Skip to content
This repository was archived by the owner on Feb 16, 2023. It is now read-only.

Feature Request: Rotation in App #228

Closed
isentropik opened this issue Dec 31, 2020 · 6 comments
Closed

Feature Request: Rotation in App #228

isentropik opened this issue Dec 31, 2020 · 6 comments

Comments

@isentropik
Copy link

Hey! I have a few images that I missed rotating while saving. Would it be possible to add a rotate feature so that the OCR doesn't spit out a garbled mess?

@jonaswinkler
Copy link
Owner

This would involve recreating the documents in paperless from scratch. With the way paperless is implemented right now, I'd essentially have to delete the document, and go through the consumption process again, since this affects everything about the document (original file changed, checksum changed, content changed, thumbnail changed, converted PDF changed, matched metadata changed due to changed content, search index changed, ...)

How many images are we talking about? You can always download the original image and correctly rotate it yourself.

@isentropik
Copy link
Author

isentropik commented Dec 31, 2020 via email

@jonaswinkler
Copy link
Owner

Rather than doing this, I'd eventually like to use the auto rotate feature of the OCR software I'm using in paperless.

https://ocrmypdf.readthedocs.io/en/latest/cookbook.html#correct-page-rotation

I just need to make sure that this is compatible with all existing configuration options.

@jonaswinkler
Copy link
Owner

I'm just moving the discussion of many related issues into one new issue.

@Patricol
Copy link

IMO there should still be a way to override an incorrect rotation judgement in the frontend; even if it isn't in-place. Of course it can be made more or less sensitive; but that isn't magic; there is no sweet spot where everything will work.

The backend procedure does indeed sound onerous, but making people export, delete, rotate, reupload, retag, etc. themselves doesn't sound like the long-term ideal way. I think the issue should remain open in case someone interested enough to implement it comes along.

@ark-
Copy link

ark- commented May 31, 2021

This would be useful as sometimes the PDF conversion gets it wrong and the OCR output is therefore fairly useless. Could fit nicely into the UI from #335

tribut pushed a commit to tribut/paperless-ng that referenced this issue Aug 18, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants