- Add TagFTSIndex service for fast SQLite FTS5-based tag search (221k+ tags)
- Implement command-mode autocomplete: /char, /artist, /general, /meta, etc.
- Support category filtering via category IDs or names
- Return enriched results with post counts and category badges
- Add UI styling for category badges and command list dropdown
Add license resolution utilities and integrate license information into model metadata processing. The changes include:
- Add `resolve_license_payload` function to extract license data from Civitai model responses
- Integrate license information into model metadata in CivitaiClient and MetadataSyncService
- Add license flags support in model scanning and caching
- Implement CommercialUseLevel enum for standardized license classification
- Update model scanner to handle unknown fields when extracting metadata values
This ensures proper license attribution and compliance when working with Civitai models.
- Added ModelMetadataProvider and CivitaiModelMetadataProvider for handling model metadata.
- Introduced SQLiteModelMetadataProvider for SQLite database integration.
- Created metadata_service.py to initialize and configure metadata providers.
- Updated CivitaiClient to register as a metadata provider.
- Refactored download_manager to use the new download_file method.
- Added SQL schema for models, model_versions, and model_files.
- Updated requirements.txt to include aiosqlite.
- Added AutoComplete class to handle input suggestions based on user input.
- Integrated TextAreaCaretHelper for accurate positioning of the dropdown.
- Enhanced dropdown styling with a new color scheme and custom scrollbar.
- Implemented dynamic loading of preview tooltips for selected items.
- Added keyboard navigation support for dropdown items.
- Included functionality to insert selected items into the input field with usage tips.
- Created a separate TextAreaCaretHelper module for managing caret position calculations.
- Added `populate_lora_from_civitai` and `populate_checkpoint_from_civitai` methods to enhance the extraction of model information from Civitai API responses.
- These methods populate LoRA and checkpoint entries with relevant data such as model name, version, thumbnail URL, base model, download URL, and file details.
- Improved error handling and logging for scenarios where models are not found or data retrieval fails.
- Refactored existing code to utilize the new methods, streamlining the process of fetching and updating LoRA and checkpoint metadata.
- Introduced MetaFormatParser class to parse metadata from images with Lora_N Model hash format.
- Implemented methods to validate metadata structure, extract prompts, negative prompts, and LoRA information.
- Enhanced error handling and logging for metadata parsing failures.
- Updated RecipeParserFactory to include MetaFormatParser for relevant user comments.
- Introduced ComfyMetadataParser class to parse metadata from Civitai ComfyUI JSON format.
- Implemented methods to validate metadata structure, extract LoRA and checkpoint information, and retrieve additional model details from Civitai.
- Enhanced error handling and logging for metadata parsing failures.
- Updated RecipeParserFactory to prioritize ComfyMetadataParser for valid JSON inputs.
- Improved reference handling in NodeMapper to support integer node IDs and added error logging for reference processing failures.
- Updated LoraLoaderMapper and LoraStackerMapper to handle lora_stack as a dictionary, ensuring compatibility with new data formats.
- Refactored trace_model_path utility to perform a depth-first search for LoRA nodes, improving the accuracy of model path tracing.
- Cleaned up unused code in parser.py related to LoRA processing, streamlining the workflow parsing logic.
- Introduced a new WorkflowParser class to streamline workflow parsing and manage node mappers.
- Added functionality to load external mappers dynamically from a specified directory.
- Refactored LoraLoaderMapper and LoraStackerMapper to handle new data formats for loras and trigger words.
- Updated recipe routes to utilize the new WorkflowParser for parsing workflows.
- Made adjustments to the flux_prompt.json to reflect changes in active states and class types.
- Deleted the `__init__.py`, `cli.py`, `extension_manager.py`, `integration_example.py`, `README.md`, `simple_test.py`, `test_parser.py`, `verify_workflow.py`, and `workflow_parser.py` files as they are no longer needed.
- Updated `.gitignore` to exclude new output files and test scripts.
- Cleaned up the node processors directory by removing all processor implementations and their registration logic.
- Expanded the prompt.json file with new configurations for KSampler, CheckpointLoaderSimple, and various CLIPTextEncode nodes.
- Introduced additional Lora management features, including a new Lora Stacker and improved trigger word handling.
- Enhanced the loras_widget.js to log the generated prompt when saving recipes directly, aiding in debugging and user feedback.
- Improved overall structure and organization of the prompt configurations for better maintainability.
- Updated the logic in ExifUtils to first identify KSampler nodes and store references to positive and negative prompt nodes.
- Added a helper function to follow these references and extract text content from CLIP Text Encode nodes.
- Implemented a fallback mechanism to extract prompts using traditional methods if references are not available.
- Improved code readability with additional comments and structured handling of node data.
