Implements issue #808 - Allow users to customize the strength
variation range for LoRA widget arrow buttons.
Changes:
- Add 'Strength Adjustment Step' setting (0.01-0.1) in settings.js
- Replace hardcoded 0.05 increments with configurable step value
- Apply to both LoRA strength and CLIP strength controls
Fixes#808
Add name pattern filtering to LoRA Pool node allowing users to filter
LoRAs by filename or model name using either plain text or regex patterns.
Features:
- Include patterns: only show LoRAs matching at least one pattern
- Exclude patterns: exclude LoRAs matching any pattern
- Regex toggle: switch between substring and regex matching
- Case-insensitive matching for both modes
- Invalid regex automatically falls back to substring matching
- Filters apply to both file_name and model_name fields
Backend:
- Update LoraPoolLM._default_config() with namePatterns structure
- Add name pattern filtering to _apply_pool_filters() and _apply_specific_filters()
- Add API parameter parsing for name_pattern_include/exclude/use_regex
- Update LoraPoolConfig type with namePatterns field
Frontend:
- Add NamePatternsSection.vue component with pattern input UI
- Update useLoraPoolState to manage pattern state and API integration
- Update LoraPoolSummaryView to display NamePatternsSection
- Increase LORA_POOL_WIDGET_MIN_HEIGHT to accommodate new UI
Tests:
- Add 7 test cases covering text/regex include, exclude, combined
filtering, model name fallback, and invalid regex handling
Closes#839
- Add `# type: ignore` comments to comfy.sd and folder_paths imports
- Remove unused imports: os, random, and extract_lora_name
- Clean up import statements across checkpoint_loader, lora_randomizer, and unet_loader nodes
- Delay torch import until needed in load_unet and load_unet_gguf methods
- This improves module loading performance by avoiding unnecessary imports
- Maintains functionality while reducing initial import overhead
Move the 'empty/no LoRA' cycling functionality from the LoRA Pool node
to the Lora Cycler widget for cleaner architecture:
Frontend changes:
- Add include_no_lora field to CyclerConfig interface
- Add includeNoLora state and logic to useLoraCyclerState composable
- Add toggle UI in LoraCyclerSettingsView with special styling
- Show 'No LoRA' entry in LoraListModal when enabled
- Update LoraCyclerWidget to integrate new logic
Backend changes:
- lora_cycler.py reads include_no_lora from config
- Calculate effective_total_count (actual count + 1 when enabled)
- Return empty lora_stack when on No LoRA position
- Return actual LoRA count in total_count (not effective count)
Reverted files to pre-PR state:
- lora_loader.py, lora_pool.py, lora_randomizer.py, lora_stacker.py
- lora_routes.py, lora_service.py
- LoraPoolWidget.vue and related files
Related to PR #861
Co-authored-by: dogatech <dogatech@dogatech.home>
- Add validation to check if Civitai API metadata contains recipe fields
- Fall back to EXIF extraction when API returns empty metadata (meta.meta=null)
- Improve error messages to distinguish between missing metadata and unsupported format
- Add _has_recipe_fields() helper method to validate metadata content
This fixes import failures for Civitai images where the API returns
metadata wrapper but no actual generation parameters (e.g., images
edited in Photoshop that lost their original generation metadata)
- Rename nodes to 'Checkpoint Loader (LoraManager)' and 'Unet Loader (LoraManager)'\n- Use os.sep for relative path formatting in model COMBO inputs\n- Update path matching to be robust across OS separators\n- Update docstrings and comments
Refactor _prepare_checkpoint_paths() to return a tuple instead of having
side effects on instance variables. This prevents extra unet paths from
being incorrectly classified as checkpoints when processing extra paths.
- Changed return type from List[str] to Tuple[List[str], List[str], List[str]]
(all_paths, checkpoint_roots, unet_roots)
- Updated _init_checkpoint_paths() and _apply_library_paths() callers
- Fixed extra paths processing to properly isolate main and extra roots
- Updated test_checkpoint_path_overlap.py tests for new API
This ensures models in extra unet paths are correctly identified as
diffusion_model type and don't appear in checkpoints list.
Add file_path as a tie-breaker for all sort modes in ModelCache, BaseModelService, LoraService, and RecipeCache to ensure deterministic ordering when primary keys are identical. Resolves issue #859.
