Benjamin Otte [Wed, 22 Jun 2011 02:29:42 +0000 (04:29 +0200)]
Revert "Add a testcase with a print dialog"
THe testcase depends on the installed printers, so is nondeterministic.
This reverts commit
5bf14cc7485fe015ad15747454e877f2fb3dabc5.
Benjamin Otte [Wed, 22 Jun 2011 02:24:32 +0000 (04:24 +0200)]
Revert "Add a filechooser testcase"
File choosers depend on the files in the current directory and the
settings the user has set for the file chooser, so the output cannot be
deterministic.
This reverts commit
f05c9e02c49f857c33e02c3d89483fcccb5df254.
Benjamin Otte [Wed, 22 Jun 2011 02:21:53 +0000 (04:21 +0200)]
Revert "Add a testcase involving a font selection dialog"
The font chooser depends on the fonts installed on the system, so the
output is nondeterministic.
This reverts commit
08fdc399762c2af07d94e42a1801e691e5d4d6ab.
Benjamin Otte [Wed, 22 Jun 2011 02:20:36 +0000 (04:20 +0200)]
tests: Don't emit a message for every unnamed object
Benjamin Otte [Wed, 22 Jun 2011 02:14:49 +0000 (04:14 +0200)]
tests: Reset the unnamed counter before a test run
This guarantees that objects are named the same all the time.
Benjamin Otte [Wed, 22 Jun 2011 01:47:25 +0000 (03:47 +0200)]
gail: Add support for action_set_description
As set_description is never called and unsupported by the at-spi, we can
omit implementing it.
This means we can also omit get_description calls in various places, as
they'd just return the default value: NULL.
Benjamin Otte [Tue, 21 Jun 2011 16:14:40 +0000 (18:14 +0200)]
tests: Fix crash due to freeing random pointers
Benjamin Otte [Tue, 21 Jun 2011 14:43:56 +0000 (16:43 +0200)]
a11y: Remove hack for old code
This code was supposed to work around a bad interaction between GOK and
Nautilus from 7 years ago.
If it still exists, the GOK developers may complain to the Nautilus
developers.
https://bugzilla.gnome.org/show_bug.cgi?id=137401
Benjamin Otte [Tue, 21 Jun 2011 11:16:11 +0000 (13:16 +0200)]
gail: Remove useless code
The ATK default code does the same thing as these two functions.
Joachim Breitner [Tue, 21 Jun 2011 08:55:28 +0000 (10:55 +0200)]
docs: Remove mention of client messages
Matthias Clasen [Wed, 22 Jun 2011 00:18:04 +0000 (20:18 -0400)]
Add an assistant testcase
Matthias Clasen [Wed, 22 Jun 2011 00:12:37 +0000 (20:12 -0400)]
Add a testcase involving an expander
Matthias Clasen [Wed, 22 Jun 2011 00:08:46 +0000 (20:08 -0400)]
Add a testcase with a print dialog
Matthias Clasen [Wed, 22 Jun 2011 00:03:08 +0000 (20:03 -0400)]
Add a messagedialog testcase
Matthias Clasen [Tue, 21 Jun 2011 23:58:25 +0000 (19:58 -0400)]
Add a testcase involving a font selection dialog
Matthias Clasen [Tue, 21 Jun 2011 23:55:39 +0000 (19:55 -0400)]
Add a testcase with an appchooser dialog
Matthias Clasen [Tue, 21 Jun 2011 23:34:55 +0000 (19:34 -0400)]
Add a README for accessibility-dump
Matthias Clasen [Tue, 21 Jun 2011 23:20:59 +0000 (19:20 -0400)]
Add an about dialog testcase
Matthias Clasen [Tue, 21 Jun 2011 23:20:28 +0000 (19:20 -0400)]
Add a color chooser testcase
Matthias Clasen [Tue, 21 Jun 2011 23:12:55 +0000 (19:12 -0400)]
Add a scrollbar to the range testcase
Matthias Clasen [Tue, 21 Jun 2011 22:56:32 +0000 (18:56 -0400)]
Add a filechooser testcase
Matthias Clasen [Tue, 21 Jun 2011 22:53:30 +0000 (18:53 -0400)]
Remove debug spew
This had crept into the expected output for the buttons
testcase.
