Commit Graph

61247 Commits

Author SHA1 Message Date
Matthias Clasen
14122d1acb x11: Export gdk_x11_surface_get_root_coords privately
This lets us avoid a roundtrip through the surface vfuncs.
2020-02-22 19:10:16 -05:00
Matthias Clasen
f93d0f8fb5 x11: Keep a ref on GdkDrag objects
It is expected that backends keep a ref on the GdkDrag
objects that they create as long as the drag is ongoing.
2020-02-22 19:02:13 -05:00
Matthias Clasen
833b564946 x11: Avoid crashes in dnd
We were forgetting to clean up the ::xevent signal
handler in some error cases. Move the signal connection
later, when we know the drag is going forward, and
use g_signal_connect_object to make sure the signal
handler is not forgotten.
2020-02-22 19:02:13 -05:00
Matthias Clasen
7edfcc37a3 Merge branch 'wip/wayland-fix-popup-grabs' into 'master'
Fix Wayland popup grabs

See merge request GNOME/gtk!1463
2020-02-22 20:51:32 +00:00
Jonas Ådahl
f0b0076a1c wayland/popup: Emit un-withdrawn event earlier
Otherwise grabbing the seat will appear to have failed.
2020-02-22 20:09:38 +01:00
Jonas Ådahl
71323a8b48 wayland/popup: Remove unnecessary checks when mapping popup
We only call this when we're not already mapped so don't check that. We
also only call this when we should be mapped so don't check that either.
2020-02-22 20:08:37 +01:00
Matthias Clasen
b8db9e124b Merge branch 'matthiasc/for-master' into 'master'
Fix a hard-to-spot typo

See merge request GNOME/gtk!1462
2020-02-22 19:01:18 +00:00
Matthias Clasen
ddd126740e Fix a hard-to-spot typo
The symptom caused by this was scales in
popovers not reacting to dragging.
2020-02-22 13:49:12 -05:00
Emmanuele Bassi
9bf03cfeec docs: Add migration notes for GtkFileChooser 2020-02-22 18:07:11 +00:00
Matthias Clasen
011f58fa64 Merge branch 'matthiasc/for-master' into 'master'
x11: Fix a crash in event handling

See merge request GNOME/gtk!1461
2020-02-22 16:41:29 +00:00
Matthias Clasen
1af94d0bce x11: Fix a crash in event handling
This was overlooked in 79b4510c6d.
2020-02-22 16:41:29 +00:00
Emmanuele Bassi
ff6772fd98 Remove preview widget from GtkFileChooser
The preview widget harks from a platform before time, when we didn't
have GIO, or a thumbnail specification.

Very few applications use it correctly, if at all; it has an horrid hack
to deal with the ownership of the widget's instance when accessed
through the getter function; it messes up the layout of the widget and
its label is less than useful when it comes to file names longer than a
dozen characters; it's a poor substitute for a proper thumbnail view.
2020-02-22 16:39:01 +00:00
Matthias Clasen
be25905a8d Fix gdk docs build 2020-02-22 10:44:21 -05:00
Emmanuele Bassi
e5ef26efa8 Remove GtkFileChooserEntry:local-only
We don't use it any more in GtkFileChooserWidget.
2020-02-22 15:22:06 +00:00
Emmanuele Bassi
ebd23737c0 Remove GtkPlacesSidebar:local-only
We don't use it any more in GtkFileChooserWidget.
2020-02-22 15:22:06 +00:00
Emmanuele Bassi
6e699e3f04 Remove GtkPlacesView:local-only
We don't use it any more from the file chooser widget.
2020-02-22 15:22:06 +00:00
Emmanuele Bassi
4a3742979d Remove GTK_FILE_CHOOSER_ACTION_CREATE_FOLDER
We don't need a whole separate action, now that the file chooser widget
can create folders; we can create use SELECT_FOLDER and create one.
2020-02-22 15:22:06 +00:00
Emmanuele Bassi
b09e7df81b Remove GtkFileChooser:local-only
Now that the whole API goes through GFile we don't have the weird split
between local-only and non-local-only modes.
2020-02-22 15:22:06 +00:00
Emmanuele Bassi
516eab5c43 Remove GtkFileChooser:show-hidden
This is a user setting and a user action, not something that ought to be
programmatically set via the file selection API.
2020-02-22 15:22:06 +00:00
Emmanuele Bassi
063ad28b1a Remove overwrite confirmation machinery from GtkFileChooser
Overwrite confirmation should not be optional, and it should not loop
into application code to create their own dialog and user response.
2020-02-22 15:22:06 +00:00
Emmanuele Bassi
d505573ee6 Drop GtkFileChooser:extra-widget
We have "choices" as a more rational (and portable) API; additionally,
the ownership semantics of the extra widget property getter are a hack.
2020-02-22 15:22:06 +00:00
Emmanuele Bassi
5f070ff233 Remove filename/URI API from GtkFileChooser
GtkFileChooser's API predates GIO by a few years, so it started off with
filenames and URI as character arrays. After introducing GIO as a
dependency, the API included GFile-based entry points.

It's much more appropriate to use GFile everywhere, as we want to
encourage people to use GIO instead of passing random bytes to low level
POSIX API.

