Timm Bäder [Thu, 12 Apr 2018 07:47:17 +0000 (09:47 +0200)]
gl renderer: Increase GArray size by 2 up front
Since we know we will add two elements anyway.
Timm Bäder [Thu, 12 Apr 2018 06:20:43 +0000 (08:20 +0200)]
widget: Avoid a dead assignment
Timm Bäder [Thu, 12 Apr 2018 06:20:28 +0000 (08:20 +0200)]
tooltip: Avoid some possibly uninitialized values
Aron Xu [Sat, 14 Apr 2018 05:10:27 +0000 (13:10 +0800)]
po: fix zh_CN translation
Benjamin Otte [Fri, 13 Apr 2018 00:13:27 +0000 (02:13 +0200)]
wayland: Don't freeze the frame clock too early
We can't freeze the frame clock on commit, but only after-paint,
otherwise the frameclock will resume in the paint stage.
So freeze the frame clock at the end of the frame if we are waiting for
a frame callback.
Note; The diff is only lage because of indentation changes due to
avoiding early returns in favor of a branch.
Benjamin Otte [Wed, 11 Apr 2018 23:42:21 +0000 (01:42 +0200)]
gdk: Remove gdk_surface_create_similar_image_surface()
It's unused.
Benjamin Otte [Wed, 11 Apr 2018 23:41:51 +0000 (01:41 +0200)]
gdk: Remove gdk_cairo_surface_create_from_pixbuf()
It's unused and people should use textures and snapshots anyway.
Benjamin Otte [Wed, 11 Apr 2018 23:40:38 +0000 (01:40 +0200)]
widget-factory: Use a pixbuf instead of a surface
This gets rid of the last user of
gdk_cairo_surface_create_from_pixbuf().
Benjamin Otte [Wed, 11 Apr 2018 23:29:15 +0000 (01:29 +0200)]
testgtk: Don't set cairo surfaces as icons
The code expects textures these days, so use those.
Benjamin Otte [Wed, 11 Apr 2018 23:02:33 +0000 (01:02 +0200)]
gdk: Remove gdk_cairo_get_drawing_context()
It's unused.
Benjamin Otte [Thu, 12 Apr 2018 11:59:55 +0000 (13:59 +0200)]
tests: Use gdk_texture_save_to_png() where appropriate
Benjamin Otte [Thu, 12 Apr 2018 11:50:33 +0000 (13:50 +0200)]
texture: Add gdk_texture_save_to_png()
It's needed for debugging Timm's code, so better have it here than
hidden in my random-patch vault.
Benjamin Otte [Wed, 11 Apr 2018 17:20:17 +0000 (19:20 +0200)]
drawingcontext: Remove unused APIs
Christoph Reiter [Wed, 11 Apr 2018 17:49:03 +0000 (19:49 +0200)]
ci: allow flatpak jobs to fail for now
We can easily revert this when things are fixed again.
Timm Bäder [Wed, 11 Apr 2018 17:14:09 +0000 (19:14 +0200)]
scrolledwindow: Fix indicator revealing
We need to increase the scrollbar bounds size.
Timm Bäder [Wed, 11 Apr 2018 16:09:30 +0000 (18:09 +0200)]
gl renderer: Slightly change ops_offset behavior
Add the given delta to the current one instead of replacing it.
Timm Bäder [Wed, 11 Apr 2018 15:57:57 +0000 (17:57 +0200)]
gl renderer: Don't draw all rounded clip nodes to a texture
We only need to do that if the rounde clip node intersects with the
outer one.
Руслан Ижбулатов [Wed, 11 Apr 2018 16:39:34 +0000 (16:39 +0000)]
GDK W32: Fix a missing weak referencing
The changes in
a82d67bb7d9290693e8d8c503e65694bb62b44ca didn't
preserve a g_object_weak_ref() call that we need to ensure the
objects in hash map don't become stale. Fix this.
Emmanuele Bassi [Wed, 11 Apr 2018 14:32:37 +0000 (14:32 +0000)]
Merge branch 'gitlab-ci-ccache-again' into 'master'
gitlab-ci: Reenable ccache but disable it during the initial meson call
See merge request GNOME/gtk!111
Emmanuele Bassi [Wed, 11 Apr 2018 13:37:57 +0000 (14:37 +0100)]
Simplify the Wayland code generation
Instead of going through an ancillary script to strip away the
`WL_EXPORT` annotation from the generated code, we should bump up the
required version of Wayland, and use the `private-code` argument for
wayland-scanner, which does the right thing for us.
