Matthias Clasen [Thu, 2 Jan 2014 04:28:23 +0000 (23:28 -0500)]
Make symlink test work installed
We can't assume that the location used for G_TEST_DIST paths
is writable, so just create the symlink in the current directory
instead.
Matthias Clasen [Thu, 2 Jan 2014 03:39:32 +0000 (22:39 -0500)]
Drop locale-dependent timeval tests
Matthias Clasen [Thu, 2 Jan 2014 03:30:24 +0000 (22:30 -0500)]
Make /contenttype/tree test work installed
The x-content/win32-software type is only recognized if
the autorun.exe file is executable. Since the file is installed
as data, we need to fix up its permissions in an
install-data-hook.
Matthias Clasen [Thu, 2 Jan 2014 03:01:15 +0000 (22:01 -0500)]
Avoid a compiler warning
Matthias Clasen [Thu, 2 Jan 2014 02:39:51 +0000 (21:39 -0500)]
Make timeval tests independent of the environment
Some of the tested formats are locale-dependent, so unset
the TZ environment variable before testing them.
Chun-wei Fan [Thu, 2 Jan 2014 01:54:18 +0000 (09:54 +0800)]
MSVC 2012 Projects: Rename a Property Sheet
Forgot to update the property sheet file name from the Visual Studio 2010
update... Sorry!
Ryan Lortie [Tue, 29 Oct 2013 18:11:28 +0000 (11:11 -0700)]
Fix races in unix signal dispatch
Fix some races introduced in
be2c7b83c4a9c9d3aa76b1499c27ab19e0f4e470
while keeping the property that multiple handlers for the same unix
signal all get dispatched.
Also fix the behaviour of the source checking for pending signals when
it's created. No matter what we do here (clear the pending flag or not)
there is something that can go wrong.
If we clear the flag, we may prevent other sources from being
dispatched. If we don't clear it, we may end up dispatching the same
source twice (if we manage to dispatch it from its own thread before the
GLib worker has a chance to run).
Instead, run the full dispatch procedure when a new source is added. It
actually doesn't matter what thread this runs in since the lock is held.
https://bugzilla.gnome.org/show_bug.cgi?id=711090
Colin Walters [Thu, 2 Jan 2014 00:05:34 +0000 (19:05 -0500)]
gio/tests: Fix usage of BUILT_SOURCES
Yosef Or Boczko [Wed, 1 Jan 2014 23:40:26 +0000 (01:40 +0200)]
Fix doc build
Matthias Clasen [Wed, 1 Jan 2014 22:42:33 +0000 (17:42 -0500)]
Improve gutils.c test coverage
Matthias Clasen [Wed, 1 Jan 2014 22:41:59 +0000 (17:41 -0500)]
Improve GDate test coverage
Matthias Clasen [Tue, 31 Dec 2013 14:19:50 +0000 (09:19 -0500)]
Improve GIOChannel test coverage
Matthias Clasen [Tue, 31 Dec 2013 14:19:22 +0000 (09:19 -0500)]
Improve hostutils test coverage
Matthias Clasen [Tue, 31 Dec 2013 14:18:50 +0000 (09:18 -0500)]
Improve GVariant test coverage
Matthias Clasen [Tue, 31 Dec 2013 14:18:30 +0000 (09:18 -0500)]
Test stdio wrappers
Matthias Clasen [Tue, 31 Dec 2013 01:18:15 +0000 (20:18 -0500)]
Trivial typo fix
Matthias Clasen [Mon, 30 Dec 2013 15:55:07 +0000 (10:55 -0500)]
Improve charset test coverage
Matthias Clasen [Mon, 30 Dec 2013 15:54:46 +0000 (10:54 -0500)]
Add tests for g_malloc
Matthias Clasen [Mon, 30 Dec 2013 15:53:32 +0000 (10:53 -0500)]
Drop memory-related trap variables
These are just more lo-tech conditional breakpoint wannabes.
Debuggers can be trusted to support conditional breakpoints
nowadays.
Matthias Clasen [Mon, 30 Dec 2013 14:49:30 +0000 (09:49 -0500)]
Improve GAsyncQueue test coverage
Matthias Clasen [Mon, 30 Dec 2013 14:48:56 +0000 (09:48 -0500)]
Improve test coverage of g_utf8_collate_key_for_filename
The existing tests were accidentally using the same test data
twice. Fix that, and add another set of tests that exercise
the filename collation special cases.