Matthias Clasen [Tue, 21 Jun 2011 22:53:06 +0000 (18:53 -0400)]
GailTreeView: deal with absence of model
Matthias Clasen [Tue, 21 Jun 2011 22:52:24 +0000 (18:52 -0400)]
GailImageCell: Deal with absence of pixbufs
...and when there are pixbufs, don't leak them.
Matthias Clasen [Tue, 21 Jun 2011 22:51:40 +0000 (18:51 -0400)]
Fix dependencies between directories
This ensures that libgtk gets relinked when I touch sources
in gtk/a11y.
Matthias Clasen [Tue, 21 Jun 2011 22:06:02 +0000 (18:06 -0400)]
Try harder to generate names for accessibles
We are still hitting the 'no name' case in GtkFileChooser for
a GailBox, otherwise.
Matthias Clasen [Tue, 21 Jun 2011 22:01:26 +0000 (18:01 -0400)]
Handle AtkValue implementations that return integer values
We do this by transforming the returned GValue to a string.
As a side-effect, the formatting for doubles changes, so
adjust expected output to match.
Matthias Clasen [Mon, 20 Jun 2011 13:03:25 +0000 (09:03 -0400)]
Don't build the libgail-util docs
The doc build for that is currently broken, and libgail-util is
undergoing reconstructive surgery anyway, currently.
Or maybe it'll turn out to be an amputation...
Benjamin Otte [Mon, 20 Jun 2011 12:22:49 +0000 (14:22 +0200)]
tests: omit fg-color and bg-color values from attributes
They depend on the environment, too.
Benjamin Otte [Mon, 20 Jun 2011 12:14:00 +0000 (14:14 +0200)]
tests: Add line breaks when dumping text attributes
This improves diff readability a lot when attributes differ.
Benjamin Otte [Mon, 20 Jun 2011 11:58:10 +0000 (13:58 +0200)]
widget: Use g_type_add_class_private()
It turns out that that funciton does indeed exist and my rant was just
outdated. Shame on me.
Benjamin Otte [Mon, 20 Jun 2011 11:57:50 +0000 (13:57 +0200)]
gail: Fix compile warning
Benjamin Otte [Mon, 20 Jun 2011 11:57:26 +0000 (13:57 +0200)]
tests: undefine GTK_DISABLE_DEPRECATED as we include deprecated types
Matthias Clasen [Mon, 20 Jun 2011 04:03:53 +0000 (00:03 -0400)]
Remove debug spew
Matthias Clasen [Mon, 20 Jun 2011 04:00:39 +0000 (00:00 -0400)]
Fix GtkSwitchAccessible type definition
It turns out that ATK_DEFINE_TYPE_WITH_CODE() is broken; it
tells GType that the class and instance size for the accessible
type are the same as for its parent type. Which is not true
if your instance struct has members such as 'description' here.
This was causing hard-to-track-down memory corruption, since
description and the GtkAccessible private pointer were sharing
the same memory location.
Matthias Clasen [Mon, 20 Jun 2011 01:29:42 +0000 (21:29 -0400)]
Add a test involving a tree view
Matthias Clasen [Mon, 20 Jun 2011 01:28:35 +0000 (21:28 -0400)]
Dump AtkTable properties
Also, make get_name() deal better with GtkAccessibles whose
widgets don't have buildable names.
Matthias Clasen [Sun, 19 Jun 2011 23:51:15 +0000 (19:51 -0400)]
Dump AtkComponent properties
Note that the spatial aspects of AtkComponent are all dependent
on fonts and other environmental factors, thus we only dump the
little bits outside of that.
Matthias Clasen [Sun, 19 Jun 2011 23:28:47 +0000 (19:28 -0400)]
Make the dumped output a bit more readable
By adding sections for each interface.
Matthias Clasen [Sun, 19 Jun 2011 23:16:26 +0000 (19:16 -0400)]
Add a test involving a text view
Matthias Clasen [Sun, 19 Jun 2011 23:15:54 +0000 (19:15 -0400)]
Dump AtkStreamableContent properties
Also, simplify the format for links slightly.
Matthias Clasen [Sun, 19 Jun 2011 23:01:57 +0000 (19:01 -0400)]
Fix the expected output for range.ui
GtkScale puts a LRM character at the start of the string.
