Refactor GL rendering to prepare for GTK4
- Remove FlGLArea - we can just use GtkGLArea directly
- Stop using gdk_cairo_draw_from_gl, it's not available in GTK4.
- Rename FlRendererGL to FlRendererGdk, this was implying that
FlRenderer could be something other than OpenGL which is not the case
currently.
- Remove unnecessary virtual methods in FlRenderer, just implement them
as standard methods.
Related: https://github.com/flutter/flutter/issues/143408
This PR changes the format check on CI to use the command added in
https://github.com/flutter/engine/pull/50747.
Additionally, while making this change, I noticed that the CI check was
not checking the formatting of all files, and that as a result, files
were present in the repo with incorrect formatting. I have fixed the
formatting and fixed the check to always check all files.
Generated by https://github.com/flutter/engine/pull/48903 (`dart ./tools/header_guard_check/bin/main.dart --fix`).
As discussed with @cbracken and @jmagman, the guards are not technically needed on the Mac/iOS code, but they (a) do not hurt and (b) still provide value if for some reason `#include` is used instead of `#import` (though I suspect we could try to add that to the tool in the future as well).
## Description
This moves the state update to only happen on realizing the window instead of at initialization time, based on [the comment from](https://github.com/flutter/flutter/issues/137262#issuecomment-1792020246) @robert-ancell .
## Related Issues
- https://github.com/flutter/flutter/issues/137262
## Tests
- I tried to add tests, but it doesn't seem possible to create a view without an actual display connected (and making a mock of it defeats the purpose of the test). I'm happy to be proven wrong, though.
This PR takes a step back and changes the GTK IM client window back from FlView to GtkApplicationWindow as it was before #33111. The window was originally changed to FlView to make the code testable by cutting a dependency to gtk_widget_translate_coordinates(). The change was hard to revert because there were several conflicting changes on top. Therefore, this PR introduces an FlTextInputViewDelegate to be able to provide coordinate mapping in a testable way.
Fixes: flutter/flutter#108832
* Make FlAccessibleNode derivable
This allows introducing specialized subclasses that implement such a11y
interfaces as AtkEditableText that should not be implemented for the
generic node.
* FlAccessibleNode: add set value, text selection & perform action
Prepare API required for the upcoming AtkEditableText implementation.
* FlViewAccessible: postpone child node creation
This changes the a11y node creation flow so that (unknown) child nodes
are not pre-created when their parent is updated but "pending" child
node IDs are collected for later. Parent-child relationships are
established at the end of update batches when all nodes have been
created.
* Add FlAccessibleTextField that implements AtkEditableText
Solves: flutter/flutter#103191
* FlEngine: add fl_engine_update_accessibility_features()
* FlSettings: add API for high-contrast & animations
* FlEngine: allow passing mock messenger at construction time
* FlSettingsPlugin: report accessibility features
Let MockBinaryMessenger internally manage the FlBinaryMessenger wrapper
instance and provide it via operator FlBinaryMessenger*() to avoid
having to create two objects in every test. The same technique was used
for MockSettings in #32618.
This PR is a continuation of #33111 and includes necessary changes to test the `TextInputType.setClient()` platform channel call. Testing the actual text input flow requires a mock IM context that will follow in a separate PR.
* [Linux] pass GdkWindow instead of FlView to FlTextInputPlugin
* [Linux] add MockBinaryMessenger::ReceiveMessage()
* [Linux] add test for TextInputType.setClient
* Drop the offscreen window
GdkOffscreenWindow requires display & screen which is a problem in
headless mode. Luckily, we don't really need a window instance in tests
because its only passed as a client window to the IM context which will
be mocked out when we get to tests that depend on it.
This PR makes the Linux key handling code a little more lenient when it comes to the order in which it receives responses to key events from the framework. I had assumed that there wasn't a case where responses could get out of order, but it seems that it is possible, given that you can mash on the keyboard and eventually get one out of order.
This changes the code so that instead of just looking at the first entry in the pending event deque, it searches the deque starting at the beginning to find the event, and remove it.
This changes the text handling so that keyboard events are sent to the framework first for handling, and then passed to the text input plugin, so that the framework has a chance to handle keys before they get given to the text field.
This is complicated by the async nature of the interaction with the framework, since GTK wants a synchronous response. So, in this change, I always tell GTK that the event was handled, and if it wasn't, then I re-dispatch the event once we know one way or the other.
Switches the Linux embedding from the standard C API to the new proctable version, to allow for unit testing of the embedding layer separately from the embedder APIs implementation.
We currently use a mix of C standard includes (e.g. limits.h) and their
C++ variants (e.g. climits). This migrates to a consistent style for all
cases where the C++ variants are acceptable, but leaves the C
equivalents in place where they are required, such as in the embedder
API and other headers that may be used from C.
Make a single-param ctor explicit in order to prevent surprising
implicit conversions.
Add a check for zero message-size and don't malloc/memcpy the incoming
message in those cases.
Add braces where they were missing.
Simplify FlDartProject by removing the path arguments - we'll assume that it's
in the standard bundle layout and add new methods later if we need to support
other cases.
