- Fixed crop() to extract original tags from text instead of reconstructing
- Added MinerU-specific logic in manual.py to handle space/tab separated tags
- Removed redundant import re that caused UnboundLocalError
- Ensures correct bbox coordinates for native images, fallback images, and page selection
- Changed fallback image generation to page-width strips (full horizontal, bbox vertical)
- Implemented smart crop() with native+fallback mixing and deduplication
- Added thresholds: max 10 images, total height <2000px
- Established native_img_map for table/image/equation priority
- Removed 120px padding logic that caused super-long stitched thumbnails
This fixes the issue where chunk thumbnails were either missing or excessively long due to:
1. MinerU not providing images for pure text blocks
2. Official crop() adding 120px padding and stitching across pages
3. Manual.py merging multiple sections into one chunk
The new approach:
- Priority 1: Use MinerU's native high-quality images (tables/equations)
- Priority 2: Use page-width fallback strips (consistent width for stitching)
- Priority 3: Use full page as last resort
- Deduplicates identical bboxes during stitching
- Limits output to reasonable dimensions for UX
- Critical bug fix: imgs list was not initialized before use (line 439)
- Without this fix, NameError would occur when cache miss triggers fallback
- Discovered during reliability audit of MinerU image generation fix
- Add _img_path_cache dict to cache line_tag -> img_path mapping
- Populate cache in _generate_missing_images for fallback text block images
- Refactor crop() to check cache first, return cached image directly
- Fallback to single-position cropping to avoid super-tall merged images
- Fix text_types to use both string literals and enums for compatibility
- Add bbox clamping to prevent cropping errors
### What problem does this PR solve?
Handle MinerU sanitized filenames when reading output. #11613, #11620.
Thanks @shaoqing404 for raising this issue.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Add fallbacks for MinerU output path. #11613, #11620.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Feat: add more chunking method #11311
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Fix: bbox not included in mineru output #11315
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Fix: manual parser with mineru #11320
Fix: missing parameter in mineru #11334
Fix: add outlines parameter for pdf parsers
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Add support for MinerU http-client/server method.
To use MinerU with vLLM server:
1. Set up a vLLM server running MinerU:
```bash
mineru-vllm-server --port 30000
```
2. Configure the following environment variables:
- `MINERU_EXECUTABLE=/ragflow/uv_tools/.venv/bin/mineru` (or the path to
your MinerU executable)
- `MINERU_BACKEND="vlm-http-client"`
- `MINERU_SERVER_URL="http://your-vllm-server-ip:30000"`
3. Follow the standard MinerU setup steps as described above.
With this configuration, RAGFlow will connect to your vLLM server to
perform document parsing, which can significantly improve parsing
performance for complex documents while reducing the resource
requirements on your RAGFlow server.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
- [x] Documentation Update
---------
Co-authored-by: writinwaters <cai.keith@gmail.com>
### What problem does this PR solve?
support local mineru api in docker instance. like no gpu in wsl on
windows, but has mineru api with gpu support.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
MinerU supports VLM-Transfomers backend.
Set `MINERU_BACKEND="pipeline"` to choose the backend. (Options:
pipeline | vlm-transformers, default is pipeline)
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
---------
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
### What problem does this PR solve?
Add MinerU parser. #3945, #8092.
Set `MINERU_EXECUTABLE` to the MinerU executable path, defaults to
`mineru`.
Set `MINERU_DELETE_OUTPUT=0` to preserve MinerU's output, default is 1,
which deletes temporary output.
Set `MINERU_OUTPUT_DIR` to choose the MinerU output directory (uses the
temporary directory if unset).
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
