Added proper backend validation for both config.json and model.safetensors to confirm model availability. Updated frontend logic to use /matting/check-model response, preventing unnecessary download notifications.
Implemented drag-to-resize functionality for the output area with visual transform handles on corners and edges. Users can now interactively resize the output area by dragging handles instead of using dialogs, with support for grid snapping and aspect ratio preservation.
Implements one-click auto adjustment of output area to fit selected layers with intelligent bounding box calculation. Supports rotation, crop mode, flips, and includes automatic padding with complete canvas state updates.
Updated mask loading to immediately use available data from connected nodes and preserve existing masks if none is provided. Backend mask data is only fetched after workflow execution, ensuring no stale data is loaded during connection.
Enhanced LayerForge input handling to strictly separate image and mask loading based on connection type. Images now only load when allowImage=true and masks only when allowMask=true, preventing unintended cross-loading between input types.
Fixed an issue where images were not immediately refreshed after background removal (matting). Now, the canvas updates instantly when the background is removed, ensuring correct display without requiring manual scaling or other actions.
Refactored tooltip logic for the clipboard switch so it now updates immediately when toggled, showing the correct template without requiring mouse movement. Added helper functions and improved event handling for better UX.
Redesigned the positioning system for the Blend Mode menu, inspired by the "Custom Output Area" logic. The menu now anchors precisely to the top-right corner of the viewport and stays in place during panning and zooming.
Refactored the "Output Area Size" button to use the new setOutputAreaSize method:
Correctly sets output area size and position based on current system logic (custom shape, extensions, outputAreaBounds)
Fully functional across all editor modes
New resizing logic:
Operates relative to the current center of the output area
Output area expands or contracts around its center without repositioning
Ensures center remains unchanged as expected
This fix provides precise control over layout dimensions and aligns with user expectations across workflows.
Refactored notification utilities for a more modern, reusable notification system and added info/success/error notifications to the background removal (matting) workflow. Removed the custom error dialog in favor of notifications, and exposed all notification types for debugging. Updated imports and cleaned up notification-related code.
Adds logic to adjust the canvas viewport to keep content centered when entering or exiting fullscreen mode. The adjustment calculates the difference in container sizes and updates the viewport position accordingly, improving user experience during fullscreen transitions.
Implemented an ESC key handler to allow users to close the fullscreen editor mode using the Escape key. Updated the editor button tooltip and the shortcuts documentation to reflect this new shortcut.
Replaces all uses of alert() with showErrorNotification or showSuccessNotification for improved user experience and consistency. Refactors canvas creation to use the createCanvas utility function across multiple files, reducing code duplication and improving maintainability. Also updates layer ID generation to use generateUUID.
Introduces a 'visible' property to layers and updates all relevant logic to support toggling layer visibility. Adds visibility toggle icons to the layers panel using a new IconLoader utility, with SVG and fallback canvas icons. Updates rendering, selection, and batch preview logic to respect layer visibility. Also improves blend mode menu UI and ensures new/pasted layers are always added on top with correct z-index.
When the preview is disabled in the canvas widget, a 1x1 transparent PNG placeholder image is now set to node.imgs. This prevents issues with missing images and ensures consistent UI behavior.
Moved all SAM Detector integration logic from CanvasView to a new SAMDetectorIntegration module for better maintainability and separation of concerns. Updated CanvasView to use the new integration functions and removed duplicate code.
Added debouncing to the updateOutput function to prevent excessive updates during rapid changes. Large images are now handled using blob URLs for better performance, while small images use data URIs. Also added logic to skip output updates when preview is disabled and improved cleanup of temporary file trackers when nodes are removed.
Introduces the addMask method to MaskTool in both JS and TS implementations, allowing new masks to be overlaid without clearing existing ones. Updates CanvasView to use addMask instead of setMask when applying SAM detector results.
Replaces polling-based monitoring of SAM Detector results with a MutationObserver that detects modal closure and style changes. Adds fallback to polling if modal is not found, and provides user notification if no mask is applied. This improves reliability and user feedback when applying masks from the SAM Detector.
Eliminates the sendCanvasToClipspace method and related UI/menu options from CanvasView and its TypeScript counterpart, as well as the associated type definition. Also removes the unused maskContext getter from MaskTool. This refactor likely reflects a change in feature requirements or a move away from Impact Pack compatibility.
Introduces automatic registration of canvas images in ComfyUI's clipspace for Impact Pack compatibility, including a new 'Send to Clipspace' menu option and a method on ComfyNode. Adds monitoring and mask application for SAM Detector results, ensuring seamless mask transfer to LayerForge. Also exposes MaskTool's maskContext property for external access.
Moved layer drawing logic into CanvasLayers._drawLayer and _drawLayers methods, replacing repeated rendering code in CanvasIO and CanvasLayers. This improves maintainability and ensures consistent handling of layer properties such as blend mode, opacity, rotation, and flipping. Also, fixed layer serialization to only generate imageId when missing, and ensured new layers have flipH/flipV set when created from matted images.
Replaces setPreviewVisibility(false) with direct assignment to previewVisible in Canvas and Canvas.ts. Adds initialization of preview state based on widget value in CanvasView and CanvasView.ts to ensure correct preview visibility on widget creation.
Adds specific handling for JSONDecodeError during model loading in Python, returning a clear error message if the model config is corrupted. Updates the JS/TS frontends to show a custom error dialog with details and copy-to-clipboard functionality instead of a simple alert, and ensures spinner removal is safe. This improves user experience and troubleshooting for matting model errors.
