[Feature Request] remove password from pdf

[christf]

Description

once in a while I receive a password-protected pdf file via email.
Paperless of course cannot ingest this. When asked to open, the user receives a query for the password, and then the file opens.
It would be nice if paperless could decrypt this kind of encrpyted PDF files and ingest them. Ghostscript is able to do the deed on the command line. My current workaround is to download the file, decrypt it and post it to paperless via API. However that causes duplicates and the uncertainty associated with it whether all encrypted documents have already been processed.

Having the ability to "decrypt this pdf using a password that was queried from the user" would resolve that uncertainty and integrate the workflow.

Other

related: #1141

[completelyClueless]

I had the same issue - found this
https://piep.tech/posts/automatic-password-removal-in-paperless-ngx/

would be happy to have such function integrated into paperless-ngx directly...

[bernarddt]

This is a very involved process, but luckily it's do-able. I would also love the integration of this simple feature to unlock pdfs. But luckily this solution works. I've also updated the base script with the changes mentioned in the comments. It helps if you understand a bit of Python coding.

Also I'm running Docker on a Windows system and had to convert the line ending chars to LF for the script to work, otherwise I get an error in the web server logs of:

/usr/bin/env: ‘python\r’: No such file or directory

[swdit]

+1 for a integrated solution.
In my world NGX would just ask "remember password" like a WebBrwoser does.

I ve seen the piep tech attempt too but it would require to know the password in advance, that a thrid party is using to mail me enrycpted PDF's.

[DrSpaldo]

+1 for me as well. I get an email with a PDF protected password, the password is the same every time from that sender. So it would be nice that it just removes the password or ingests the document by using that same password.

[DrSpaldo]

PS. I ended up adding the script above to my Paperless NGX instance. It uses Proxmox with the VE Helper scripts. I just had to make a few minor tweaks from the comments. Ie. python3

[Alienraider]

I feel like I've tried everything but I cannot get the piep.tech script to work. Most recently:

1. Set up passwords.txt (and saved it to paperless's /data folder)
2. Set up removepassword.py (and saved it to the same folder) - changed line 8 to refer to /mnt/user/appdata/paperless-ngx/data/removepassword.py
3. Set a variable in Paperless' Docker PAPERLESS_PRE_CONSUME_SCRIPT: with value /mnt/user/appdata/paperless-ngx/data/removepassword.py

Any suggestions as to what I'm doing wrong?

[nroach44]

In the PAPERLESS_PRE_CONSUME_SCRIPT, you might need to use the path that paperless would see (i.e /data/removepassword.py ?)

[Alienraider]

Hmmn - thanks - I neglected to mention that I'd tried that... It's a real pickle!

It does seem like this would be a great feature to have built in, hopefully the team can do something in the space. It's an incredible piece of software as it is, this would be the icing on the cake for me though.

[CapoD]

I suggest that paperless manages a list of passwords and tries to decrypt consumed documents using this list until it succeeds.

[DrSpaldo]

That is what the script linked above does. It is just some mucking around, which is why it would be great if it was built into the program

[swdit]

well thats nice as long as you now the passwords prior to consuming into ngx.
For me this is usually not the situation.
I would love the see the PDF password handling as part of a version function.
=> Make unenrypted version => enter pdf password => original has bin decrypted and unencrypted version has been saved.
This way we would keep the original for tax authority and revision purposes and in parallel have an easy to handle version inside our ngx and can decide which one to forward to where.

[christf]

It seems to be a reasonable assumption, that the genuinely intended receiver of an encrypted document will know the password in advance.

If reprocessing failed imports is possible, we could achieve Stefans use case too.

[nbranquinho]

+1.
Nowadays, we more frequently receive PDFs that are protected by a password.
If Paperless-ngx can handle this natively that would be ideal. (Or even remove the password from the file if the user so wishes)

[popy2k14]

+1 for this. Also receiving some protected pdf's and would be nice if this is integrated directly as a feature.