Introduces a new video file, ShowTimeLayerForge.webm, to the Showtime directory. This asset may be used for media playback or demonstration purposes within the project.
Replaces direct calls to canvas.render() and node.onResize() with setTimeout to defer execution, ensuring UI updates occur after DOM changes. This helps prevent rendering issues and improves responsiveness when toggling mask controls and editor view.
Replaces all references to 'softness' with 'hardness' for brush controls in both CanvasView.js and MaskTool.js. Updates UI labels, slider IDs, and internal logic to use 'hardness' terminology, clarifying that a higher value means a harder brush edge.
Introduces a brush preview overlay using a separate preview canvas in MaskTool. Mouse event handlers in CanvasInteractions and MaskTool are updated to support passing both world and view coordinates, enabling accurate brush preview rendering. The preview is shown or hidden appropriately on mouse enter/leave and while drawing.
Set logging level to DEBUG in both Python and JavaScript loggers for improved debugging. Updated MaskTool to accept an onStateChange callback and trigger it after mask state changes, allowing Canvas to respond to mask updates.
Changed the default log level to LogLevel.NONE in both Python and JavaScript logger utilities. Updated LayerForgeLogger to fully disable logging when the effective level is NONE, ensuring no logs are emitted for disabled modules.
Converted the publish workflow from a job-based GitHub Actions workflow to a reusable composite action. Added inputs for personal access token and skip_checkout, included Python setup and comfy-cli installation steps, and updated the publish command to use the new structure.
Refactors the .painter-tooltip CSS for better appearance, responsiveness, and usability, including new scaling classes, table-based layout, and improved scrollbar styling. Updates shortcut help content to use tables instead of lists for clarity. Enhances tooltip positioning logic to prevent overflow outside the viewport and ensures proper display near screen edges.
Expanded the feature list to include persistent state, advanced masking, AI-powered matting, memory management, and improved workflow integration. Rewrote and reorganized the controls section for clarity, adding new shortcuts and detailed descriptions for canvas, layer, clipboard, and mask mode interactions.
Split help tooltip content into standard and mask mode shortcuts. Tooltip now dynamically displays relevant shortcuts based on whether the mask tool is active, improving user guidance during mask editing.
Eliminates the 'optional' input_image and input_mask parameters from the CanvasNode class and its process_canvas_image method. Fallback logic for these parameters is also removed, simplifying the code and relying solely on WebSocket cache for image and mask data.
Added 'title' attributes to various canvas control buttons to provide tooltips with descriptive information about each button's function. This improves accessibility and user experience by clarifying the purpose of each control.
Replaces direct DOM removal with a dedicated closeBlendModeMenu method for consistency. Increases menu z-index for better stacking, and appends the menu to the canvas container if available instead of always using document.body.
Replaces the onInteractionEnd callback with onStateChange for more consistent state change notifications. Adds a removeSelectedLayers method to Canvas for cleaner layer removal logic. Updates UI event handlers to use the new methods and callbacks, and ensures state is saved after relevant operations. Cleans up redundant updateOutput calls and streamlines output update logic.
Eliminated the 'cache_enabled' and 'output_switch' parameters and related logic from CanvasNode. This simplifies the node's configuration and processing flow by removing unused or redundant options.
Introduces an optional onInteractionEnd callback to the Canvas class, which is triggered at the end of user interactions. CanvasView now uses this callback to update output after interactions, improving responsiveness to user actions.
Moved sys.path modification from canvas_node.py to __init__.py for better package management. Improved logger formatting for colored output and enhanced file logging configuration with error handling for log directory creation. Added python/__init__.py to make the 'python' directory a package.
This commit applies consistent code formatting across multiple files, including spacing, indentation, and object destructuring. No functional changes were made; the update improves code readability and maintainability.
Removed redundant comments and streamlined logic across canvas-related modules, including mask positioning, garbage collection, and WebSocket communication. Improved code readability and maintainability by eliminating unnecessary explanations and clarifying intent in both Python and JavaScript files. No functional changes were made; this is a cleanup and refactor for better developer experience.
