Feature Request: OCR support for digitally signed dcouments.

[ShakataGaNai]

Running v3.1 out of docker containers for testing (per https://docs.papermerge.io/3.1/setup/docker-compose/ ). When you upload and attempt to OCR a digitally signed document, the process fails silently. Looking at the logs (from the worker) finds a logical error message:

[2024-03-02 00:20:56,933: ERROR/ForkPoolWorker-8] Task papermerge.core.tasks.ocr_document_task[77be2d59-9703-42df-a3cc-bf920a61eab4] raised unexpected: DigitalSignatureError()
Traceback (most recent call last):
  File "/core_app/.venv/lib/python3.10/site-packages/celery/app/trace.py", line 477, in trace_task
    R = retval = fun(*args, **kwargs)
  File "/core_app/.venv/lib/python3.10/site-packages/celery/app/trace.py", line 760, in __protected_call__
    return self.run(*args, **kwargs)
  File "/core_app/papermerge/core/tasks.py", line 79, in ocr_document_task
    ocr_document(
  File "/core_app/papermerge/core/ocr/document.py", line 86, in ocr_document
    _ocr_document(
  File "/core_app/papermerge/core/ocr/document.py", line 54, in _ocr_document
    ocrmypdf.ocr(
  File "/core_app/.venv/lib/python3.10/site-packages/ocrmypdf/api.py", line 337, in ocr
    return run_pipeline(options=options, plugin_manager=plugin_manager, api=True)
  File "/core_app/.venv/lib/python3.10/site-packages/ocrmypdf/_sync.py", line 388, in run_pipeline
    validate_pdfinfo_options(context)
  File "/core_app/.venv/lib/python3.10/site-packages/ocrmypdf/_pipeline.py", line 204, in validate_pdfinfo_options
    raise DigitalSignatureError()
ocrmypdf.exceptions.DigitalSignatureError: Input PDF has a digital signature. OCR would alter the document,
invalidating the signature.

I can't find any mention of this anywhere, but supporting OCR for digitally signed documents would be nice. Perhaps the version dropdown can indicate something like "Version X w/ OCRed and w/o digital signature". Honestly, I don't even care about accessing a version of the document with OCR'd text, so long as the text is there for full text search. Especially when dealing with a multiplicity of signed legal documents.

[ciur]

Thank you for opening this ticket.

Would you mind uploading a digitally signed document that I can experiment with? Of course, I mean document without sensitive information. One page document (digitally signed) with a couple of words would do the job just fine.

This will help me understand your request better and, of course, validate the feature while developing it.

[ShakataGaNai]

Attaching 3. One is a digital document pushed right through docusign. One is the same document printed then scanned, and through docusign. The third is the same print/scan document signed with Adobe Acrobat (which I'm least confident in working, because Adobe...)

Lipsum scan - adobe signed.pdf
Lipsum scan - docusign.pdf
lipsum - docusign.pdf

[bluekitedreamer]

Possibly a simple issue to fix, see another issue recently filed here with solution suggestion (#614 (comment)).