Matthias Clasen [Wed, 10 May 2023 20:55:16 +0000 (16:55 -0400)]
gsk: Fix some renderer gotchas
unrealize() needs to clean up the surface that was set
in realize(), and both need to emit property change
notification.
Matthias Clasen [Wed, 10 May 2023 14:38:13 +0000 (10:38 -0400)]
gsk: Improve test coverage
Add more tests for gsk transforms, render node
and renderer apis.
Matthias Clasen [Wed, 10 May 2023 18:59:58 +0000 (14:59 -0400)]
gsk: Drop an unused function
Matthias Clasen [Wed, 10 May 2023 11:36:22 +0000 (07:36 -0400)]
testsuite: Close displays in the gsk tests
This gives us better coverage of the dispose
codepath for a driver that has actually been
used.
Matthias Clasen [Wed, 10 May 2023 11:34:23 +0000 (07:34 -0400)]
gsk: Dispose the driver when the display is closed
This was the intention, but the object data by itself
does not achieve that: We do run dispose on the display
when it is closed, but object data is only cleared in
finalize. So listen to the ::closed signal and remove
the driver ourselves.
Fix up the drivers dispose implementation enough for
that to actually work.
Matthias Clasen [Wed, 10 May 2023 17:44:24 +0000 (13:44 -0400)]
Use gsk_matrix_transform_point3d consistently
Matthias Clasen [Wed, 10 May 2023 10:38:07 +0000 (10:38 +0000)]
Merge branch 'matthiasc/for-main' into 'main'
tests: Stop copying the tool tests
See merge request GNOME/gtk!5947
Matthias Clasen [Wed, 10 May 2023 03:02:48 +0000 (23:02 -0400)]
tests: Don't disable vertex arrays
It doesn't work, and that makes tests fail.
Matthias Clasen [Wed, 10 May 2023 03:02:30 +0000 (23:02 -0400)]
Improve coverage for half-float fallbacks
Matthias Clasen [Wed, 10 May 2023 01:39:20 +0000 (21:39 -0400)]
build: Rename another option
Matthias Clasen [Wed, 10 May 2023 01:28:52 +0000 (21:28 -0400)]
build: Deprecate an option
Rename the demos option to build-demos, to match
the other options for building optional parts.
Matthias Clasen [Wed, 10 May 2023 01:33:21 +0000 (21:33 -0400)]
build: Require meson 0.63
This is needed for option deprecation.
Matthias Clasen [Wed, 10 May 2023 01:21:39 +0000 (21:21 -0400)]
tests: Stop copying the tool tests
No point in doing that, and the meson feature
we are using here is deprecated.
Matthias Clasen [Wed, 10 May 2023 00:48:33 +0000 (20:48 -0400)]
Post-release version bump
Matthias Clasen [Tue, 9 May 2023 17:36:21 +0000 (13:36 -0400)]
4.11.2
Matthias Clasen [Tue, 9 May 2023 20:42:38 +0000 (16:42 -0400)]
Update baseline docs a bit
Matthias Clasen [Tue, 9 May 2023 20:53:59 +0000 (20:53 +0000)]
Merge branch 'deprecate-align-value' into 'main'
Deprecate GTK_ALIGN_BASELINE
See merge request GNOME/gtk!5946
Matthias Clasen [Tue, 9 May 2023 20:30:37 +0000 (16:30 -0400)]
Update tests
Matthias Clasen [Tue, 9 May 2023 17:08:34 +0000 (13:08 -0400)]
Deprecate GTK_ALIGN_BASELINE
We can deprecated enum values now.
Matthias Clasen [Tue, 9 May 2023 17:08:08 +0000 (13:08 -0400)]
Use GTK_ALIGN_BASELINE_FILL
Replace GTK_ALIGN_BASELINE with its new name
everywhere.
Matthias Clasen [Tue, 9 May 2023 18:23:40 +0000 (18:23 +0000)]
Merge branch 'wip/ebassi/gen-version-macros' into 'main'
Generate version and deprecation macros at build time
See merge request GNOME/gtk!5945
Emmanuele Bassi [Tue, 9 May 2023 18:01:21 +0000 (19:01 +0100)]
build: Internal tests must set GTK_COMPILATION
Otherwise they won't be able to use the appropriate function visibility
on Windows.