Matthias Clasen [Mon, 30 Dec 2013 14:48:29 +0000 (09:48 -0500)]
Add tests for g_spaced_primes_closest
Matthias Clasen [Mon, 30 Dec 2013 14:48:10 +0000 (09:48 -0500)]
Improve GOptionContext test coverage
Matthias Clasen [Mon, 30 Dec 2013 14:47:49 +0000 (09:47 -0500)]
Improve GBookmarkFile test coverage
Matthias Clasen [Tue, 31 Dec 2013 17:45:09 +0000 (12:45 -0500)]
GBookmarkFile: Streamline error handling a bit
Matthias Clasen [Tue, 31 Dec 2013 17:48:05 +0000 (12:48 -0500)]
GBookmarkFile: Remove dead code
The icon element is actually parsed further up, and this
case is never hit, so remove it.
Matthias Clasen [Mon, 30 Dec 2013 14:45:53 +0000 (09:45 -0500)]
Make a bookmark testfile roundtrippable
This will be used in new tests in the following commit.
Matthias Clasen [Mon, 30 Dec 2013 13:57:04 +0000 (08:57 -0500)]
Trivial formatting fixes
Matthias Clasen [Sun, 29 Dec 2013 18:47:43 +0000 (13:47 -0500)]
Improve glib-unix test coverage
Matthias Clasen [Sun, 29 Dec 2013 18:46:00 +0000 (13:46 -0500)]
Improve GHashTable test coverage
Matthias Clasen [Sun, 29 Dec 2013 18:45:38 +0000 (13:45 -0500)]
Add a test for g_markup_parse_context_get_element_stack
Matthias Clasen [Sun, 29 Dec 2013 18:43:28 +0000 (13:43 -0500)]
Remove broken attempt at parsing date-only
g_time_val_from_iso8601 was attempting to parse strings
having only a date, but failed to actually set the timeval
despite returning TRUE. Since the docs state that the function
only parses strings containing a date and a time, just return
FALSE in this case.
Also remove an incomplete testcase for this behaviour that was
just checking the boolean return value, but not timeval.
Matthias Clasen [Sun, 29 Dec 2013 12:55:14 +0000 (07:55 -0500)]
Add missing includes
config.h needs to be included in every source file, first.
Matthias Clasen [Sun, 29 Dec 2013 12:23:19 +0000 (07:23 -0500)]
Improve mainloop test coverage
Matthias Clasen [Sun, 29 Dec 2013 12:23:01 +0000 (07:23 -0500)]
Add tests for GTimer and GTimeVal
Matthias Clasen [Thu, 26 Dec 2013 04:52:15 +0000 (23:52 -0500)]
Improve GTree test coverage
Matthias Clasen [Thu, 26 Dec 2013 04:51:58 +0000 (23:51 -0500)]
Improve logging test coverage
Matthias Clasen [Thu, 26 Dec 2013 04:51:26 +0000 (23:51 -0500)]
Improve testing test coverage
In particular, test the just-fixed g_test_trap_reached_timeout.
Matthias Clasen [Thu, 26 Dec 2013 04:50:37 +0000 (23:50 -0500)]
Fix g_test_trap_reached_timeout
The function was returning the opposite of what the
documentation says.
Matthias Clasen [Wed, 25 Dec 2013 04:35:31 +0000 (23:35 -0500)]
Improve GSlice test coverage
Matthias Clasen [Wed, 25 Dec 2013 04:33:28 +0000 (23:33 -0500)]
Improve GNode test coverage
Matthias Clasen [Wed, 25 Dec 2013 04:26:25 +0000 (23:26 -0500)]
Improve fileutils test coverage
Add some tests for g_file_read_link.
Matthias Clasen [Wed, 25 Dec 2013 04:24:40 +0000 (23:24 -0500)]
Unify error reporting in gfileutils.c
Use the set_file_error helper function in more places,
saving some 50 lines.
Matthias Clasen [Tue, 24 Dec 2013 05:06:00 +0000 (00:06 -0500)]
Improve GResourceFile test coverage
Matthias Clasen [Tue, 24 Dec 2013 05:05:36 +0000 (00:05 -0500)]
Improve gsettings test coverage
Matthias Clasen [Tue, 24 Dec 2013 05:05:01 +0000 (00:05 -0500)]
Improve gdbus test coverage
Matthias Clasen [Tue, 24 Dec 2013 05:04:44 +0000 (00:04 -0500)]
Improve converter test coverage
Matthias Clasen [Tue, 24 Dec 2013 05:04:14 +0000 (00:04 -0500)]
Add tests for remote actions
This improves the GApplication test coverage.
