I think this library does too much

[guettli]

Here is some feedback. Ignore it, if you don't care.

I think your library does too much.

In my case I am only interested in one part: make scanned PDFs searchable (by using tesseract ocr)

I think all other things are useful. Each has its own use case, which is very unrelated to the next.

Someone might be interested in the evernote upload, the next user wants to sort PDFs according to keywords into sub directories ....

[guettli]

May be the net result would be better, if you join forces with other developers.

Here is a project which focuses on "create searchable PDF":

https://github.com/jbarlow83/OCRmyPDF