Introduces a WebSocket-based mechanism for CanvasNode to send and receive canvas image and mask data in RAM, enabling fast, diskless data transfer between frontend and backend. Adds a new WebSocketManager utility, updates CanvasIO to support RAM output mode, and modifies CanvasView to send canvas data via WebSocket before prompt execution. The backend (canvas_node.py) is updated to handle WebSocket data storage and retrieval, with improved locking and cleanup logic. This change improves workflow speed and reliability by avoiding unnecessary disk I/O and ensuring up-to-date canvas data is available during node execution.
Introduces x/y coordinates to MaskTool for world-space positioning, allowing the mask to extend beyond the output area. Updates mask drawing, export, and rendering logic to account for mask position. Ensures mask position is updated when moving or resizing the canvas, and preserves mask content during canvas resizing. Improves mask extraction and rendering accuracy.
Introduces an operation counter and threshold in ImageReferenceManager to trigger automatic garbage collection after a set number of canvas operations. Canvas now increments the operation count on save, undo, and redo, and exposes methods to set the operation threshold and retrieve stats including operation count. CanvasView displays the operation count and threshold after manual garbage collection.
Introduced ImageReferenceManager to track and clean up unused images from the database and cache. Added manual garbage collection controls to the UI and exposed related stats and cleanup methods in Canvas. Updated db.js with a method to retrieve all image IDs for cleanup purposes.
Refactored method and variable names from 'canvas size' to 'output area size' across multiple files for clarity. Updated UI labels and related function calls to reflect this terminology change.
Eliminated the applyBlendMode method from Canvas and CanvasLayers, along with related unused imports in CanvasIO and CanvasLayers. This streamlines the code by removing dead code and unnecessary dependencies.
Renamed Mask_tool.js to MaskTool.js and Canvas_view.js to CanvasView.js for consistency. Removed unused imports and redundant properties/methods from Canvas.js and CanvasState.js to streamline the codebase.
Removed Polish-language and redundant comments from multiple modules, including Canvas, CanvasInteractions, CanvasLayers, CanvasRenderer, CanvasState, Canvas_view, Mask_tool, and CommonUtils. This improves code readability and maintainability by reducing clutter and focusing on self-explanatory code.
This commit removes unnecessary and redundant comments from multiple JavaScript modules, streamlining the code and improving readability. No functional changes were made; only comment cleanup and minor formatting adjustments.
Moved CommonUtils.js, ImageUtils.js, and LoggerUtils.js to a new utils/ directory and updated all import paths accordingly throughout the codebase. Also increased the default canvas width in Canvas.js from 512 to 1024.
Refactors CanvasState to maintain independent undo/redo stacks for mask editing and layer editing. Updates all relevant logic to use the correct history depending on the active mode, ensuring undo/redo and history buttons work as expected in both modes. MaskTool now saves history on activation, clear, and mouse up, and history info is reported per mode. Improves user experience when switching between mask and layer editing.
Replaces calls to this.canvas.snapToGrid with the global snapToGrid function throughout CanvasInteractions.js. This simplifies the code and ensures consistent snapping behavior.
Extracted and centralized common utility functions such as createCanvas, normalizeToUint8, and generateUniqueFileName in CommonUtils.js to reduce code duplication. Added createImageFromSource to ImageUtils.js. Refactored db.js to use a helper for IndexedDB requests, simplifying error handling and reducing boilerplate. Updated CanvasIO.js, CanvasLayers.js, and Canvas_view.js to use the new utilities.
Replaced manual logger setup in CanvasRenderer, ImageCache, Mask_tool, and db modules with the createModuleLogger utility for cleaner and more consistent logging initialization.
Moved all input/output related methods from Canvas.js to a new CanvasIO.js module and delegated corresponding Canvas class methods to CanvasIO. This improves code organization and separation of concerns, making Canvas.js more focused on core canvas logic and CanvasIO.js responsible for I/O operations.
Moved all rendering-related methods from Canvas.js to a new CanvasRenderer.js module. Canvas now delegates rendering to CanvasRenderer, improving separation of concerns and maintainability.
Replaces per-layer mask processing with a single visibility canvas that composites all layers, then generates the mask based on the resulting alpha values. This simplifies the logic and ensures correct handling of partial transparency across all layers.