Matthias Clasen [Sun, 19 Jun 2011 22:55:12 +0000 (18:55 -0400)]
Add a test involving links
Matthias Clasen [Sun, 19 Jun 2011 22:53:56 +0000 (18:53 -0400)]
Dump AtkHyperLinkImpl properties
Matthias Clasen [Sun, 19 Jun 2011 22:53:24 +0000 (18:53 -0400)]
Fix some problems with the GailLinkButton implementation
Matthias Clasen [Sun, 19 Jun 2011 21:58:57 +0000 (17:58 -0400)]
Add a test involving ranges
Matthias Clasen [Sun, 19 Jun 2011 21:58:31 +0000 (17:58 -0400)]
Dump AtkValue properties
Matthias Clasen [Sun, 19 Jun 2011 21:19:34 +0000 (17:19 -0400)]
Add a test involving a selection
Matthias Clasen [Sun, 19 Jun 2011 21:18:47 +0000 (17:18 -0400)]
Dump AtkSelection properties
Also make get_name() deal with object that are not GtkAccessible,
but implement AtkText. An example of this is GailNotebookPage.
Matthias Clasen [Sun, 19 Jun 2011 20:53:16 +0000 (16:53 -0400)]
Dump AtkAction properties
Adjust expected output to match.
Matthias Clasen [Sun, 19 Jun 2011 19:30:01 +0000 (15:30 -0400)]
Dump AtkImage properties too
Update expected output to match.
Matthias Clasen [Sun, 19 Jun 2011 19:22:08 +0000 (15:22 -0400)]
More text properties
Matthias Clasen [Sun, 19 Jun 2011 18:56:05 +0000 (14:56 -0400)]
Dump AtkText properties too
...would be nice if these actually _were_ properties.
Update the expected output to match.
Matthias Clasen [Sun, 19 Jun 2011 18:11:46 +0000 (14:11 -0400)]
Remove remnants of GnomeCanvas
There were some random places where gail was poking at types
to see if they were a GnomeCanvas. Just remove this.
Matthias Clasen [Sun, 19 Jun 2011 18:05:06 +0000 (14:05 -0400)]
Fix make check
Not really the best fix; for now just ignore accessible parents
that are not GtkAccessibles - it seems that something causes
GailToplevels to show up as parents of GailWindows, randomly.
Matthias Clasen [Sun, 19 Jun 2011 18:04:29 +0000 (14:04 -0400)]
Remove an extra line from expected output
Matthias Clasen [Sun, 19 Jun 2011 17:36:20 +0000 (13:36 -0400)]
Don't forget to free an empty attribute set
Matthias Clasen [Sun, 19 Jun 2011 17:20:32 +0000 (13:20 -0400)]
Add a more convenient way to dump
For initial creation of the results, allow to run
accessibility-dump --generate file.ui,
which will dump just the expected output to stdout,
without any extra decorations.
Matthias Clasen [Sun, 19 Jun 2011 15:32:43 +0000 (11:32 -0400)]
Dump more accessibility data
This commit adds dumping of parents, attributes and states.
The expected dump output has been adjusted to match.
Matthias Clasen [Sun, 19 Jun 2011 06:11:11 +0000 (02:11 -0400)]
Add new tests to EXTRA_DIST
Matthias Clasen [Sun, 19 Jun 2011 06:10:27 +0000 (02:10 -0400)]
Add a test that sets ::accessible-name
Matthias Clasen [Sun, 19 Jun 2011 06:03:24 +0000 (02:03 -0400)]
Add a test that has a labeled-by relation in it
Benjamin Otte [Sun, 19 Jun 2011 04:11:38 +0000 (06:11 +0200)]
tests: Add first shot at an accessibility dump tool
The tool works like this:
./accessibility-dump [FILE ...]
If no files are given, all files with the extension ".ui" in the current
directory are taken. For every file "test.ui", the following steps are
performed:
1) test.ui is loaded using GtkBuilder
2) The accessible for the window is loaded
3) The information of accessible is converted into a string using a
syntax defined in this test file
4) The generated string is diffed with the file "test.txt"
5) If the diff is empty, the test is a success, if not, the test fails.