Christoph Reiter [Wed, 11 Apr 2018 10:18:55 +0000 (12:18 +0200)]
gitlab-ci: Reenable ccache but disable it during the initial meson call
There where some problems (??) with ccache not detecting changes during meson
checks. Setting CCACHE_DISABLE during the meson execution makes ccache not use
the cache and pass things directly to the compiler.
Timm Bäder [Wed, 11 Apr 2018 09:20:50 +0000 (11:20 +0200)]
gl renderer: Render everyhing according to MV scale
And not the surface's scale factor. This way the magnifier works.
Timm Bäder [Wed, 11 Apr 2018 08:10:38 +0000 (10:10 +0200)]
gl renderer: Remove unused field
Emmanuele Bassi [Wed, 11 Apr 2018 12:58:47 +0000 (13:58 +0100)]
ci: Update container to Fedora 28
We're going to need updated dependencies that we cannot build as
subprojects, like wayland-protocols.
Benjamin Otte [Wed, 11 Apr 2018 02:04:37 +0000 (04:04 +0200)]
widget: Not all toplevels are containers
Guard against GtkInvisible.
Benjamin Otte [Wed, 11 Apr 2018 01:08:55 +0000 (03:08 +0200)]
widget: Only start/stop idle sizers for toplevels
They are the only widgets that have idle sizers.
Benjamin Otte [Wed, 11 Apr 2018 01:03:35 +0000 (03:03 +0200)]
widget: Move frame clock handling to vfunc
Instead of connecting to / disconnecting from the frame clock, do it
inside the vfuncs next to changing the priv->realized boolean.
This removes a race between those 2 cases that could cause child
widgets' unrealize handlers to reconnect this widget to the frame clock
because it was still marked as realize when the widget had already
disconnected from the frame clock.
Fixes #168
Benjamin Otte [Wed, 11 Apr 2018 00:48:43 +0000 (02:48 +0200)]
widget: Remove gtk_widget_set_realized()
Everybody chains up now.
Benjamin Otte [Wed, 11 Apr 2018 00:43:19 +0000 (02:43 +0200)]
gtk: Chain up in toplevels' realize
Don't call set_realized() manually.
Benjamin Otte [Wed, 11 Apr 2018 00:38:00 +0000 (02:38 +0200)]
widget: Make gtk_widget_real_realize() stricter
But in turn, also allow it to work on widgets with their own surface.
This way, we can chain up from everywhere and won't have to export
gtk_widget_set_realized().
Benjamin Otte [Wed, 11 Apr 2018 00:33:23 +0000 (02:33 +0200)]
widget: Assert unrealizing
Instead of just setting the flag, assert that it was set.
Benjamin Otte [Tue, 10 Apr 2018 22:08:33 +0000 (00:08 +0200)]
widget: Remove unused member
Benjamin Otte [Tue, 10 Apr 2018 20:20:10 +0000 (22:20 +0200)]
container: Remove gtk_container_maybe_start_idle_sizer()
Fold it into gtk_container_start_idle_sizer()
Benjamin Otte [Tue, 10 Apr 2018 20:08:10 +0000 (22:08 +0200)]
container: Remove gtk_container_queue_resize_handler()
It's the same as gtk_container_maybe_start_idle_sizer()
Benjamin Otte [Tue, 10 Apr 2018 20:07:36 +0000 (22:07 +0200)]
container: Use maybe_start_idle_sizer()
That way, we don't accidentally start an idle sizer for unrealized
widgets.
Benjamin Otte [Tue, 10 Apr 2018 20:02:31 +0000 (22:02 +0200)]
window: Don't queue resize handler
It's not needed, because the queue_allocate() call right above it does
this work.
Benjamin Otte [Tue, 10 Apr 2018 20:01:18 +0000 (22:01 +0200)]
widget: Don't queue resize handler twice
We already queue it when we set_alloc_needed(), so there's no need to do
it again.
Benjamin Otte [Tue, 10 Apr 2018 19:45:09 +0000 (21:45 +0200)]
container: Don't store the resize clock
Use gtk_widget_get_frame_clock() instead
Benjamin Otte [Tue, 10 Apr 2018 18:59:49 +0000 (18:59 +0000)]
Merge branch 'gitlab-ci-mingw' into 'master'
gitlab-ci: mingw build
See merge request GNOME/gtk!105
Christoph Reiter [Fri, 6 Apr 2018 16:15:38 +0000 (18:15 +0200)]
gitlab-ci: Add a 32bit mingw build
Emmanuele Bassi [Tue, 10 Apr 2018 13:51:27 +0000 (14:51 +0100)]
Terminate strncpy() buffers correctly
When using strncpy() with a buffer we need to account for the
terminating NUL character. GCC 8 started warning when using PPD_MAX_NAME
as the buffer length for strncpy() because the buffer we're copying into
has the same length — which means that the terminating NUL may be
skipped if the source string has a length of PPD_MAX_NAME.
