Instead of using datastructure that are tailored to our line by line
renderer and draw functions, use more generic and future proof data that
are easier to handle by hardware accelerator, and may be easier to
stabilize.
The screenshots from the screenshots test have to be re-done
because the rounding is different
Note: the C++ API DrawTextureArgs is not exposed yet
This adds a `unstable-wgpu-24` feature that exposes WGPU types in the
GraphicsAPI enum, adds `require_wgpu_24()` to the backend selector,
and adds a conversion from `wgpu::Texture` to `slint::Image`.
The `require_wgpu_24()` function in the selector will be extended in the
future (before the next release) to permit specifying additional aspects
of the WGPU configuration.
Co-Authored-By: Olivier Goffart <olivier.goffart@slint.dev>
... by changing the resolution for the `WindowItem` to traverse the
item tree from the current item, instead of going to the window.
This doesn't quite fix#4298 because `rem` resolution is still missing.
That requires the built-in default font size function to be fixed as
well, which is non-trivial.
cc #4298
Conversion from negative float to unsigned is saturating to 0 in rust
and undefined behavior in C++, we should therefore handle the case
properly
Fixes#8222
Commit cd6f2e2 reformated the .toml, but the 80 char width column is
judged too small to be practical
Add a .taplo.toml file
Also do not split feature array
* C++: add `notify_` prefix to the Model functions and add docs
The `notify_*` function are used by the model to notify the view.
As opposed to the function without the prefix, which are used to get
notified when a source model change
Fixes#3888
CC #3945 (for the docs about thread safety)
... using taplo with default settings
I tried this with 4 spaces indentation, but the patch is almost as
big as this one, so I went with default settings instead as that
is just easier:-)
Right now we always `lock().value()` it which is the equivalent of
`upgrade().unwrap()` in rust, this helps because it keeps the parent
alive when we are calling function in it.
Ideally we should also check that it wasn't deleted, but that's another
issue.
Fixes#7880
Most of the slint.h file is used for model code, so before doing some
work on the model i wanted to refactor the code in its own file.
Since the model need access to the ComponentHandle, i also had to move
that in its own file
I found `mise` a tool to help manage development projects.
https://mise.jdx.dev/
It basically can manage ENV vars based on the directory you are in,
manages tools you need (and makes thoser available based on the
directory you are in), and allows to define simple tasks that can
then be shared between developers.
Tools can be found in npm, pipx, aqua, binaries on github, ...
and it tries to verify signatures and all that (if supported by the
repo the data comes from).
I replaces the entire autofix workflow with mise tooling and tasks,
just to give it a try :-)
To reproduce:
```sh
> cargo install mise # to get the tool itself
# Follow the necessary step
# https://mise.jdx.dev/installing-mise.html#shells
# to intergate into your shell
> cd /your/slint/folder
# Mise will now ask whether or not to trust this dir and prints the
# command needed to do so. Run that.
> mise install # Install all the tools defined in .mise/config.toml
# Add a .mise.local.toml with local overrides. Git will ignore this file.
# Or add tasks into .mise/tasks/local ... Git will also ignore those.
> mise run 'ci:autofix:**:all' # To run all the ci:autofix tasks.
```
It is so much fatser to see these checks fail locally than it is to
bother CI with them :-)
This changes the default if the user has selected
`CONFIG_BSP_LCD_COLOR_FORMAT_RGB888` in their sdkconfig,
(which will result in build error if the buffer were passed as
Rgb565Pixel, or in garbage of the screen if the screen was configured
programatically to use rgb565 without calling `bsp_display_new` which is
unlikely if they set CONFIG_BSP_LCD_COLOR_FORMAT_RGB888)