6) The diff is output when the test runner is run with --verbose
So to add a test named "test", create a file called "test.ui", put it
into this directory. Then create the expected output file "test.txt",
put it into this directory too. You can create the initial version of
this file by invoking "./accessibility-dump --verbose test.ui". The
output will contain the expected text and can be copy/pasted into the
text file.
Benjamin Otte [Sat, 18 Jun 2011 07:51:18 +0000 (09:51 +0200)]
API: Add gtk_widget_class_set_accessible_type()
The function is supposed to bypass the ATK registry. For 2 reasons:
1) We get rid of a lot of boilerplate madness.
2) The registry allows creating multiple accessibles per widget and we
don't.
The old code for registries is still there.
Benjamin Otte [Sat, 18 Jun 2011 07:11:11 +0000 (09:11 +0200)]
widget: Add a GtkWidgetClassPrivate
As long as glib doesn't provide class privates, we'll have to ship our
own...
Benjamin Otte [Wed, 15 Jun 2011 12:57:20 +0000 (14:57 +0200)]
a11y: Remove unused HTML objects
Seems they were for GtkHTML and never used since GAIL got imported into
GTK.
Matthias Clasen [Fri, 17 Jun 2011 04:15:27 +0000 (00:15 -0400)]
No need for gail.h
Matthias Clasen [Fri, 17 Jun 2011 04:14:11 +0000 (00:14 -0400)]
NO_GAIL not used anymore
Matthias Clasen [Fri, 17 Jun 2011 04:11:00 +0000 (00:11 -0400)]
We're not a module anymore
So no need to implement module entry points; and libgnome is
dead too, so no need to export functions for it either.
Matthias Clasen [Fri, 17 Jun 2011 03:13:38 +0000 (23:13 -0400)]
Add a first, trivial test for accessibles
Matthias Clasen [Fri, 17 Jun 2011 02:40:21 +0000 (22:40 -0400)]
And remove it from the build too
Matthias Clasen [Fri, 17 Jun 2011 02:38:48 +0000 (22:38 -0400)]
Remove the old gail tests
This is not what we are looking for in tests, and it hasn't been
touched in ages.
Benjamin Otte [Wed, 15 Jun 2011 10:46:50 +0000 (12:46 +0200)]
gtk: Include gail by default, don't build it as a module
It is now no longer possible to disable it.
This doesn't matter though because GTK will not instantiate a11y
objects until you actually use it. So nothing changes in practice.
Benjamin Otte [Tue, 14 Jun 2011 20:27:11 +0000 (22:27 +0200)]
gtk: Move a11y tests fro a11y/tests to tests/a11y
Benjamin Otte [Tue, 14 Jun 2011 20:08:55 +0000 (22:08 +0200)]
gail-util: Move into toplevel directory
Benjamin Otte [Tue, 14 Jun 2011 18:29:15 +0000 (20:29 +0200)]
gail: Copy gail-util functions into gail
Otherwise we get a circular dependency if we move libgail into GTK:
GTK depends on gail-util depends on gail (is part of GTK)
Benjamin Otte [Tue, 14 Jun 2011 15:54:27 +0000 (17:54 +0200)]
gail: Move from modules/other/gail to gtk/a11y
Benjamin Otte [Tue, 14 Jun 2011 14:09:35 +0000 (16:09 +0200)]
gail: No need to include modules/other in CFLAGS anymore
Now that we include headers directly, ew don't need it anymore. And the
current directory is included automatically.
Benjamin Otte [Tue, 14 Jun 2011 13:35:34 +0000 (15:35 +0200)]
gail: Include files directly
Instead of including <gail/gailfoo.h>, include "gailfoo.h"
Private headers should be included like this (to make it esily visible
that the headers are private) and we want to move this directory to a
different name, so hardcoding the directory name is counterproductive to
that.
Arash Mousavi [Tue, 5 Jul 2011 17:48:59 +0000 (22:18 +0430)]
Updated Persian translations
Matthias Clasen [Tue, 5 Jul 2011 15:04:06 +0000 (11:04 -0400)]
Post-release version bump
Up to 3.1.9
Matthias Clasen [Tue, 5 Jul 2011 15:03:23 +0000 (11:03 -0400)]
3.1.8
Matthias Clasen [Tue, 5 Jul 2011 13:33:51 +0000 (09:33 -0400)]
Update NEWS
Benjamin Otte [Tue, 5 Jul 2011 13:28:18 +0000 (15:28 +0200)]
reftests: Use an offscreen window to avoid artifacts
When tests are larger than the screen size and no compositing is
enabled, the window will be clipped to the screen size and all areas
outside of the screen have undefined contents.