Emmanuele Bassi [Tue, 9 May 2023 16:02:09 +0000 (17:02 +0100)]
Minimize the amount of inclusions in public headers
Public headers should mainly include gdktypes.h, which already include
the symbol visibility and versioning macros; we can also modify
gdktypes.h to include the enumerations.
Matthias Clasen [Tue, 9 May 2023 16:36:55 +0000 (16:36 +0000)]
Merge branch 'wip/otte/sections' into 'main'
Add GtkSectionModel
See merge request GNOME/gtk!5818
Matthias Clasen [Tue, 9 May 2023 15:17:26 +0000 (11:17 -0400)]
Use #pragma once
We switched over to using #pragma once, so lets
use it consistently, in generated headers as well.
Matthias Clasen [Tue, 9 May 2023 15:16:31 +0000 (11:16 -0400)]
Drop the gdk/gdkversionmacros.h header
It was just added to avoid touching too many files.
Touching them now.
Emmanuele Bassi [Tue, 9 May 2023 14:31:27 +0000 (15:31 +0100)]
demos: Drop symbol visibility macros from demo types
Since the corresponding source files do not include "config.h", and are
not compiled with `GTK_COMPILATION`, they will generate the wrong symbol
exporting under Windows.
Emmanuele Bassi [Tue, 9 May 2023 12:13:44 +0000 (13:13 +0100)]
build: Minor style nitpicks
Emmanuele Bassi [Tue, 9 May 2023 12:13:00 +0000 (13:13 +0100)]
build: Use Meson's gnu_symbol_visibility
Instead of injecting `-fvisibility=hidden` depending on a compiler check
ourselves, let Meson do it for us.
This also avoids us having to filter `-fvisibility=hidden` when reusing
the common compiler flags.
Matthias Clasen [Mon, 8 May 2023 20:03:16 +0000 (16:03 -0400)]
Add a missing file
Matthias Clasen [Mon, 8 May 2023 19:59:41 +0000 (15:59 -0400)]
Fix install location
Matthias Clasen [Mon, 8 May 2023 19:53:42 +0000 (15:53 -0400)]
Use GTK_COMPILATION throughout
It is a bit ugly that it doesn't fit the neat
namespace schema, but it is what we use everywhere.
Matthias Clasen [Mon, 8 May 2023 19:53:27 +0000 (15:53 -0400)]
Don't excessively define GDK_EXTERN
Once is enough.
Matthias Clasen [Mon, 8 May 2023 19:43:08 +0000 (15:43 -0400)]
Make it build
Fix the circular dependency by moving the generated
headers to gdk/version/, and build that directory
first.
Misc other fixes, such as putting the custom targets
as sources, not depedencies, and using the correct
major version in the generator script.
Emmanuele Bassi [Mon, 8 May 2023 18:39:35 +0000 (19:39 +0100)]
Generate version and deprecation macros at build time
Let's poach the same script used by GLib to avoid having to add all the
version macros by hand every time we increment the GTK version.
This is a work in progress:
- need to rename the GLIB_STATIC_COMPILATION check
- circular dependency: libgtkcss depends on gdkversionmacros.h, but libgdk
depends on libgtkcss
Benjamin Otte [Tue, 9 May 2023 15:16:43 +0000 (15:16 +0000)]
Merge branch 'wip/otte/no-vsync' into 'main'
Add GDK_DEBUG=no-vsync
See merge request GNOME/gtk!5944
Benjamin Otte [Wed, 12 Apr 2023 15:00:08 +0000 (17:00 +0200)]
listview: Make inertness not depend on factory
The inertness concept introduced in
62e9d1e470b assumed a listview was
inert when no factory was set. This has 2 problems:
1. columnview uses a listview without factories.
2. header factory being set but factory not being set technically makes
the listiew inert - but should it?
So for now, make inertness only depend on visibility and root.
A side benefit is that this matches columnview semantics.
