Changelog for 2.x branch


version 2.0

  • Replaced imagemagick with netpbm
  • Invoke giffix to fix broken gifs before conversion
  • Support png images
  • Analyze the file to detect the format without content-type
  • Added several configuration parameters

version 2.1

  • Added scoring for wrong content-type
  • Added scoring for broken gif images
  • Added configuration for helper applications
  • Added autodisable_score feature to disable the OCR engine if the message has already enough points

version 2.1b

  • Rule bugfix to avoid warnings

version 2.1c

  • Applied patch provided by Howard Kash to fix problems with spamassassin + Mailscanner + FuzzyOcr?
  • Removed '-' from jpegtopnm arguments to provide backwards compatibility for older netpbm versions
  • Fixed typo (treshold -> threshold)

version 2.2

  • Small bugfix in content-type check for jpeg (jpg was not matching), thanks to Matthias Keller
  • Added more error handling
  • Removed debug files, added logfile instead
  • More messages with verbose = 2

version 2.3

  • Multiple scans with different pnm preprocessing and gocr arguments possible
  • Support for interlaced gifs
  • Support for animated gifs
  • Temporary file handling reorganized
  • External wordlist support
  • Personalized wordlist support
  • Spaces are now stripped from wordlist words and OCR results before matching
  • Experimental MD5 Database feature

version 2.3b

  • MD5 Database replaced by different feature database
  • Corrupted images are now handled better
  • Added a timeout function to avoid lockups
  • Added threshold overriding on word basis in wordlist
  • Various bugfixes