See: #2455
2020-02-22 15:22:06 +00:00
Benjamin Otte
b002572824 Merge branch 'wip/otte/dnd' into 'master'
stuff

See merge request GNOME/gtk!1460
2020-02-22 14:53:13 +00:00
Matthias Clasen
767849d6d2 Remove GdkSeatGrabPrepareFunc from headers
It is not used in public api anymore.
2020-02-22 08:11:34 -05:00
Matthias Clasen
76e115b910 Add a warning when a grab fails
This reveals that the grabs for popover menus
always fail now, causing the menus to not properly
hide.
2020-02-22 08:11:34 -05:00
Benjamin Otte
8a098b1343 tests: Guard variables with necessary macros 2020-02-22 07:44:52 +01:00
Benjamin Otte
9d915ff431 treeviewcolumn: Inline variable into return_if_fail() 2020-02-22 07:44:52 +01:00
Benjamin Otte
6f0a0ce7f1 testsuite: Don't use g_return_if_fail()
Use g_assert()
2020-02-22 07:44:52 +01:00
Benjamin Otte
18d7187e4b testsuite: Don't use return_if_fail()
Use assert() instead
2020-02-22 07:35:23 +01:00
Benjamin Otte
d9e3eaaec8 treeview: Guard code by right compiler macros 2020-02-22 07:33:41 +01:00
Benjamin Otte
0c4c88b449 scrolledwindow: Inline variable into return_if_fail() 2020-02-22 07:33:41 +01:00
Benjamin Otte
2cb0007b89 iconview: Inline variables into return_if_fail() 2020-02-22 07:33:41 +01:00
Benjamin Otte
eb70173646 levelbar: Guard function by right compiler macros 2020-02-22 07:33:41 +01:00
Benjamin Otte
2e941acb29 filesystemmodel: Guard variable with correct compiler flags 2020-02-22 07:33:41 +01:00
Benjamin Otte
42d71f1016 combobox: Inline variables into return_if_fail() 2020-02-22 07:33:41 +01:00
Benjamin Otte
99fc31865a comboboxtext: Inline functions into return_if_fail() 2020-02-22 07:33:40 +01:00
Benjamin Otte
f9b7825a47 device: Inline function into return_if_fail() 2020-02-22 07:33:40 +01:00
Benjamin Otte
e3aee62b18 drop: Guard variable with correct compiler flags 2020-02-22 07:33:40 +01:00
Benjamin Otte
ce3ed45e81 surface: Inline function
Fixes compile error with -DG_DISBALE_CHECKS
2020-02-22 07:33:40 +01:00
Benjamin Otte
a12e563cd2 main: Don't synthesize crossing events when nothing changed
This was especially bad because it was confusing the event controllers
so much, they'd emit leave + enter events every time the mouse moved.
2020-02-22 07:33:40 +01:00
Benjamin Otte
a5e2275a72 stackswitcher: Use GdkDropControllerMotion
... and use one controller per button instead of using it on the
switcher and then going through lots of pain attempting to find the
right button for the location under the mouse.
2020-02-22 07:33:40 +01:00
Benjamin Otte
d680e2e344 Port simple cases to GtkDropControllerMotion 2020-02-22 07:33:40 +01:00
Benjamin Otte
2a8fd25a4b testsuite: Check proper notify emissions, too
This is in particular relevant for the ::is-focus property, because
updating that one doesn't cause enter/leave events.

But it also checks that notify and enter/leave happen in the right
order.
2020-02-22 07:33:40 +01:00
Benjamin Otte
34e5f36fdf Add GtkDropControllerMotion 2020-02-22 07:33:40 +01:00
Benjamin Otte
9223d1ee04 eventcontrollermotion: Fix docs 2020-02-22 07:33:40 +01:00
Benjamin Otte
fc43ec0bbc gtk: Bubble drag events like motion events
Emit crossing events - with a new GTK_CROSSING_DROP type - like we do
for motion events. There is no more special casing for them.

Note that the gesture has not been updated yet, so some obscure behavior
may occur.
2020-02-22 07:33:40 +01:00
Benjamin Otte
be4b34aa8f events: Use GdkDrop as event sequence
This allows treating drop events like touch events, which GTK groups by
event sequence.

It's a bit ugly that we just case the GdkDrop pointer, but event
sequences are only meant to be unique pointer ids, so it's fine.
2020-02-22 07:33:40 +01:00
Benjamin Otte
6f73a750a7 widget: Simplify adjust_allocation()
And in particular, only do it if the widget doesn't use ALIGN_FILL.

This avoids lots of measuring in the common case and speeds up
size_allocate() by about 25%.

And because size_allocate() is the bottleneck in the fishbowl, this also
gets ~25% more fishies.
2020-02-22 07:33:40 +01:00
Benjamin Otte
4128ee88b0 widget: Pull margin computation out of adjust_allocation
It's way cheaper to just do it.

Also simplifies adjust_size_allocation a lot.
2020-02-22 07:33:23 +01:00
Benjamin Otte
418bdc87ef sizerequest: Stop clamping for_size to natural size
Widgets should be given the actual size they will be allocated, so they
can do something with it.

If they want to clamp themselves to their natural size, nothing's
stopping them, they know their natural size after all.
2020-02-22 07:33:23 +01:00