Benjamin Otte [Mon, 8 May 2023 18:47:39 +0000 (20:47 +0200)]
gtk-demo: Add alternative settings demo
This one displays a long list of all settings with sections and allows
filtering the list to quickly search for keys.
Benjamin Otte [Tue, 11 Apr 2023 02:46:40 +0000 (04:46 +0200)]
gtk-demo: Split out code from the settings demo
I want to add another demo using settings and share relevant code.
Benjamin Otte [Tue, 11 Apr 2023 02:46:24 +0000 (04:46 +0200)]
theme: Add some random CSS for section headers
Benjamin Otte [Tue, 11 Apr 2023 02:45:30 +0000 (04:45 +0200)]
listitemmanager: Create header items for sections
Now, finally, listitemmanager has all the necessary support to make
section headers work.
Benjamin Otte [Mon, 10 Apr 2023 12:32:24 +0000 (14:32 +0200)]
listview: GtkListHeader and gtk_list_view_set_header_factory()
Adds infrastructure for handling factories for list headers.
For now, listitemmanager doesn't instantiate them, so they aren't used.
Benjamin Otte [Mon, 10 Apr 2023 21:18:56 +0000 (23:18 +0200)]
listitemmanager: Add GtkListItemChange
... for tracking widgets during changes.
This just pulls all the different disjointed parts into one struct with
a sensible API.
Benjamin Otte [Sun, 9 Apr 2023 12:24:26 +0000 (14:24 +0200)]
listitemmanager: Add GtkListHeaderBase and vfuncs
This is plumbing for section handling.
The vfuncs don't do anything, this is just the infrastructure.
Benjamin Otte [Mon, 10 Apr 2023 21:17:41 +0000 (23:17 +0200)]
listitembase: exit early
When update() is called but no values are changed, exit early.
Benjamin Otte [Sun, 9 Apr 2023 03:29:14 +0000 (05:29 +0200)]
listitemmanager: Make sections configurable
Add a gtk_list_item_manager_set_has_sections() (default: FALSE) that
requires explicit turning on for the listitemmanager to gain section
support.
Benjamin Otte [Sun, 9 Apr 2023 02:26:51 +0000 (04:26 +0200)]
listitemmanager: Add/remove sections as needed
When ensuring widgets, ensure that their section is known. This will
be relevant when we use section widgets.
Also ensure that sections that don't cover any widget get destroyed.
Benjamin Otte [Sun, 9 Apr 2023 02:24:47 +0000 (04:24 +0200)]
testsuite: assert widgets are in matched sections
All widgets that we display should have their sections known, so assert
that that is the case.
Benjamin Otte [Sun, 9 Apr 2023 02:23:35 +0000 (04:23 +0200)]
testsuite: Add a print function for the listitemmanager
This is very useful when debugging.
Benjamin Otte [Sat, 8 Apr 2023 01:59:17 +0000 (03:59 +0200)]
listitemmanager: Add gtk_list_tile_get_next_skip()
... and gtk_list_tile_get_previous_skip() and use them.
Allows skipping over REMOVED and FILLER tiles which makes the code a lot
more readable.
Benjamin Otte [Sat, 8 Apr 2023 01:47:13 +0000 (03:47 +0200)]
listitemmanager: Add gtk_list_tile_set_type()
Benjamin Otte [Sun, 9 Apr 2023 02:42:12 +0000 (04:42 +0200)]
testsuite: Add verbose output to listitemmanager test
Benjamin Otte [Fri, 7 Apr 2023 18:50:08 +0000 (20:50 +0200)]
listitemmanager: Properly handle sections during add/remove
We don't insert sections ourselves yet, but we handle the existing one
when items get added or removed.
Benjamin Otte [Fri, 7 Apr 2023 18:49:06 +0000 (20:49 +0200)]
listitemmanager: Augment header/footer info
Benjamin Otte [Fri, 7 Apr 2023 02:00:00 +0000 (04:00 +0200)]
listitemmanager: Add section tiles
For now, we just have a HEADER at the start and a FOOTER at the end.
That's hard enough to get right.
Benjamin Otte [Thu, 6 Apr 2023 00:28:26 +0000 (02:28 +0200)]
testsuite: Improve listitemmanager test
It now has trackers!
