The type of thep property is `Point`, which existed before. It was
mapped to `slint::private_unstable_api::re_exports::Point` (euclid) and
is now mapped to slint::LogicalPosition (also in C++).
Turn the instance variable into a Promise itself, where the first
promise that's assigned to it creates the component and shows
it, before resolving the promise with it.
Some associated helper functions that use this.#instance - which is now a Promise - are also async.
In some call sites it doesn't matter, in others an await was added.
With the following snippet there was a race:
```
if (instance === null) {
instance = await create();
await instance.show();
} else {
instance.create_with_existing_window(...);
}
```
When awaiting for the promise that show() returns to settle, we would re-enter
this function (due to a broken recursion guard, that's now removed).
When re-entering, create_with_existing_window() is called, which - in order to
access the slint::Window to re-use - must create the window adapter.
Since that didn't happen from within an event loop invocation, winit would panic.
Consequently, this patch set also makes create_with_existing_window() return
a promise, as well as hide() (for the sake of it).
The broken recursion guard is easy to fix (with a .finally() handler setting
the boolean back), but that would mean we end up not always rendering
the very latest sources. Plus, the same issue exists in wasm_preview.ts.
Now, in case of subsequent re-entrancy in this code, we end up creating
a bunch of create_with_exiting_window() promises, which will eventually
settle and then we go back to the original show() call that will show the canvas.
Note: the addition of MinimalSoftwareWindow::set_size is there because
it would be a breaking change for user who called set_size on the
MinimalSoftwareWindow while also using the WindowAdapter trait.
This was the case of a test.
The same problem theorically exist with set_position and position, but
is unlikely to be a problem because i don't think people would use the
position with a MinimalSoftwareWindow
The renderer() is now public as well. That's because I want to make sure
that the scealed trait don't have non-provided method
The interpreter now creates the window typically in show(), so similar
to commit a8fcb5acd6 make show() return a
promise and invoke show() from within the event loop.
Amends d98c6773e1
When the TextInput item is read-only, it should not report availability
or any other status to the input method via the window adapter.
This also fixes the order of events when clicking on a TextInput: We
would send the ime position update request before enabling the ime,
because we set the focus _after_ setting the cursor pos.
Fixes#2812
With the latest Corrosion update test binaries don’t depend on libslint_cpp.so with path anymore that contains a slash. Therefore rpath kicks in and this workaround should not be needed anymore.
Plus, this gives us test coverage for rpath.
This patch adds a `close()` function that can be called to close a popup
window, and a `close-to-click` boolean that can be set to false to
disable the default behavior.
The added functions enable mixing colors and manipulating the opacity
of colors and brushes.
They enable the behavior of some of the available functions from SASS and are
added for future use for adding the Adwaita style (future PR).
After commit 7df902b53c, the winit window
is not created when calling show() anymore but it's now created at
component creation time. That means the event loop workaround removed
in 459f810bd8 is now needed at
construction time.
Since this is a winit and wasm specific issue, it's now dealt with in
the wasm interpreter implementation, by invoking the creation from the
event loop from there and returning a promise in the API.
This changes the API therefore: create() can only be called after the
event loop is running.
In the compiler this is still very primitive, but an attempt to start a
generic interface. The basic assumption is that all item functions will
eventually need access to the window adapter and itemrc. Support for
additional arguments is still missing.
Also missing is support for the function access via rtti in the
interpreter, hence the hardcoding at the moment.
