Only the interpreter is implemented so far
MacOs won't work yet because we don't disable the default winit menubar
The viewer don't support removing the MenuBar yet
If the effectively drop shadow size would be zero, the FemtoVG and Skia item renderers would panic. The following test-case reproduces this:
```slint
export component Test inherits Window {
Rectangle {
drop-shadow-color: black;
drop-shadow-offset-x: 1px;
drop-shadow-blur: -0.5px;
width: 0.5px;
height: 0.5px;
background: blue;
}
}
```
ChangeLog: [Renderers] Fix panic with FemtoVG and Skia renderers for certain drop shadows.
This is the plumbing, where a `PopupWindow` directly becomes a child of wherever it's declared in, when show() is called. Further changes may be necessary to make this really useful, such as new roles or improved dialog handling. (see ticket for comments)
Fixes#7209
We shouldn't ungrab the mouse when the mouse distance is smaller than the threshold
Fixes#7152
ChangeLog: Fixed Flickable hard to scroll when starting slow
Based on API review, PlatformBuilder becomes BackendSelector with
a slightly smaller API surface but more options, such as selecting
Metal or Vulkan rendering.
Make it possible to programatically to switch to full-screen mode
via a new property in the Windows item.
The SLINT_FULLSCREEN environment variable will default set this
property to true. However settings this property in the slint code
will override the SLINT_FULLSCREEN.
Fixes#6665
We were actually skipping two element (the popup and the popup parent)
Also this broke the positioning of sub-popup.
The "fix" is to not get a parent for popup window.
This will also have the side effect of fixing the focus issue
- 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
This fixes drawing the background of the live preview with the software
renderer.
The problem was that when a svg image is tiled (and the svg is not
pre-rendered), we would render the svg in a duffer the size of the
target. But that's not the size at which it should be rendered as it is
going to be tiled over this surface.
Too big images are not supported by the software renderer, it would
panic in overflows in diverse places. So this also fixes panics
When a TextInput has a selection and a forward or backward movement
is performed (i.e. pressing Left or Right) without keeping the anchor
(i.e. Shift is not pressed), the selection collapses either to the
beginning or end of the selection based on the movement direction.
ChangeLog: Improved deselection behavior when pressing Left/Right in TextInput
Fixes#6511
* WIP: swrenderer: use fixed point for the pixmap font coordinate
* swrenderer: signed distance field: move the glyph to the middle
* swrenderer: round the advance instead of truncating it in distance field
* swrenderer: actually align the gplyph on the sub-pixel precision
sub-pixel within the source.
* swrenderer: adapt the threshold for signed distance field
sqrt(2) is the distance to the diagonal, seems like this gives sharper fonts
* Fix bug in the elision, and re-upload the screenshort
the screenshot changed because the afvanced is now rounded intead of
truncated
When rounded out, the dirty region might end up bigger than the size
of the screen, which would then result in panic while trying to draw a
pixel out of the screen.
Fixes#6932
ChangeLog: Added function to set the XDG app id on Wayland/X11. This needs to be added with respective function names in the language sections.
Fixes#1332
This also fix changed event not working as they rely on the init to the setup
Fixes#6836
ChangeLog: fix init and changed callback not always being called in ListView
That code was touched recently while doing the signed distance field
rendering. The offset is now within the source rectangle.
So the SharedBufferCommand's source_rect must include the whole pixmap.
Unfortunately, VectorFont are not covered by the screenshot tests
Popups are stored in a HashMap and are assigned an ID so popup.close(); closes the correct popup and so a single PopupWindow cannot be opened multiple times
Fixes#4356
Still not perfect:
- Calling several times `popup.show()` on the same popup, will open
that popup multiple times (instead of being a noop once opened)
- Calling `some-popup.close()` will always close the top of the stack,
without considering if it is the `some-popup` or another popup.
Both problems are because we don't remember whether a particular popup
is open and we don't associate `close()` with a particular popup
This currently doesn't have public API to enable it yet.
TODO:
- Error handling in the compiler
- Public API in the compiler configuration
- Documentation
Instead of using the trick of taking a sub-slice and pretending the
image was smaller than it is.
Then this is more uniform with all the other rendering path.
And this should prevent the panic in #6643Fixes#6643