- Replace multiple consecutive spaces with single underscore for tag matching
(e.g., 'looking to the side' → 'looking_to_the_side')
- Support prefix/suffix matching for flexible multi-word autocomplete
(e.g., 'looking to the' → 'looking_to_the_side')
- Add comprehensive test coverage for multi-word scenarios
Test coverage:
- Multi-word exact match (Danbooru convention)
- Partial match with last token replacement
- Command mode with multi-word phrases
- Multiple consecutive spaces handling
- Backend LOG10 popularity weight validation
Fixes: 'looking to the side' input now correctly replaces with
'looking_to_the_side, ' (or 'looking to the side, ' with space replacement)
Make preview file discovery case-insensitive so files with uppercase
extensions like .WEBP are found on case-sensitive filesystems. Also
explicitly list image/webp in the file picker accept attribute for
broader browser compatibility.
https://claude.ai/code/session_01SgT2pkisi27bEQELX5EeXZ
- Add LOG10(post_count) weighting to BM25 score for better relevance ranking
- Prioritize tag_name prefix matches above alias matches using CASE statement
- Remove frontend re-scoring logic to trust backend排序 results
- Fix pagination consistency: page N+1 scores <= page N minimum score
Key improvements:
- '1girl' (6M posts) now ranks #1 instead of #149 for search '1'
- tag_name prefix matches always appear before alias matches
- Popular tags rank higher than obscure ones with same prefix
- Consistent ordering across pagination boundaries
Test coverage:
- Add test_search_tag_name_prefix_match_priority
- Add test_search_ranks_popular_tags_higher
- Add test_search_pagination_ordering_consistency
- Add test_search_rank_score_includes_popularity_weight
- Update test data with 15 tags starting with '1'
Fixes issues with autocomplete dropdown showing inconsistent results
when scrolling through paginated search results.
- Delete examples/metadata/ directory and all example files
- Real metadata.json files in model roots are better examples
- Examples were artificial and could become outdated
- Maintenance burden outweighs benefit
- Remove 'Complete Examples' section from docs/metadata-json-schema.md
- Remove reference to example files in 'See Also' section
Rationale:
Users have access to real-world metadata.json files in their actual
model directories, which contain complete Civitai API responses with
authentic data structures (images arrays with prompts, files with hashes,
creator information, etc.). These are more valuable than simplified
artificial examples.
- Replace static imports of deprecated ComfyButton and ComfyButtonGroup with dynamic imports
- Only loads legacy API files when frontend version < 1.33.9 (backward compatibility path)
- Frontend >= 1.33.9 users no longer see deprecation warnings since legacy code is never loaded
- Preserves full backward compatibility for older ComfyUI frontend versions
- All existing tests pass (159 JS + 65 Vue tests)
- Create docs/metadata-json-schema.md with complete field reference
- All base fields for LoRA, Checkpoint, and Embedding models
- Complete civitai object structure with Used vs Stored field classification
- Model-level fields (allowCommercialUse, allowDerivatives, etc.)
- Creator fields (username, image)
- customImages structure with actual field names and types
- Field behavior categories (Auto-Updated, Set Once, User-Editable)
- Add .specs/metadata.schema.json for programmatic validation
- JSON Schema draft-07 format
- oneOf schemas for each model type
- Definitions for civitaiObject and usageTips
- Add example metadata files for each model type
- lora-civitai.json: LoRA with full Civitai data
- lora-custom.json: User-defined LoRA with trigger words
- lora-no-triggerwords.json: LoRA without trigger words
- checkpoint-civitai.json: Checkpoint from Civitai
- embedding-custom.json: Custom embedding
Key clarifications:
- modified: Import timestamp (Set Once, never changes after import)
- size: File size at import time (Set Once)
- base_model: Optional with actual values (SDXL 1.0, Flux.1 D, etc.)
- model_type: Used in metadata.json (not sub_type which is internal)
- allowCommercialUse: ["Image", "Video", "RentCivit", "Rent"]
- civitai.files/images: Marked as Used by Lora Manager
- User-editable fields clearly documented (model_name, tags, etc.)
- Add BatchImportService with concurrent execution using asyncio.gather
- Implement AdaptiveConcurrencyController with dynamic adjustment
- Add input validation for URLs and local paths
- Support duplicate detection via skip_duplicates parameter
- Add WebSocket progress broadcasting for real-time updates
- Create comprehensive unit tests for batch import functionality
- Update API handlers and route registrations
- Add i18n translation keys for batch import UI
- Update get_lora_info() to check both loras_roots and extra_loras_roots
- Add fallback logic to return trigger words even if path not in recognized roots
- Ensure Trigger Word Toggle node displays trigger words for LoRAs from extra folder paths
Fixes issue where LoRAs added from extra folder paths would not show their trigger words in connected Trigger Word Toggle nodes.
* Fixed a bug where `prompt` and `negativePrompt` were both being
added directly to HTML without escaping them. Given prompts are
allowed to have HTML characters (e.g. `<lora:something:0.75>`), by
forgetting to escape them some tags were missing in the metadata
views for example images using those characters.