Benjamin Otte [Wed, 5 Apr 2023 13:10:16 +0000 (15:10 +0200)]
listitemmanager: Add tile types
We have a FILLER and a REMOVED type now.
Also makes gc() more sensitive to types.
Benjamin Otte [Wed, 5 Apr 2023 00:49:03 +0000 (02:49 +0200)]
listitemmanager: Add a type to tiles
There's only one type for now, but hey, you have to start somewhere.
Benjamin Otte [Sat, 26 Feb 2022 20:46:42 +0000 (21:46 +0100)]
testsuite: Add tests for sections to filterlistmodel
Matthias Clasen [Sat, 26 Feb 2022 23:55:16 +0000 (18:55 -0500)]
Add a test for filterlistmodel sections
Matthias Clasen [Sat, 26 Feb 2022 19:59:43 +0000 (14:59 -0500)]
filterlistmodel: Support sections
Propagate sections from the child model to
the filter model.
Benjamin Otte [Sat, 19 Feb 2022 02:39:13 +0000 (03:39 +0100)]
sortlistmodel: add a fast path for get_section()
Benjamin Otte [Sat, 19 Feb 2022 01:37:53 +0000 (02:37 +0100)]
testsuite: Add section tests to sortlistmodel test
Matthias Clasen [Sat, 26 Feb 2022 23:52:49 +0000 (18:52 -0500)]
Add a test for sortlistmodel sections
Benjamin Otte [Fri, 18 Feb 2022 02:21:48 +0000 (03:21 +0100)]
sortlistmodel: Implement GtkSectionModel
The get_section() implementation is a slow and steady implementation
that has to be careful to not screw up when an incremental sort is only
partially sorted.
Benjamin Otte [Sat, 26 Feb 2022 03:52:16 +0000 (04:52 +0100)]
flattenlistmodel: Implement GtkSectionModel
Each child model is reported as one section.
Benjamin Otte [Tue, 15 Feb 2022 01:03:00 +0000 (02:03 +0100)]
Implement GtkSectionModel for all selection models
Benjamin Otte [Mon, 14 Feb 2022 02:18:36 +0000 (03:18 +0100)]
Add GtkSectionModel
Prototyping the interface to be used for sections in listview, so people
can review and play with it.
Benjamin Otte [Wed, 5 Apr 2023 23:55:18 +0000 (01:55 +0200)]
testsuite: Add a listitemmanager test
Matthias Clasen [Tue, 9 May 2023 14:48:18 +0000 (14:48 +0000)]
Merge branch 'reftest-msvc' into 'main'
reftest-snapshot.c: Fix running reftests on Visual Studio
See merge request GNOME/gtk!5937
Matthias Clasen [Tue, 9 May 2023 14:47:51 +0000 (14:47 +0000)]
Merge branch 'inspect-a11y-bounds' into 'main'
Inspector accessibility pane improvements
See merge request GNOME/gtk!5941
Benjamin Otte [Tue, 9 May 2023 12:41:24 +0000 (14:41 +0200)]
wayland: Don't assert when requesting another frame
When GDK_DEBUG=no-vsync is on, we might have more than one outstanding
frame. Don't assert when that hapens. Just request a frame callback for
the first and skip the others.
Benjamin Otte [Tue, 9 May 2023 12:38:24 +0000 (14:38 +0200)]
frameclock: Keep more history
Not all frames get timing info with GDK_DEBUG=no-vsync, so make sure
that even when we render tons of frames, the one frame that does get
timing info is still there when the timing info arrives.
I set it to 128 from 16 now.
This is roughly good enough to go to 5000fps from on a 60Hz monitor.
Benjamin Otte [Tue, 9 May 2023 12:36:45 +0000 (14:36 +0200)]
gtk-demo: Update fps with higher priority
We want to make sure to always update it, so make sure it has a higher
priority than redraws.
Benjamin Otte [Tue, 9 May 2023 12:55:54 +0000 (14:55 +0200)]
gdk: Add GDK_DEBUG=no-vsync
That turns off waiting and freezing in the frame clock and in
surfaces and triggers redraws asap.
