Move FFmpeg decode thread, frame buffering, and lifecycle management
out of CBackground into a dedicated CVideoBackend class. Background.hpp
no longer includes FFmpeg headers. CBackground interacts with video
purely via open(), swapFrame(), and frameW/H().
Extend the background widget to natively decode and display video
files (mp4, mkv, webm, avi, mov, gif, and more) using FFmpeg, with
no external tools required.
- Detection is extension-based; existing image paths are unaffected
- A background decode thread paces frames to their PTS timestamps
and publishes them to the render thread via an O(1) mutex-swap,
avoiding any memcpy per frame
- sws_getCachedContext handles codecs that only report their pixel
format after the first decoded frame
- Videos loop seamlessly via av_seek_frame at EOF
- blur_passes works on video frames the same as on images
- No new config keys: path = /path/to/video.mp4 is sufficient
Properly lock AWP<IWidget> weak pointers before calling onAssetUpdate()
to prevent use-after-free when widgets are destroyed during shutdown or
output removal. Guard timer callback against null g_asyncResourceManager.
Fix destruction order to join threads before resetting globals.
* core: use Hyprgraphics::CAsyncResourceGatherer
* core: move screencopy frame generation to the new resource manager
* core: introduce a dedicated onAssetUpdate callback
* check for unloaded before finished and some cleanup
* also allow for dynamic label resource deduplication
use a simple counter instead of a timestamp to allow the same
widget on a different monitor to reuse a text cmd resource.
I didn't do this before, because I was worried about two labels that use
the same command with different reload times. I mitigated that by just
incrementing the revision by the time interval. This should be sufficent
to avoid clashes.
* don't render within onAssetUpdate to avoid duplicate renders
another much improvement for multi monitor setups.
allows updating within the same frame for most labels.
* remove nvidia workaround :)
I tested and it seems like the resource manager revision makes
the nvidia workaround obsolete.
* widget: add click handling and point containment methods to IWidget interface
* core: add onClick method to handle mouse click events
- renderer: move getOrCreateWidgetsFor method declaration to public section
* core: update mouse event handling to track mouse location and button clicks
* widget: add onclick command handling and point containment to CLabel
- config: add onclick special config value to label
* assets: add label configuration for keyboard layout switching
* config: add onclick configuration for label widgets
- add CLICKABLE macro for onclick configuration
- replace direct onclick assignment with CLICKABLE macro
* core: fix cursor shape initialization and pointer handling
- ensure pointer is available before setting cursor shape
- initialize cursor shape device if not already done
* core: add hover handling and cursor shape updates
- implement onHover method to manage widget hover states
- update cursor shape based on hover status
- ensure all outputs are redrawn after state changes
* widgets: add hover state management and bounding box calculations
- add setHover and isHovered methods to manage hover state
- implement containsPoint method for hit testing
- override getBoundingBox in CLabel for accurate positioning
- add onHover method in CLabel to change cursor shape
* core: add hover handling in pointer motion
- invoke onHover method with current mouse location
* widgets: add hover handling and bounding box for password input field
- add getBoundingBox method to calculate the widget's bounding box
- implement onHover method to update cursor shape on hover
* widgets: update hover behavior for label widget
- modify cursor shape setting to only apply when onclickCommand is not empty
* core: optimize hover handling and rendering for lock surfaces
- Improve hover state tracking for widgets
- reduce unnecessary redraw calls by tracking hover changes
- remove redundant renderAllOutputs() call
* widgets: add onclick and hover to shape and image
* core: trigger hover and onclick only for the currently focused surface
* core: handle fractionalScale in onclick and hover
* core: don't trigger onclick or hover when hide_cursor is set
* misc: remove braces
* core: run onclick commands asnychronously
---------
Co-authored-by: Memoraike <memoraike@gmail.com>
Makes more sense than clearing the input buffer in the auth impl.
Also added a check for the password buffer length to reset the fail
color as soon as the password length > 0.