To avoid this, we can use an offscreen window.
Benjamin Otte [Tue, 5 Jul 2011 13:18:14 +0000 (15:18 +0200)]
reftests: Fix label-sizing reference for Pango
See https://bugzilla.gnome.org/show_bug.cgi?id=649783 for a discussion
of the problem. Should this bug be fixed, this patch can just be
reverted.
Benjamin Otte [Tue, 5 Jul 2011 13:17:23 +0000 (15:17 +0200)]
reftests: load/save file once in Glade
This way the output is canonicalized for further editing. No changes
have been done.
Yaron Shahrabani [Sat, 2 Jul 2011 07:26:09 +0000 (10:26 +0300)]
Updated Hebrew translation.
Cosimo Cecchi [Fri, 1 Jul 2011 23:57:27 +0000 (01:57 +0200)]
reftests: Add test for box brokenness
This is a tests for the last commit.
Benjamin Otte [Fri, 1 Jul 2011 23:25:45 +0000 (01:25 +0200)]
box: Remove cache for sibling paths
We can't cache this path as widgets might dynamically add or remove
style classes without telling us. At which point the stored path does
not work.
Cosimo Cecchi [Fri, 1 Jul 2011 16:33:26 +0000 (12:33 -0400)]
toolbar: don't leak the sibling path
Cosimo Cecchi [Thu, 30 Jun 2011 21:00:32 +0000 (17:00 -0400)]
toolbar: don't count invisible widgets when building the sibling path
Also, fix RTL nth-child handling for toolbars.
Cosimo Cecchi [Thu, 30 Jun 2011 19:51:10 +0000 (15:51 -0400)]
toolbar: invalidate order when an item visibility changes
Matthias Clasen [Sat, 18 Jun 2011 17:14:18 +0000 (13:14 -0400)]
Test box and toolbar regions in parallel
This makes it easy to see discrepancies.
Plus, it is fun
Matthias Clasen [Sat, 18 Jun 2011 16:45:56 +0000 (12:45 -0400)]
toolbar: nth-child support for GtkToolbar
This doesn't quite work right yet.
Cosimo Cecchi [Thu, 30 Jun 2011 20:19:57 +0000 (16:19 -0400)]
box: follow children visibility when building the sibling path
This fixes nth-child not working for widgets which were not
visible before being added to the box, but it's also a better fix
for https://bugzilla.gnome.org/show_bug.cgi?id=652769
Cosimo Cecchi [Fri, 1 Jul 2011 15:47:41 +0000 (11:47 -0400)]
Revert "gtkbox: Invalidate order on changes, even if the child is not visible"
This reverts commit
8cb0cc58e321641ebda699b9cbdc0067aeb56d3f.
Matthias Clasen [Thu, 30 Jun 2011 23:18:26 +0000 (19:18 -0400)]
Fix up docs for predefined style classes.
The table was incomplete and out of date. Instead, just
put a list of links in that place, and move all the extra
documentation to the macros. Bug 653785
Matthias Clasen [Thu, 30 Jun 2011 18:45:16 +0000 (14:45 -0400)]
GdkKeymap: Prevent an infinite loop in the non-XKB case
Somehow the increment got lost, as comparison with the 2.x code
shows.
Vincent Untz [Tue, 28 Jun 2011 09:47:37 +0000 (11:47 +0200)]
gtkbox: Invalidate order on changes, even if the child is not visible
When we build the sibling path for the order, we do not skip hidden
children (since, quoting the comment, "we cannot reliably detect changes
in widget visibility"). So we need to invalidate the order when hidden
children are reordered and removed.
https://bugzilla.gnome.org/show_bug.cgi?id=652769
John (J5) Palmieri [Wed, 29 Jun 2011 18:53:33 +0000 (14:53 -0400)]
[gi] add more annotations to force drag and drop api to be methods