Closed rmusser01 closed 4 months ago
Will use pandoc, have the user supply the converted txt file:
https://pandoc.org/MANUAL.html#epubs
pandoc -f epub -t plain -o filename.txt filename.epub
Closing this as I don't want to integrate pandoc right now. Don't feel its a big ask considering everything else necessary to get this working...
As a user, I would like to be able to select / upload an epub document/book, have the text content of the document extracted, and ingested into the database with keyword tagging.
I would further like ability to search across it and use it for some sort of Q/A RAG thing in the future. (Will split off into its own issue once epub ingestion is supported)
Calibre cli: (Calibre CLI -> convert X book format to raw text -> Ingest into DB) https://manual.calibre-ebook.com/generated/en/ebook-convert.html https://manual.calibre-ebook.com/conversion.html#conversion
Existing projects: https://github.com/Medusa-ML/Book-Summarizer/tree/main
https://unix.stackexchange.com/questions/647686/convert-epub-to-txt-and-preserve-original-formatting