GitLab

Update pdf_extract_kit.py

Modify line 397 to ensure compatibility with CPU execution, addressing the issue where specifying 'cpu' in config.json still results in a ValueError for expecting a cuda device but getting 'cpu' during demo execution.

Release 0.9.0

fix(pdf_parse): improve span filtering

feat(pdf_parse): improve span filtering and add new block types

- Refactor remove_outside_spans function to filter spans more accurately
- Add image_footnote, index, and list block types to output file documentation
- Update draw_span_bbox to use preproc_blocks instead of para_blocks
- Bump version to 0.9.0

fix(pdf_parse): improve span removal logic for all content types

fix(pdf_parse): improve span removal logic for all content types 

- Update remove_outside_spans function to handle all content types
- Add processing for text and equation spans
- Improve overlap calculation for better accuracy

- Update remove_outside_spans function to handle all content types
- Add processing for text and equation spans
- Improve overlap calculation for better accuracy

- Update remove_outside_spans function to handle all content types
- Add processing for text and equation spans
- Improve overlap calculation for better accuracy

fix(pdf_parse): optimize span processing by removing outside spans

fix(pdf_parse): optimize span processing by removing outside spans

- Add new function `remove_outside_spans` to filter spans based on image and table blocks
- Reorder span processing steps to improve efficiency
- Update imports to include `calculate_overlap_area_in_bbox1_area_ratio`

fix(magic_pdf): handle missing image_path in spans

fix(magic_pdf): handle missing image_path in spans

# Conflicts:
#	magic_pdf/dict2md/ocr_mkcontent.py

- Add check for 'image_path' in spans to avoid errors when it's missing
- Update image handling in both paragraph text and content dictionary
- Improve error handling and make the code more robust

- Update image content extraction to iterate through all spans in a block
- Add support for extracting table content from spans within a block
- Handle multiple content types within table spans (latex, html, image)
- Refactor code to be more modular and easier to maintain

(docs&build): switch to Aliyun PyPI mirror

(docs&build): switch to Aliyun PyPI mirror

- Update PyPI mirror from Tsinghua to Aliyun in multiple Dockerfiles and installation scripts
- This change may improve package download speed and reliability for users in China

Dev->0.9 release

docs(README): update model download instructions for PDF-Extract-Kit 1.0 

- Update README.md and README_zh-CN.md to include new model download instructions
- Provide detailed steps on how to download models after PDF-Extract-Kit 1.0 repository change
- Emphasize the need to re-download models due to repository change

- Update README.md and README_zh-CN.md to include new model download instructions
- Provide detailed steps on how to download models after PDF-Extract-Kit 1.0 repository change
- Emphasize the need to re-download models due to repository change

refactor(table): disable StructEqTable support and add TableMaster support

- Remove import and usage of StructTableModel- Add support for TableMaster model- Update table model initialization logic to support TableMaster
- Log error and exit if StructEqTable is selected, as it's under upgrade
- Update README files to reflect changes in table parsing capabilities

fix: add priority match rule

perf: table model update with PP OCRv4

Dev