The appropriate way to handle the case where we're copying a source with
a length bigger than of PPD_MAX_NAME is, as reported in the strncpy()
documentation, to copy `PPD_MAX_NAME - 1` bytes, and explicitly NUL
terminate the destination buffer. This has the additional benefit of
avoiding the compiler warning.
Benjamin Otte [Tue, 10 Apr 2018 13:10:56 +0000 (15:10 +0200)]
wayland: Factor out frame timings request code
Also call the code everywhere we draw and not just with Cairo.
Benjamin Otte [Tue, 10 Apr 2018 12:22:16 +0000 (14:22 +0200)]
filechooserbutton: Fix reserved pointers
So gcc stops complaining about unnecessary parenthesis.
Benjamin Otte [Tue, 10 Apr 2018 12:21:50 +0000 (14:21 +0200)]
ffmpeg: Fix compile warnings for newest version
Timm Bäder [Tue, 10 Apr 2018 07:53:48 +0000 (09:53 +0200)]
vulkancontext: Add new error enum
Timm Bäder [Sun, 8 Apr 2018 09:35:39 +0000 (11:35 +0200)]
widgetfocus: Use gtk_widget_compute_bounds
Instead of gtk_widget_get_allocation.
Timm Bäder [Sat, 7 Apr 2018 11:37:24 +0000 (13:37 +0200)]
widget: Remove _get_own_allocation
Replace all usages of it with _compute_bounds
Timm Bäder [Sat, 7 Apr 2018 08:13:36 +0000 (10:13 +0200)]
widget: Remove gtk_widget_get_own_allocation
Use compute_bounds everywhere.
Timm Bäder [Sat, 7 Apr 2018 08:10:57 +0000 (10:10 +0200)]
widget: Remove get_own_allocation usages
Timm Bäder [Sat, 7 Apr 2018 08:02:02 +0000 (10:02 +0200)]
notebook: Use compute_bounds
Timm Bäder [Sat, 7 Apr 2018 07:29:17 +0000 (09:29 +0200)]
popover: Use gtk_widget_compute_bounds
Instead of the private get_own_allocation
Timm Bäder [Sat, 7 Apr 2018 07:26:31 +0000 (09:26 +0200)]
toolbar: Avoid a get_own_allocation call
We want the content allocation here since we are allocation child
widgets.
Timm Bäder [Sat, 7 Apr 2018 07:26:12 +0000 (09:26 +0200)]
widget: Fix typo in compute_bounds docs
Timm Bäder [Sat, 7 Apr 2018 08:05:38 +0000 (10:05 +0200)]
gdk: Protect against negative GDK_SCALE values
Using those should not happen.
Руслан Ижбулатов [Mon, 9 Apr 2018 20:16:23 +0000 (20:16 +0000)]
GDK W32: use clock after_paint signal for timing updates
* Previous commit had misleading info. The code was
added to begin_paint() instead of end_paint(). Though
that did not affect its performance in any visible way.
* Company advised to move the code to an "after_paint" signal
handler, so that it works on all renderers, not just Cairo.
This change caused high fluctuation in FPS values in fishbowl
when it is put in a situation where it cannot achieve 60fps
(such as using Cairo renderer at ultra-high resolution).
This seems to be deliberate and not a bug.
Emmanuele Bassi [Mon, 9 Apr 2018 19:10:47 +0000 (20:10 +0100)]
docs: Link to GResource
Since we're referring to GLib's functionality, might as well save some
time to the reader, and link to the GResource API reference page.
Daniel Boles [Mon, 9 Apr 2018 19:55:14 +0000 (20:55 +0100)]
docs/reference/gtk/getting_started: cgit => GitLab
master version of
b685eb177932c021975a4b337e37d5911c537de0
Руслан Ижбулатов [Mon, 9 Apr 2018 19:07:10 +0000 (19:07 +0000)]
GDK W32: set update frequency and timestamp
There is no easily apparent way of being notified when frame updates
happene exactly, so we just query frame info at the end of each paint.
If we query too often (faster than DWM refresh rate), we just get
the same values twice in a row, but that is, hopefully, highly unlikely.
Daniel Boles [Mon, 9 Apr 2018 18:43:45 +0000 (19:43 +0100)]
AccelGroup: Remove @See_also of deleted function
gtk_item_factory_new() is not a thing that exists.