Benjamin Otte [Tue, 9 May 2023 12:55:59 +0000 (14:55 +0200)]
frameclock: Add gdk_frame_clock_idle_is_frozen()
This is in preparation for a future commit.
Also turn some macros into inline functions and pass the clock instead
of its priv pointer.
Matthias Clasen [Tue, 9 May 2023 12:51:29 +0000 (12:51 +0000)]
Merge branch 'fix-wgl' into 'main'
Few fixes for WGL after recent updates
Closes #5808
See merge request GNOME/gtk!5936
Matthias Clasen [Tue, 9 May 2023 11:29:11 +0000 (11:29 +0000)]
Merge branch 'fix-gsk-crash' into 'main'
Fix a refcounting mishap in gsk
Closes #5823
See merge request GNOME/gtk!5943
Matthias Clasen [Tue, 9 May 2023 10:51:53 +0000 (06:51 -0400)]
gsk: Fix a refcounting mishap
This may fix #5823
Chun-wei Fan [Tue, 9 May 2023 10:10:34 +0000 (18:10 +0800)]
gdkglcontext-win32-wgl.c: Always request alpha bits
...when we are using wglChoosePixelFormatARHB(). This ensures that we
hvae a HDC with a pixel format that will really support alpha bits, as
we did for the traditional ChoosePixelFormat().
Thanks to Patrick Zacharias for testing and pointing things out.
Benjamin Otte [Sat, 29 Apr 2023 05:00:45 +0000 (07:00 +0200)]
win32: Turn off multisampling
We know we draw beautiful Windows, but drawing them once is totally
enough.
Benjamin Otte [Wed, 26 Apr 2023 23:27:50 +0000 (01:27 +0200)]
win32: Actually increase the counter
We were sending random junk to ChoosePixelFormat().
Also assert that we don't overflow the array. That might be usefu to
know if we carelessly add attributes later.
Chun-wei Fan [Thu, 4 May 2023 16:09:43 +0000 (00:09 +0800)]
GDK/Win32: Use dummy WGL contexts more
... for creating the actual WGL contexts, so that we can cut down on the
number of times where we need to create the base, legacy WGL contexts in
order to create the WGL contexts with attributes. We could just use the
dummy context that we have to make it current to create the needed
WGL contexts.
Chun-wei Fan [Wed, 3 May 2023 09:58:54 +0000 (17:58 +0800)]
gdkglcontext-win32-wgl.c: Fix using wglChoosePixelFormatARB()
If we are querying the best supported pixel format for our HDC via
wglChoosePixelFormatARB() (i.e. we have the WGL_ARB_pixel_format extension),
it may return a pixel format that is different from the pixel format that we
used for the dummy context that we have setup, in order to, well, run
wglChoosePixelFormatARB(), which sadly requires a WGL context (HGLRC) to be
current in order to use it, which means the dummy HDC already has a pixel
format that has been set (notice that each HDC is only allowed to have its
pixel format to be set *once*). This is notably the case on Intel display
drivers.
Since we are emulating surfaceless GL contexts, we are using the dummy GL
context (and thus dummy HDC that is derived from the notification HWND used in
GdkWin32Display) for doing that, we would get into trouble if th actual HDC
from the GdkWin32Surface has a different pixel format set.
So, as a result, in order to fix this situation, we do the following:
* Create yet another dummy HWND in order to grab the HDC to query for the
capabilities the GL drivers support, and to call wglChoosePixelFormatARB() as
appropriate (or ChoosePixelFormat()) for the final pixel format that we use.
* Ditch the dummy GL context, HDC and HWND after obtaining the pixel format.
* Then set the final pixel format that we obtained onto the HDC that is derived
from the HWND used in GdkWin32Display for notifications, which will become our
new dummy HDC.
* Create a new dummy HGLRC for use with the new dummy HDC to emulate surfaceless
GL support.