Matthias Clasen [Tue, 24 Dec 2013 05:03:20 +0000 (00:03 -0500)]
Improve GIOModule test coverage
Add a dedicated test for GIOModule.
Matthias Clasen [Tue, 24 Dec 2013 05:01:22 +0000 (00:01 -0500)]
Correct a doc comment
Matthias Clasen [Tue, 24 Dec 2013 04:59:56 +0000 (23:59 -0500)]
GIOModule: Add some more details to the docs
Matthias Clasen [Tue, 24 Dec 2013 04:59:37 +0000 (23:59 -0500)]
Fix a typo in a comment
The D-Bus interface is org.freedesktop.DBus.Peer, not
org.freedesktop.DBus.
Matthias Clasen [Mon, 23 Dec 2013 20:47:45 +0000 (15:47 -0500)]
Add a generic property test
This test is inspired by its namesake in GTK+. We instantiate
all types, and check the default values of their properties,
with some exceptions for types that are known not to work.
Matthias Clasen [Mon, 23 Dec 2013 20:01:20 +0000 (15:01 -0500)]
GTlsPassword: Make declared defaults match
String properties typically start out with a value of NULL.
No reason to deviate from that convention here.
Matthias Clasen [Mon, 23 Dec 2013 20:00:55 +0000 (15:00 -0500)]
Make GFileIcon more robust
No reason not to survive g_object_unref (g_object_new (...))
Matthias Clasen [Mon, 23 Dec 2013 20:00:18 +0000 (15:00 -0500)]
Make GEmblem more robust
No reason not to survive g_object_unref (g_object_new (...)).
Matthias Clasen [Mon, 23 Dec 2013 20:00:02 +0000 (15:00 -0500)]
Trivial formatting fixes
Matthias Clasen [Mon, 23 Dec 2013 17:11:03 +0000 (12:11 -0500)]
Add tests for x-content type sniffing
This test exercises the tree matching parts of gcontenttype.c
Matthias Clasen [Mon, 23 Dec 2013 17:09:46 +0000 (12:09 -0500)]
Simplify fileattributematcher tests a bit
Just use assertions instead of pretty-printed error messages.
Matthias Clasen [Mon, 23 Dec 2013 17:08:47 +0000 (12:08 -0500)]
Improve GApplicationCommandline test coverage
Add tests for many GApplicationCommandline apis to the
existing application tests.
Matthias Clasen [Mon, 23 Dec 2013 17:08:08 +0000 (12:08 -0500)]
Improve GSubprocess test coverage
Test the synchronous communicate functions.
Matthias Clasen [Mon, 23 Dec 2013 17:07:11 +0000 (12:07 -0500)]
Improve GInitable test coverage
This just adds a bit more variety to the uses of GInitiable
in other tests.
Jasper St. Pierre [Thu, 19 Sep 2013 19:41:09 +0000 (15:41 -0400)]
gobject.py: Simplify or_join_array
Jasper St. Pierre [Thu, 19 Sep 2013 19:24:16 +0000 (15:24 -0400)]
gobject.py: Simplify and reduce code a bit
Jasper St. Pierre [Thu, 19 Sep 2013 18:59:44 +0000 (14:59 -0400)]
gobject.py: Remove old hack for stripping IA__ symbols
We don't use IA__ symbols anymore; they've been replaced with -Bsymbolic
Jasper St. Pierre [Thu, 19 Sep 2013 15:55:47 +0000 (11:55 -0400)]
gobject.py: Fix indentation
Jasper St. Pierre [Tue, 31 Dec 2013 20:13:06 +0000 (15:13 -0500)]
codegen: direction defaults to "in"
Michael Catanzaro [Sun, 29 Dec 2013 20:20:13 +0000 (14:20 -0600)]
docs: fix a typo
Chun-wei Fan [Fri, 27 Dec 2013 04:25:18 +0000 (12:25 +0800)]
Visual Studio Projects: Cleanup Property Sheets
Make the projects include a single property sheet as necessary, which will
in turn include the other property sheet(s) as needed, so that we can avoid
warnings where we include the same property sheets twice in a single
project.
Also make the copying of pre-configured headers into custom build rules so
that they can be removed upon the clean command from the IDE and that they
can be re-copied if their counterpart *.h.win32 are updated.
Chun-wei Fan [Fri, 27 Dec 2013 02:50:35 +0000 (10:50 +0800)]
Visual Studio Projects: Use Unix Line Endings
...where possible, to make application of patches easier in the future.