Also, maybe the lack of space after the colon was what stopped the link
from rendering.
Benjamin Otte [Mon, 9 Apr 2018 18:19:17 +0000 (20:19 +0200)]
css: Clipping changes don't need a queue_allocate() anymore
Now that queue_draw() isn't restricted to clip anymore, we don't need to
care about clip in the CSS engine either.
We do keep GTK_CSS_AFFECTS_CLIP around though because GtkWindow does
care for the window's size.
Benjamin Otte [Mon, 9 Apr 2018 12:45:04 +0000 (14:45 +0200)]
entry: Stop being activatable
Activating entries does not do anything.
Activate the default widget instead.
Daniel Boles [Mon, 9 Apr 2018 17:25:08 +0000 (17:25 +0000)]
Merge branch 'master' into 'master'
gdk_x11_display_get_monitor: fix monitor number test logic
See merge request GNOME/gtk!107
Sébastien Villemot [Mon, 9 Apr 2018 13:36:36 +0000 (15:36 +0200)]
gdk_x11_display_get_monitor: fix monitor number test logic
When asked for a nonexistent (positive) monitor number,
gdk_x11_display_get_monitor would (at best) return an uninitialized pointer,
instead of returning NULL.
Benjamin Otte [Sun, 8 Apr 2018 22:58:31 +0000 (00:58 +0200)]
glcontext: Store the buffer age regions in the GL context
That way, we can store the right region there: The actual painted area
instead of the exposed area (which is way too small).
Also, the GL context is the only user of this data, so storing it there
seems way smarter.
Benjamin Otte [Sun, 8 Apr 2018 22:40:47 +0000 (00:40 +0200)]
vulkan: Use new resize vfunc to recreate swapchain
... instead of checking sizes for every frame.
Benjamin Otte [Sun, 8 Apr 2018 22:35:32 +0000 (00:35 +0200)]
drawcontext: Add a surface_resized() vfunc
Call this vfunc whenever the surface's size has changed.
Benjamin Otte [Sun, 8 Apr 2018 22:25:03 +0000 (00:25 +0200)]
surface: Track all draw contexts created for the surface
Benjamin Otte [Sun, 8 Apr 2018 18:09:45 +0000 (20:09 +0200)]
widget: Don't do a 0-size check before snapshotting
Even widgets with an empty allocation may still want to draw stuff.
Examples include shadows or child widgets with negative margins.
Fixes GtkEntry's progressbar not showing up anymore.
Daniel Boles [Sun, 8 Apr 2018 17:39:59 +0000 (18:39 +0100)]
testmodelbutton: Test :centered, :icon, :inverted
Test more stuff.
Daniel Boles [Sun, 8 Apr 2018 17:31:03 +0000 (18:31 +0100)]
testmodelbutton: Fix wrongly overridding MB child
I was setting a GtkButton:label, which resulted in the ModelButton's own
custom children getting lost. That doesn't make for a useful test...
Daniel Boles [Sun, 8 Apr 2018 17:13:33 +0000 (18:13 +0100)]
ModelButton: Use WARN_INVALID_PROPERTY, not assert
Do the same thing we do everywhere else for invalid properties.
Daniel Boles [Sun, 8 Apr 2018 16:08:11 +0000 (17:08 +0100)]
ModelButton: Fix type Wether => Whether
Daniel Boles [Sun, 8 Apr 2018 15:07:35 +0000 (16:07 +0100)]
testmodelbutton: Add explanatory comments to CSS
and load at an earlier, more sensible place, before creating any widgets
Daniel Boles [Sun, 8 Apr 2018 11:43:07 +0000 (12:43 +0100)]
ModelButton: Protect against nonsensical gbooleans
the same way we do normally.
Daniel Boles [Sun, 8 Apr 2018 12:07:38 +0000 (13:07 +0100)]
tests/testmodelbutton: Add this
and test whether :action-name, :role, and CSS nodes work as expected
https://gitlab.gnome.org/GNOME/gtk/issues/163
Timm Bäder [Sun, 8 Apr 2018 07:30:24 +0000 (07:30 +0000)]
Merge branch 'wip/sadiq/fixes' into 'master'
listbox: Fix signal name is documentation
See merge request GNOME/gtk!106
Mohammed Sadiq [Sun, 8 Apr 2018 05:07:31 +0000 (10:37 +0530)]
listbox: Fix signal name in documentation
Rico Tzschichholz [Fri, 6 Apr 2018 14:20:48 +0000 (16:20 +0200)]
Fix some g-i annotation warnings
Timm Bäder [Thu, 5 Apr 2018 18:47:25 +0000 (20:47 +0200)]
menu: Avoid an invalid read
Timm Bäder [Sat, 31 Mar 2018 15:30:03 +0000 (17:30 +0200)]
lockbutton: Remove priv pointer
Timm Bäder [Sat, 31 Mar 2018 12:54:14 +0000 (14:54 +0200)]
widgetbowl: Add switch demo
Benjamin Otte [Thu, 5 Apr 2018 22:42:17 +0000 (00:42 +0200)]
