v0.12

Changes:

• PdfJS improved to be compliant with 'Image Detection' module.
• Allow Abbyy as pdf extractor.
• Python3 client improved.
• Added Spanish & Portuguese Readme.
• Headings detection improved.
• APIpostDocument with optional defaultConfig.
• Several Bug fixing.

0.11.2

Fixes:

• Fixed PdfMiner wrong split words
• Fixed #372 & #367

v0.11.1

Changes:

• Fixed dependencies with security vulnerability detected
• Several bug fixes

v0.11

Changes

• Advanced Image detection module that allows scan images using OCR's
• Improved data extraction & reconstruction when a document has pages with rotated content
• Parsr bare-metal installation process automated using just one NodeJs script
• Removed GraphicsMagick & pdf2pic dependencies
• Updated documentation
• Several bug fixes

0.10.1

Security vulnerability fixed

Bump bleach from 3.1.0 to 3.1.1 in /demo/jupyter-notebook

0.10

Changes

• New input file *.docx
• New 'Table of contents' processing module
• UI added button for outputs download
• Added compatibility for PdfMiner '20200124'
• Improved PdfMiner extraction time using xml stream reader
• Allow to run new Ocr's using API by extending configuration file
• Several bug fixes

Breaking changes

• Deprecated pipeline configuration property 'extractor.img'

0.9: Merge branch 'develop'

Changes

• Integrated new OCR's in GUI

  • Google Vision
  • Amazon Textract
  • Microsoft Cognitive Services
  • Abbyy

• Updated GUI: Added oficial Logo and fixed some cosmetic issues

• Several bug fixing

• Updated Readme.md

v0.8: Merge pull request #293 from axa-group/feature/Image_Module_Off

Changes

• Simple Image detection using PdfMiner.
• Allowed *.elm as input to be parsed (message body and attachments are used to extract data).
• GUI can display page margins by activating just a switch.
• Readme in French.

v0.7.1: Merge pull request #263 from axa-group/feature/better-error-trace

Changes

1. Removed 'sharp' dependency from API
2. Improved errors handling
3. Allow Tesseract to run multi pages PDF's
4. Some JS vulnerabilities fixed
5. Improved Jupyter Notebook document versioning display

v0.7

Changes

1. Optimisation of images before tesseract scan (detect rotation & removed shadows)
2. New input module option Pdf.js (recommended for large Pdf's)
3. Jupyter Notebook: Added document versioning & comparison
4. Javascript vulnerability Fixed
5. Several GUI & Server bug fixes