- Implement version detection using __COMFYUI_FRONTEND_VERSION__ and /system_stats API
- Add version parsing and comparison utilities
- Dynamically register extension based on frontend version
- Use actionBarButtons API for frontend >= 1.33.9
- Fallback to legacy ComfyButton approach for older versions
- Add comprehensive version detection tests
- Import and use escapeHtml and escapeAttribute in SidebarManager.js
- Escape data-path and title attributes in folder tree and breadcrumbs
- Use CSS.escape() for attribute selectors in updateTreeSelection
- Fixes issue #843 where folders with double quotes broke navigation
Add @wheel event listener to AutocompleteTextWidget textarea to enable canvas zoom when textarea has no scrollbar.
The onWheel handler:
- Forwards pinch-to-zoom (ctrl+wheel) to canvas
- Passes horizontal scroll to canvas
- When textarea has vertical scrollbar: lets textarea scroll
- When textarea has NO scrollbar: forwards to canvas for zoom
Behavior now matches ComfyUI built-in multiline widget.
Fixes#850
Change node-selector z-index from 1000 to var(--z-overlay) (2000)
to ensure the model selector UI appears above the recipe modal
when sending checkpoints to workflow with multiple targets.
Backend _relative_path_matches_tokens() removes extensions from paths
before matching (commit 43f6bfab). This fix ensures frontend also
removes extensions from search terms to avoid matching failures.
Fixes issue where send model to workflow would receive absolute
paths instead of relative paths because the API returned empty
results when searching with file extension.
- Add GET /api/lm/example-workflows endpoint to list available templates
- Add GET /api/lm/example-workflows/{filename} to retrieve specific workflow
- Add 'New Tab Template Workflow' setting in LoRA Manager settings
- Automatically apply 80% zoom level when loading template workflows
- Override workflow's saved view settings to prevent visual zoom flicker
The feature allows users to select a template workflow from example_workflows/
directory to load when creating new workflow tabs, with a hardcoded 0.8 zoom
level for better initial view experience.
Temporarily remove width constraints when measuring content to prevent
scrollWidth from being limited by narrow container. This fixes the issue
where dropdown width was incorrectly calculated as ~120px.
Also update test to match maxItems default value (100).
Add missing offset parameter to MockTagFTSIndex to support
pagination changes from commit a802a89.
- Update search() signature to include offset=0
- Implement pagination logic with offset/limit slicing
Remove .safetensors/.ckpt/.pt/.bin extensions from model names in autocomplete
suggestions to improve UX and search relevance:
Frontend (web/comfyui/autocomplete.js):
- Add _getDisplayText() helper to strip extensions from model paths
- Update _matchItem() to match against filename without extension
- Update render() and createItemElement() to display clean names
Backend (py/services/base_model_service.py):
- Add _remove_model_extension() helper method
- Update _relative_path_matches_tokens() to ignore extensions in matching
- Update _relative_path_sort_key() to sort based on names without extensions
Tests (tests/services/test_relative_path_search.py):
- Add tests to verify 's' and 'safe' queries don't match all .safetensors files
Fixes issue where typing 's' would match all .safetensors files and cluttered
suggestions with redundant extension names.
Move clear button from top-right to bottom-right to avoid
obscuring text content. Add hover visibility for cleaner UI.
Reserve bottom padding in textarea for button placement.
Implement search query variation generation to improve matching for multi-word tags:
- Generate multiple query forms: original, underscore (spaces->_), no-space, last token
- Execute up to 4 parallel queries with result merging and deduplication
- Add smart matching with symbol-insensitive comparison (blue hair matches blue_hair)
- Sort results with exact matches prioritized over partial matches
This allows users to type natural language queries like 'looking to the side' and
find tags like 'Looking_to_the_side' while maintaining backward compatibility
with continuous typing workflows.
- Add syncChanges() function to recipeApi.js for quick refresh without cache rebuild
- Implement dropdown menu UI in recipes page with quick refresh and full rebuild options
- Add initDropdowns() method to RecipeManager for dropdown interaction handling
- Update AGENTS.md with more precise instruction about running sync_translation_keys.py
- Integrate sync changes functionality as default refresh behavior
- Add missing translations for modelTypes, recipe refresh, and sync notifications
- Translate for all supported languages (zh-CN, zh-TW, ja, ko, fr, de, es, ru, he)
- Run sync_translation_keys.py to ensure key consistency
Refactor force_refresh path to use thread pool execution instead of blocking
the event loop shared with ComfyUI. Key changes:
- Fix 1: Route force_refresh through _initialize_recipe_cache_sync() in thread pool
- Fix 2: Add GIL release points (time.sleep(0)) every 100 files in sync loops
- Fix 3: Move RecipeCache.resort() to thread pool via run_in_executor
- Fix 4: Persist cache automatically after force_refresh
- Fix 5: Increase yield frequency in _enrich_cache_metadata (every recipe)
This eliminates the ~5 minute freeze when rebuilding 30K recipe cache.
Fixes performance issue where ComfyUI became unresponsive during recipe
scanning due to shared Python event loop blocking.