gtk-demo: Redo viewing of resources
1. Don't rely on GtkImage as a detector of file type anymore.
Instead, hardcode all extensions that are in use.
2. Add a display method for videos.
Benjamin Otte [Thu, 5 Apr 2018 22:41:44 +0000 (00:41 +0200)]
video: Add autoplay and loop boolean properties
Piotr Drąg [Thu, 5 Apr 2018 20:54:03 +0000 (22:54 +0200)]
Update POTFILES.in
Carlos Garnacho [Fri, 23 Mar 2018 17:24:58 +0000 (18:24 +0100)]
gtkmenushell: Port to using GtkEventControllerKey
Carlos Garnacho [Fri, 23 Mar 2018 15:46:40 +0000 (16:46 +0100)]
gtkcellrendereraccel: Port to using GtkEventControllerKey
Carlos Garnacho [Fri, 23 Mar 2018 15:22:05 +0000 (16:22 +0100)]
gtkentry: Use key controller to track focus changes
Carlos Garnacho [Fri, 23 Mar 2018 15:09:37 +0000 (16:09 +0100)]
gtkcalendar: Use key controller to track focus changes
Carlos Garnacho [Fri, 23 Mar 2018 15:02:44 +0000 (16:02 +0100)]
gtktextview: Use key controller for focus changes
Carlos Garnacho [Thu, 22 Mar 2018 17:32:19 +0000 (18:32 +0100)]
gtkeventcontrollerkey: Add ::focus-in/out signals
And handle GDK_FOCUS_CHANGE events in order to emit those.
Carlos Garnacho [Thu, 22 Mar 2018 16:54:52 +0000 (17:54 +0100)]
gtkeventcontrollerkey: Add get_group() call
Callers can use this function on a key-pressed/released signal
to find out the key event group, useful in a few places.
Carlos Garnacho [Sun, 11 Mar 2018 12:56:32 +0000 (13:56 +0100)]
gtkmenu: Stop using ::key-press-event
Use GtkEventControllerKey for the task
Carlos Garnacho [Sun, 11 Mar 2018 12:56:21 +0000 (13:56 +0100)]
gtkiconview: Stop using ::key-press-event
Use GtkEventControllerKey for the task
Carlos Garnacho [Sun, 11 Mar 2018 12:56:02 +0000 (13:56 +0100)]
gtkcellrendererspin: Stop using ::key-press-event
Use GtkEventControllerKey for the task
Carlos Garnacho [Sun, 11 Mar 2018 12:55:45 +0000 (13:55 +0100)]
appchoooserdialog: use search bar key capture API
Carlos Garnacho [Sun, 11 Mar 2018 12:54:54 +0000 (13:54 +0100)]
inspector: Use search bar API to capture key events from the toplevel
Carlos Garnacho [Sun, 11 Mar 2018 12:53:17 +0000 (13:53 +0100)]
gtksearchbar/entry: Add [gs]et_key_capture_widget() API calls
This lets these widgets actively pull events from a widget, instead
of passively being fed events.
Carlos Garnacho [Sun, 11 Mar 2018 12:47:11 +0000 (13:47 +0100)]
gtkwindow: Perform key activation in gtk_main_do_event()
In a similar spirit to the previous commit, make window shortcut
activation happen by default on toplevels, without relying on
legacy controllers.
Carlos Garnacho [Sun, 11 Mar 2018 12:44:37 +0000 (13:44 +0100)]
gtkwidget: Activate key bindings through run_controllers()
Deferring a bit further making those a standalone controller, make
binding activation happen on run_controllers(), so it happens by
default on widgets (unless the key event was consumed earlier)
without the need of a legacy event controller.
Carlos Garnacho [Sun, 11 Mar 2018 12:41:10 +0000 (13:41 +0100)]
gtkwidget: Break early if non-gesture controllers handle the event
Non gesture controllers have no means to collaborate with other
controllers, thus should be considered standalone entities. It makes
no sense to propagate any further if scroll/key controllers handled
the event.