`__CARGO_FIX_YOLO=1` is a hack, but it does help a lot with the tedious fixes where the result is fairly clear.
See https://rust-lang.github.io/rust-clippy/master/index.html#needless_lifetimes
```
__CARGO_FIX_YOLO=1 cargo clippy --fix --all-targets --workspace --exclude gstreamer-player --exclude i-slint-backend-linuxkms --exclude uefi-demo --exclude ffmpeg -- -A clippy::all -W clippy::needless_lifetimes
cargo fmt --all
```
`__CARGO_FIX_YOLO=1` is a hack, but it does help a lot with the tedious fixes where the result is fairly clear.
See https://rust-lang.github.io/rust-clippy/master/index.html#/needless_borrow
```
__CARGO_FIX_YOLO=1 cargo clippy --fix --all-targets --workspace --exclude gstreamer-player --exclude i-slint-backend-linuxkms --exclude uefi-demo --exclude ffmpeg -- -A clippy::all -W clippy::needless_borrow
cargo fmt --all
```
Added image-default-formats with all the format supported by image by
default, and enable that feature by default.
Also put that feature in compat-1-2 for compatibility with user that
have used image 0.24 with enabled features.
Make a new compat-1-10 feature that does not enable default format by
default
ChangeLog: upgraded image crate to 0.25, added a new cargo feature
to enable all image formats. (that feature is enabled by default with
compat-1-2, added compat-1-10 to disable it
Fixes https://github.com/slint-ui/slint/issues/7251
Previously, the dark mode button was especially hard to hit on
embedded devices.
The TouchArea of the SideBarIcon used for this button was much smaller
than the others. By exposing a new property on the SideBarIcon:
icon-colorize, we can customize the colorize behaviour of the Image,
as well as the source. This allows us to re-use the internal Image for
the dark mode button. This allows us to use the default width, height
and x position of the icon. The touch areas of the sidebar buttons are
now uniform.
Use a string instead of the JobStatus enum.
This makes the PrinterQueue code compatible with the printerdemo code,
allowing us to switch between mcu and non-mcu versions of this demo,
depending on what will fit on a device.
A string is used, rather than porting the non-mcu demo to an enum,
because the non-mcu demo needs to work with the cpp_interpreted demo.
Zephyr have recently added this device tree entry to provide a common
interface for input handlers.
This means we can remove the overlay required for the i.MX RT1170
shield, and remove our modifications some of our modifications for the
native sim. It also means that we don't have to make device tree
modifications to support other boards in the future.
This gives us access to the new `zephyr,touch` device tree entry,
which will allow us to produce a more generic touch input handler.
INPUT_CALLBACK_DEFINE and the input callback now require `void *`
argument for user data. For now we leave this unused.
When this driver is using the DMA API, we cannot do partial updates.
Previously, this code would then perform a full-screen DMA transfer for
each region the renderer changes. Now we make modifications to the pixel
data of the dirty regions, if required, and then do one fill-scren DMA
transfer of the final render buffer.
Workaround for #6715
In the wasm interpreted, we use an `<img>` tag to load images, but that
cannot get the default svg size without an explicit width and height
- Rename `accessible-selectable` and `accessible-selected` to `accessible-item-{selectable,selected}`.
Because the property is for items in list rather than eg Text
- Rename `accessible-position-in-set` to `accessible-item-index`.
- Rename `accessible-size-of-set` to `accessible-item-count` and move
the property to the container element rather than the item itself
(For me this was also reproducible on desktop, but oddly not for
Olivier)
Letters like é were stripped from the noto sans font used by the demo,
and with FemtoVG we assume that presense of a character in the common
script implies full coverage (to avoid having to check every character).
Fix this by fixing the font itself. We wouldn't want those letters to
come from DejaVu :)
Don't redraw the entire scene underneath every time the LEDs are
animated. Instead, cache it into one large layer. The scene is
reasonabley complex that this justifies the expense of the additional
memory throughput.
Continue to run build-on-demand.mjs to permit a fallback to a
from-source build for users who run `npm install slint-ui` on a platform
where we don't provide binaries for (such as linux-armv7/64). However,
this is only done when run from the installed package, not from git.
From git, the prestart trick remains in use.
Fix for incorrect sized SVG on web.
Fix for incorrect gradient on HVAC for web.
Fix for opacity animation not being triggered.
Fix for README having incorrect copy paste info.