Luca Bacci [Tue, 9 May 2023 07:59:17 +0000 (07:59 +0000)]
Merge branch 'fix-5808' into 'main'
gdkglcontext-win32-wgl.c: Fix WGL context realization on 32-bit
Closes #5808
See merge request GNOME/gtk!5942
Chun-wei Fan [Mon, 8 May 2023 02:59:51 +0000 (10:59 +0800)]
gdkglcontext-win32-wgl.c: Fix WGL context realization on 32-bit
We are currently using g_clear_pointer() on the intermediate WGL contexts
(HGLRC)'s that we need to create in the way, which means that we need to ensure
that the correct calling convention for wglDeleteContext() is being applied.
To be absolutely safe about it, use the gdk_win32_private_wglDeleteContext()
calls, which will in turn call wglDeleteContext() directly from opengl32.dll
(using the OpenGL headers from the Windows SDK) instead of going via libepoxy,
which will assure us that the correct calling convention is applied.
Fixes issue #5808.
Jason Francis [Tue, 9 May 2023 01:33:04 +0000 (21:33 -0400)]
inspector: show GtkAccessible bounds in a11y pane
Jason Francis [Tue, 9 May 2023 01:30:10 +0000 (21:30 -0400)]
inspector: hide a11y object path when not using AT-SPI backend
Benjamin Otte [Mon, 8 May 2023 18:19:03 +0000 (18:19 +0000)]
Merge branch 'wip/otte/for-main' into 'main'
Revert "TEST TEST TEST"
See merge request GNOME/gtk!5940
Benjamin Otte [Mon, 8 May 2023 17:51:08 +0000 (19:51 +0200)]
Revert "TEST TEST TEST"
This reverts commit
97aff588d553b147220a1af8eb0d5fad08fefcf3.
That was a test commit for demonstrating things that shouldn't have made
it into main.
So throw it out again.
Emmanuele Bassi [Mon, 8 May 2023 15:07:26 +0000 (15:07 +0000)]
Merge branch 'new-align-value' into 'main'
filterchooser: Maintain filter combo visibility
See merge request GNOME/gtk!5939
Matthias Clasen [Mon, 8 May 2023 14:35:37 +0000 (10:35 -0400)]
filterchooser: Maintain filter combo visibility
This code broke when it was ported to listmodels
since we always have a non-NULL lis model of filters.
It can just be empty.
Matthias Clasen [Mon, 8 May 2023 13:28:15 +0000 (13:28 +0000)]
Merge branch 'new-align-value' into 'main'
Introduce GTK_ALIGN_BASELINE_FILL/CENTER
See merge request GNOME/gtk!5935
Matthias Clasen [Mon, 8 May 2023 11:00:34 +0000 (07:00 -0400)]
testsuite: Handle duplicate enum values
Our notify tests would fall over if there was
a duplicate enum value (within the first 10 values).
Make it handle that, by skipping the duplicate value.
Chun-wei Fan [Mon, 8 May 2023 07:10:17 +0000 (15:10 +0800)]
reftest-snapshot.c: Fix running reftests on Visual Studio
On Visual Studio-like builds, shove a ".dll" suffix for the reftest
module so that we look for the correct DLL when running the tests.
Matthias Clasen [Mon, 8 May 2023 01:26:09 +0000 (21:26 -0400)]
Introduce GTK_ALIGN_BASELINE_FILL/CENTER
It turns out that the old behavior of GTK_ALIGN_BASELINE
was actually used in libadwaita, so bring it back, and
introduce a new GtkAlign value for the new behavior.
Matthias Clasen [Sun, 7 May 2023 13:59:42 +0000 (13:59 +0000)]
Merge branch 'node-editor-smarts' into 'main'
node-editor: Add some editor smarts
See merge request GNOME/gtk!5934
Matthias Clasen [Sun, 7 May 2023 13:32:09 +0000 (09:32 -0400)]
Cleanup
Matthias Clasen [Sun, 7 May 2023 02:31:58 +0000 (22:31 -0400)]
node-editor: Add some editor smarts
Allow control-clicks on some fields to bring up
a more specific UI. This functionality is also
available via Ctrl-E and the context menu.
At this point, it can edit colors, fonts and
files in some places, as well as a few enums.
Matthias Clasen [Sun, 7 May 2023 02:29:44 +0000 (22:29 -0400)]
Fix a typo in the node format definition