v2.0.0

What's Changed

• Fix typo in README.md by @konstantinschulz in #25
• Remove unused batch files by @stweil in #26
• Fix dockerimage creation by @joschrew in #27
• rewrite: Pythonic, ocrd v3, utilise page-level annotation by @bertsky in #28
  • use Python instead of bashlib: faster, more flexible
  • include a distribution of the PRImA PDF converter as package data
  • instead of just the original image files, extract image data from the PAGE annotation, including any AlternativeImage
  • for that, introduce params image_feature_selector and image_feature_filter (e.g. cropped,deskewed,binarized)
  • support processing with METS Server and all new ocrd>=3.0 user-configurable features (page-parallel processing, page timeouts, error handling)
  • extend negative2zero to full PAGE validation and repairs for coordinates
  • back the font parameter by downloadable resources (ocrd resmgr); provide a variety of preconfigured fonts
  • multipage: add setting pagelabels=pagelabels for @ORDER and @ORDERLABEL from physical structMap
  • multipage: add parameter multipage_only to only keep the document-wide PDF, not the page-wise PDF files
  • multipage: add logical structMap divs as outline labels (PDF bookmarks)
  • multipage: improve and add more metadata, use proper formatting (string encoding, dates)
  • multipage: add MODS as extra XMP metadata payload
  • improve logging and relaying error messages
  • add processor ocrd-altotopdf (with limited features) besides ocrd-pagetopdf
  • add regression tests, CI and CD

Full Changelog: v1.1.0...v0.2.0