included in the length)
4. the body.
Note that 3. is a no-op unless the body type is an 8-byte aligned type
(uint64, int64, double, struct, dict_entry), since you are always on a
4-byte boundary from aligning and writing the length.
So, an empty aax (that is, an array containing zero arrays of int64)
is serialized as follows:
1. align to a 4-byte boundary
2. length of the contents of this (empty) array, in bytes (0)
3. align to a 4-byte boundary (the child array's alignment requirement)
4. there is no body.
But previously, GDBus would recurse in step three to align not just for
the type of the child array, but for the nonexistent child array's
contents. This only affects the algorithm when the grandchild type has
8-byte alignment and the reader happened to not already be on an 8-byte
boundary, in which case 4 bytes were spuriously skipped.
https://bugzilla.gnome.org/show_bug.cgi?id=673612
Signed-off-by: David Zeuthen <davidz@redhat.com>
Will Thompson [Tue, 10 Apr 2012 11:58:52 +0000 (12:58 +0100)]
GDBusMessage: print more debug info about alignment
Signed-off-by: David Zeuthen <davidz@redhat.com>
Yaron Shahrabani [Tue, 10 Apr 2012 17:06:51 +0000 (20:06 +0300)]
Updated Hebrew translation.
Мирослав Николић [Tue, 10 Apr 2012 09:06:03 +0000 (11:06 +0200)]
Updated Serbian translation
Jasper St. Pierre [Mon, 9 Apr 2012 23:07:13 +0000 (20:07 -0300)]
gclosure: Support return values of GVariants
https://bugzilla.gnome.org/show_bug.cgi?id=673803
Cosimo Cecchi [Mon, 9 Apr 2012 19:47:29 +0000 (15:47 -0400)]
tests: always use four digit years for date parsing tests
Or some system with different locale settings might get confused whether
a two digit year is to be parsed with regard to the current century or
as an absolute year.
Aurimas Černius [Mon, 9 Apr 2012 18:52:10 +0000 (21:52 +0300)]
Updated Lithuanian translation
Piotr Drąg [Mon, 9 Apr 2012 15:24:00 +0000 (17:24 +0200)]
Updated POTFILES.skip
Fran Diéguez [Sun, 8 Apr 2012 19:56:22 +0000 (21:56 +0200)]
Updated Galician translations
Matthias Clasen [Sun, 8 Apr 2012 14:23:16 +0000 (10:23 -0400)]
binding: Improve test coverage
Matthias Clasen [Sun, 8 Apr 2012 14:21:50 +0000 (10:21 -0400)]
spawn: Improve test coverage
Add a test that excercises the script execution code.
Unfortunately, much of this code only runs in the forked
child, and therefore its execution does not get caught
by gcov.
Matthias Clasen [Sun, 8 Apr 2012 14:21:27 +0000 (10:21 -0400)]
mainloop: Improve test coverage
Matthias Clasen [Sun, 8 Apr 2012 14:20:46 +0000 (10:20 -0400)]
settings: Improve test coverage
Matthias Clasen [Sun, 8 Apr 2012 14:20:10 +0000 (10:20 -0400)]
gdbusserver: Improve test coverage
Matthias Clasen [Sun, 8 Apr 2012 14:19:20 +0000 (10:19 -0400)]
icon: Improve test coverage
Matthias Clasen [Sun, 8 Apr 2012 14:17:32 +0000 (10:17 -0400)]
contenttype: Improve test coverage
Excercise more parts of the guessing machinery.
These tests rely on the mime database being present.
Matthias Clasen [Sun, 8 Apr 2012 14:17:04 +0000 (10:17 -0400)]
networkaddress: Improve test coverage
Matthias Clasen [Sun, 8 Apr 2012 14:15:29 +0000 (10:15 -0400)]
networkaddress: Fix userinfo parsing in uris
The code that is checking the userinfo part was accidentally
given a pointer to the end of the userinfo, so it was not
checking the right portion of the string at all.
Matthias Clasen [Sat, 7 Apr 2012 15:32:36 +0000 (11:32 -0400)]
utils: Improve test coverage
Add a test for some of our gettext variants.
Matthias Clasen [Sat, 7 Apr 2012 15:28:42 +0000 (11:28 -0400)]
fileutils: Improve test coverage
Add tests for corner cases of g_path_get_basename, and
for the errno-to-file-error conversion.
Matthias Clasen [Sat, 7 Apr 2012 15:28:04 +0000 (11:28 -0400)]
sort: Improve test coverage
Test the code paths that sort 'large' items.
Matthias Clasen [Sat, 7 Apr 2012 15:27:15 +0000 (11:27 -0400)]
date: Improve test coverage
Add tests that exercise g_date_order, g_date_compare, g_date_clamp,
and more of the date parsing functions.
Matthias Clasen [Sat, 7 Apr 2012 15:23:47 +0000 (11:23 -0400)]
convert: Improve test coverage
Add a test that excercises the 'no conversion' code path.
This uncovered that we don't treat errno properly in this path,
and as a consequence, the returned error code is unreliable.
Matthias Clasen [Sat, 7 Apr 2012 15:22:13 +0000 (11:22 -0400)]
collate: Improve test coverage
Add some strings that exercise the non-ASCII parts of our
collation support.