PDF/A compliance

[carstenblt]

ERPNext PDFs don't seem to comply with PDF/A:

[0xD0M1M0]

I believe this not directly ERPNext related. Factur-X creates the final PDF/A-3. Still doesn't change the fact, that errors are raised.

[0xD0M1M0]

Background info: https://www.pdflib.com/de/pdf-know-how/zugferd-und-factur-x/

[carstenblt]

Factur-X cannot fix the PDF, if it's already non-conformant: akretion/factur-x#1

ERPNext still creates PDFs with wkhtmltopdf, right? I fear then there is not a lot you can do about it.

[0xD0M1M0]

Checking the direct ERPNext Output, the "error" seems to be present there already:

• PDF/A requires if DeviceGrey, DeviceRGB or DeviceCMYK is used, that an Output Intent with a Colorprofile has to be present.
• ID-Entry with two Hex-Values is missing in the File End

[0xD0M1M0]

Factur-X cannot fix the PDF, if it's already non-conformant: akretion/factur-x#1

Ok, understood.

ERPNext still creates PDFs with wkhtmltopdf, right? I fear then there is not a lot you can do about it.

wkhtmltopdf is used for some things and pypdf for others:

https://github.com/frappe/frappe/blob/9fd891d01eb6b7d4a03fa0fe1762ae1a8ce8fe50/frappe/utils/print_format.py#L7

Maybe a step in between can fix the mess and convert the output into PDF/A-3 before merging:
https://www.python-forum.de/viewtopic.php?t=44286

[timschneider]

I have implemented a conversion using Ghostscript https://ghostscript.readthedocs.io/en/latest/VectorDevices.html#vectordevices-creating-pdfa in the PR #39. In this case, the PDF file is converted to PDF/A-3 by ERPNext before the XML data is added. This requires Ghostscript to be installed on the system.