Actually, it is safe to say that Papermerge was entirely rewritten – that’s why this release took 18 months to reach beta!
In any case, here is in nutshell what’s new:
- frontend is separate application
- frontend features desktop like look and feel
- frontend features dual panel mode
- frontend features realtime document OCR status updates
- application is fully RESTul
- REST API documentation (so called openapi schema) is generated from code
- backend is kubernetes ready (i.e. can be configured via environment variables) – basically application is now cloud native 😉
- download searchable PDFs
- use OCRmyPDF to extract OCR data
- use elasticsearch as search engine
- pages can be rotated
- two documents can be merged into one
What’s was Temporary Removed
Well, there are some parts which didn’t make it to this version, but will make it in next releases:
- document sharing between users
What was Removed
All code related to importing documents (import from IMAP account or local folder) was removed from backend. Any sort of document importing feature e.g. from IMAP account, S3 account, FTP account, local folder etc etc and will be provided via 3rd parties which will implement import specific code and upload document via Papermerge REST API to be backend server.
Basically importing documents from X, or Y (e.g. x=S3 location, Y=IMAP account) will be implemented by external applications.
In order to quickly start Papermerge 2.1.0b1 locally use docker compose as described in documentation
Docker images are available on: