JS Math.random() only has 53 bits of integer precision, making the
64-bit range produce non-uniform values. 32-bit (0xFFFFFFFF) is the
standard ComfyUI seed range and works correctly in both Python and JS.
Also removes the redundant mode input from SeedHistoryNode, relying on
ComfyUI's built-in control_after_generate widget instead.
Refs #43
ComfyUI is converting nodes_latent.py to V3 Schema on October 8th, which will
break direct imports of LatentBatch. This commit replaces the import with a
local implementation copied directly from ComfyUI source code.
Changes:
- Removed: from comfy_extras.nodes_latent import LatentBatch
- Added: Local batch_latents() and reshape_latent_to() functions
- Updated: latentbatch.batch() calls to use batch_latents()
- Added: torch and comfy.utils imports for tensor operations
- Added: Comprehensive unit tests for latent batching functionality
The local implementation is functionally identical to the original and ensures
the node will continue working after the V3 schema migration.
Test Coverage:
- 5 new tests in TestLatentBatchingFunctions class
- All 16 tests passing (11 existing + 5 new)
- Tests cover tensor operations, batch indexing, and reshape logic
- Add auto-batching functionality to split large LoRA collections into manageable chunks
- Implement batch_size parameter to control number of LoRAs per batch (default: 25)
- Add batch_index parameter to select which batch to process
- Include batch tracking metadata in LORA_PARAMS for visualization
- Display batch info in plot parameters when available
- Update lora_list output to show batch header when auto-batching is enabled
- Add comprehensive tests for batching functionality
- Update documentation with detailed auto-batching usage instructions
This feature prevents UI disconnection issues when processing large numbers of LoRAs
by allowing users to process them in smaller batches sequentially.
- Remove debug code and console.log statements
- Fix test suite to properly mock folder_paths module
- Update test expectations to match actual implementation
- Add comprehensive README documentation with feature list
- Add placeholder images for documentation screenshots
- Include diagnostic scripts for testing embedding paths
- All tests passing (338 passed, 2 skipped)
Features implemented:
- Autocomplete for embeddings, LoRAs, and custom tags
- Custom word list loading from URL (with security validation)
- Configurable triggers and settings
- Auto-insert comma, replace underscores, Tab/Enter selection
- Smart scrolling in suggestion list
- Secure content validation to prevent XSS attacks
Credits to pythongosssss/ComfyUI-Custom-Scripts for inspiration
- Add debug/test panel to the node with helpful usage hints
- Display counts of available embeddings and LoRAs
- Show sample items and status information
- Include clear instructions for triggering autocomplete
- Update display name with 🫶 branding
- Make node an OUTPUT_NODE to display information
- Implement centralized settings registry for all KikoTools
- Create KikoEmbeddingAutocomplete node with backend API
- Add frontend JavaScript autocomplete widget with ComfyUI integration
- Support for embeddings and LoRAs with smart filtering
- Configurable settings in ComfyUI UI with 🫶 branding
- Include keyboard navigation and real-time suggestions
- Implement film grain effect with customizable parameters (scale, strength, saturation, toe, seed)
- Use pure PyTorch operations for better GPU utilization (no OpenCV dependencies)
- Apply ITU-R BT.709 color space conversion for accurate grain distribution
- Implement screen blend mode for better highlight preservation
- Add channel-specific weighting matching real film characteristics (3x blue, 2x red)
- Preserve alpha channel when present
- Add comprehensive test suite (20 tests covering all functionality)
- Include documentation and example workflow
- Register node under ComfyAssets/image category
Improvements over reference implementation:
- More efficient memory management avoiding numpy/OpenCV conversions
- Better grain mixing algorithm with proper color science
- Improved performance through PyTorch-native operations
BREAKING CHANGE: Node categories now use emoji-based organization
Add 6 new xyz-helper nodes adapted from comfyui-essentials-nodes:
- FluxSamplerParams: FLUX-optimized parameter generator with batch support
- LoRAFolderBatch: Batch process multiple LoRAs from folders
- PlotParameters: Visualize parameter effects with graphs
- SamplerSelectHelper: Intelligent sampler selection with recommendations
- SchedulerSelectHelper: Optimal scheduler selection for samplers
- TextEncodeSamplerParams: Combined text encoding and parameter management
Changes:
- Port and enhance nodes from comfyui-essentials (now in maintenance mode)
- Add comprehensive documentation with attribution to original author (cubiq)
- Create example workflows for xyz-helpers tools
- Update all node categories to use emoji-based organization
- Fix all unit tests to pass with new category system
- Update README with xyz-helpers section and attribution
Attribution: xyz-helpers adapted from github.com/cubiq/ComfyUI_essentials
All tests passing (318 pass, 2 skip)
- Add new image_scale_down_by tool for downscaling images/latents
- Add display_any.js web component for node display
- Include comprehensive unit tests for the new tool
- Add AnyType class that inherits from str and overrides __ne__ to always return False
- This matches ComfyUI's type checking system for wildcard inputs
- Based on implementation from ComfyUI_essentials
- Add comprehensive tests for AnyType behavior
- Fixes type mismatch errors when connecting any node type
- Change from ('*', {}) to ('*') for proper ComfyUI wildcard type
- Update test to match the corrected syntax
- Fixes 'Return type mismatch' error when connecting nodes
- Universal input acceptance for any data type
- Two display modes: raw value and tensor shape
- Extracts tensor shapes from nested structures
- Comprehensive unit tests with 100% coverage
- Full documentation with usage examples
- OUTPUT_NODE for UI display functionality
- Update pyproject.toml to use black's default line-length of 88
- This matches what the CI workflow expects (black --check without args)
- Reformat all Python files to comply with the new line length
- This will prevent CI failures due to formatting discrepancies
- Add GeminiPromptNode for AI-powered prompt engineering
- Integrates with Google's Gemini API for prompt generation
- Includes various prompt templates and generation modes
- Add comprehensive tests and documentation
- Register node in ComfyAssets category
- Add code quality tools: flake8, mypy, black, pre-commit
- Add .gitattributes for line ending consistency
- Add .secrets.baseline for secret scanning
- Update GitHub workflows for better CI/CD
- Update documentation formatting and examples
- Add CLAUDE.md for AI assistant guidance
- Add scripts directory for automation tools
- Update project configuration in pyproject.toml
- Improve type hints and code formatting across all modules
- Update test configurations and fixtures
- Add ImageToMultipleOfNode for image dimension adjustment
- Ensures image dimensions are multiples of specified values
- Supports both padding and cropping modes
- Useful for model-specific dimension requirements
- Add tests and documentation
- Register node in ComfyAssets category
Vito Sansevero