Initialize text_recognizer in prepare

Remove unused

Optimize emias parser

Prevent fail when whole page is empty

Update version

Improve text utils

Rework ufa parser

Improve qms segmenter

Improve rais layout

Improve text utils

Support page indexes

Remove play.py

Add todo

Add package building

Update api of pdf_create_slice_doc

Rework rais analyzer

Improve text extraction

Improve qms layout

Use real pages count

Update perm_own segmentation

Rework qms segmentation

Add base ufa_rename blocks parser

Add cache roles

Add caching

progress

Remove debug visualization

fix progress

Apply formatting

Improve text recognition

Rework blocks roles

Add rais segmentation

Test unstructured lib

Add qms lab results parser

Use absolute modules import

Reorganize package

Refactor

Add basic emias segmentation

Store progress

Support multi part qms analyzer

Add multi page blocks

Improve text recognition, remove unused

Improve arch, optimize memory usage