GitLab

feat(ocr): filter out low confidence ocr results

- Add confidence score threshold to filter out low confidence OCR results
- Improve OCR accuracy by ignoring less certain detections

refactor(txt_spans_extract_v2): optimize span processing and OCR logic

- Add checks for uppercase character start in the first span of a block

- Optimize character sorting for accurate text assembly
- Handle empty char scenarios to prevent errors
- Remove unnecessary comments and improve code readability
- Enhance OCR text content handling by removing low-confidence spans

- Merge useful_spans and unuseful_spans handling
- Simplify overlap ratio calculation and block type checking
- Remove unnecessary span removal and re-addition

fix(pdf_parse): Move the logic for filling text content into spans before the discarded_block recognition to fix the issue of empty text blocks in discarded_block.

fix(pdf_parse): Move the logic for filling text content into spans before the discarded_block recognition to fix the issue of empty text blocks in discarded_block.

master -> dev

Release 0.10.1

test: batch process demo PDFs- Update test block to iterate through multiple demo PDF files

- Use os.path.join to construct file paths for better cross-platform compatibility
- Remove hardcoded file path

feat(demo): add visualization bbox parameter and refactor parsing process

- Add is_draw_visualization_bbox parameter to enable/disable visualization of bounding boxes
- Refactor the parsing process to improve code readability and maintainability
- Update function documentation to reflect new parameter
- Simplify test code by using a more generic variable name

Fix/demo

master -> dev

Release 0.10.0

fix(pdf_parse): improve OCR result handling

fix(pdf_parse): improve OCR result handling

- Add null check for OCR results to prevent errors on empty lists
- Enhance robustness of OCR text processing in the magic-pdf project

fix(table): add null check for OCR result in rapid table prediction 

refactor(model): move page total time logging to custom model analysis

- Move page total time logging to doc_analyze_by_custom_model.py
- Remove page total time logging from pdf_extract_kit.py
- Add page_start timing variable to custom model analysis
- Update logger output format for page total time

fix(table): add null check for OCR result in rapid table prediction

- Add a null check for OCR result in the predict method
- Return None values if OCR result is None to prevent further processing

feat(README): update for v0.10.0 

- Introduced hybrid OCR text extraction capabilities in v0.10.0
- Significantly improved parsing performance in complex text distribution scenarios- Combined advantages of accurate content extraction and faster speed in text mode with more precise span/line region recognition in OCR mode
- Updated both English and Chinese README files

refactor(para): improve line stop flag and remove unused debug mode

- Add '-' and '–' to LINE_STOP_FLAG in pdf_parse_union_core_v2.py
- Remove unused debug_mode parameter from para_split function in para_split_v3.py

* delete unused pipeline file

* add json test circle

* add size reduction test case

* add serializable test case

* add invalid json compress test case

* add empty test case

* add special char test case