The README.txt's and the .sln files are still in Windows/DOS line endings
as they need to be so.
Chun-wei Fan [Fri, 27 Dec 2013 02:32:01 +0000 (10:32 +0800)]
Visual Studio Property Sheets: Use Unix Line Endings
This makes it easier to apply patches in the future.
Antoine Jacoutot [Thu, 26 Dec 2013 10:03:45 +0000 (11:03 +0100)]
kqueue: fix a potential deadlock
In case an object is already monitored, we lock then return without
unlocking it which can then result in a self deadlock. So properly
unlock before returning.
https://bugzilla.gnome.org/show_bug.cgi?id=721074
Mikhail Gusarov [Thu, 26 Dec 2013 16:22:05 +0000 (11:22 -0500)]
gio-2.0.pc: Add -lselinux when linking statically if so configured
https://bugzilla.gnome.org/show_bug.cgi?id=721087
Morten Welinder [Thu, 26 Dec 2013 12:37:17 +0000 (07:37 -0500)]
gsubprocesslauncher: Use "env" instead of "environ"
The latter may come from system headers.
https://bugzilla.gnome.org/show_bug.cgi?id=721059
Matthias Scheler [Wed, 25 Dec 2013 18:57:56 +0000 (13:57 -0500)]
gio: Fix return of value from void function
https://bugzilla.gnome.org/show_bug.cgi?id=721034
Colin Walters [Mon, 18 Nov 2013 19:28:15 +0000 (14:28 -0500)]
Revert "gsettings m4: check for .xml in src/builddir"
This reverts commit
b3593693d918f0ae97094f6712d817180b8eea6a.
See: https://bugzilla.gnome.org/show_bug.cgi?id=712171#c3
See: https://bugzilla.gnome.org/show_bug.cgi?id=712171#c4
Tested using both srcdir == builddir and srcdir != builddir in hotssh.
Conflicts:
m4macros/gsettings.m4
https://bugzilla.gnome.org/show_bug.cgi?id=712630
Ryan Lortie [Fri, 22 Nov 2013 18:06:54 +0000 (13:06 -0500)]
gdbus tool: print GVariant errors in context
Use g_variant_parse_error_print_context() to format the error message
from the GVariant parser.
There is a slightly dubious interaction with the "parse me harder"
functionality here. We're probably going to have to deal with that
separately.
https://bugzilla.gnome.org/show_bug.cgi?id=715028
Ryan Lortie [Fri, 22 Nov 2013 17:57:21 +0000 (12:57 -0500)]
gapplication(1): print GVariant errors in context
Use g_variant_parse_error_print_context() to format the error message
from the GVariant parser.
https://bugzilla.gnome.org/show_bug.cgi?id=715028
Ryan Lortie [Fri, 22 Nov 2013 17:55:10 +0000 (12:55 -0500)]
gsettings tool: print GVariant errors in context
Use g_variant_parse_error_print_context() to format the error message
from the GVariant parser.
https://bugzilla.gnome.org/show_bug.cgi?id=715028
Ryan Lortie [Sun, 22 Dec 2013 16:41:19 +0000 (11:41 -0500)]
GVariant: add way to print context of parse errors
This was a feature intended from the very beginning that somehow never
got written. It's a way to replace these sort of error messages out of
the GVariant parser:
1-2,10-15:unable to find a common type
with something in the style of the Vala compiler:
unable to find a common type:
[1, 2, 3, 'str']
^ ^^^^^
https://bugzilla.gnome.org/show_bug.cgi?id=715028
Ryan Lortie [Sun, 22 Dec 2013 16:33:07 +0000 (11:33 -0500)]
Use POSIX-specified <poll.h> over <sys/poll.h>
POSIX specifies that <poll.h> is the correct header to include for
poll(), so let's do that instead.
https://bugzilla.gnome.org/show_bug.cgi?id=141251
Murray Cumming [Tue, 17 Sep 2013 09:10:28 +0000 (11:10 +0200)]
GVariant: Add g_variant_parse_error_quark()
Most GErrors, such as GSomethingError, have a function to get
their quark that looks like g_something_error_quark(),
so bindings (such as gtkmm) would expect GVariantParseError
to have g_variant_parse_error_quark(). Instead this had
g_variant_parser_get_error_quark().
This deprecates the old function and adds the correct one,
making life easier for gtkmm (and maybe others).
https://bugzilla.gnome.org/show_bug.cgi?id=708212
Matthias Clasen [Sun, 22 Dec 2013 05:44:56 +0000 (00:44 -0500)]
More settings backend tests
This adds some more test coverage around writability changes
in the keyfile and delayed settings backends.
Matthias Clasen [Sun, 22 Dec 2013 05:43:44 +0000 (00:43 -0500)]
GKeyfileSettingsBackend: Use a directory monitor
GKeyfileSettingsBackend was using a file monitor to monitor
the parent directory of its keyfile.
Matthias Clasen [Sat, 21 Dec 2013 21:49:52 +0000 (16:49 -0500)]
More test coverage for settings backends
Matthias Clasen [Sat, 21 Dec 2013 21:16:16 +0000 (16:16 -0500)]
Simplify coverage rules
This works just fine, no need for the manual removal of the
info file.
Matthias Clasen [Sat, 21 Dec 2013 20:44:51 +0000 (15:44 -0500)]
Fix a typo
Matthias Clasen [Sat, 21 Dec 2013 18:20:44 +0000 (13:20 -0500)]
Fix a !srcdir problem with installed tests
When installing scripts which are not generated in a
srcdir != builddir build, $< will have ../.. goo in it.
So we need to strip the directory parts before using
the filename to construct a new path.
Matthias Clasen [Sat, 21 Dec 2013 17:51:22 +0000 (12:51 -0500)]
Improve gsettings test coverage
Deal with corner cases around delay.
Matthias Clasen [Sat, 21 Dec 2013 16:56:24 +0000 (11:56 -0500)]
contenttype tests: better assertions
I recently had to track down why these tests failed. Turned
out that some rogue package on my system had installed mime
types that declared all files with 3 letter names to be
'chemical/x-turbomole-vibrational'.
This change will make it more obvious what is going on by
mentioning the mime types in the assertion message.
Matthias Clasen [Sat, 21 Dec 2013 15:48:40 +0000 (10:48 -0500)]
Slightly improve coverate for notification tests
Matthias Clasen [Sat, 21 Dec 2013 06:10:59 +0000 (01:10 -0500)]
Improve test coverage for GPropertyAction
Matthias Clasen [Sat, 21 Dec 2013 06:00:10 +0000 (01:00 -0500)]
Test some gsettings corner case
Setting a strv to NULL was not tested before.
Matthias Clasen [Sat, 21 Dec 2013 05:04:14 +0000 (00:04 -0500)]
Some tweaks to coverage support
Rename the configure option to --enable-coverage, and make
it quiet by default.
Fran Diéguez [Thu, 19 Dec 2013 00:17:25 +0000 (01:17 +0100)]
Updated Galician translations
Dan Winship [Wed, 18 Dec 2013 14:59:54 +0000 (09:59 -0500)]
Make g_test_run() return 77 if all tests are skipped
Change g_test_run() to return 1 on failure (rather than the number of
failed tests), and 77 if all tests are skipped (since automake and
some other test harnesses recognize that status code).
Previously g_test_run() returned the number of failed tests, but this
behavior was not documented, and at any rate, prior to 2.39,
g_test_run() would normally not return at all if an error occurred.
https://bugzilla.gnome.org/show_bug.cgi?id=720263
Dan Winship [Wed, 18 Dec 2013 14:29:29 +0000 (09:29 -0500)]
gtestutils: rename test_skip_count to test_startup_skip_count
https://bugzilla.gnome.org/show_bug.cgi?id=720263
Dan Winship [Wed, 11 Dec 2013 15:32:11 +0000 (16:32 +0100)]
gtestutils: skipping a test should count as success, not failure
In particular, the test program as a whole should exit with status 0
if you skipped some tests but did not fail any.
https://bugzilla.gnome.org/show_bug.cgi?id=720263
Bastien Nocera [Mon, 16 Dec 2013 16:21:41 +0000 (17:21 +0100)]
gdbus-codegen: Fix crasher in goa-using apps
When replacing a version of goa-daemon (from gnome-online-accounts)
by a newer version with some added interfaces, evolution-data-server
and the gvfs-goa volume monitor might crash as there's no interface
definition for this new interface.
Work-around this by returning earlier from the _notify() implementation,
rather than accessing invalid memory.
https://bugzilla.gnome.org/show_bug.cgi?id=720539
Tim Lunn [Tue, 17 Dec 2013 20:14:38 +0000 (07:14 +1100)]
Make gdb pretty-printers compatible with Python3
On some systems gdb is linked against python3 where "long" no longer
exists. In this case should be using int.
https://bugzilla.gnome.org/show_bug.cgi?id=720635