Feature Request: Allow "FPDF.add_font" to accept a bytes/buffer object

[EmperorArthur]

Hello,

This is more of a nice to have than anything, but in playing around, I found this pymupdf-fontsconvenient package. and thought it would be great if it could be loaded into the code without an extra step.

In addition, accepting bytes or buffers would allow for use with setuptools Resource Manager API when using this as part of a larger project.

[Lucas-C]

Interesting!

I'll try to check out soon how fpdf2 could work along this pymupdf-fonts library

[Lucas-C]

Just a quick heads-up: I haven't been able to install pymupdf on my laptop (the required steps are a bit convoluted) so I haven't been able to progress on this subject yet.

Anyone is very welcome to have a look a it!

[EmperorArthur]

Oh, I didn't bother with that. I just installed the fonts package. https://pypi.org/project/pymupdf-fonts/

I believe they're ignoring the whole api I was talking about and just base64 encoding the data.

[Lucas-C]

Alright, but the code to parse & load those font buffers are defined in MuPDF:
https://github.com/kobolabs/mupdf/blob/master/include/mupdf/fitz/font.h#L81

Without PyMuPDF, it will be difficult to extract those fonts...

[Lucas-C]

On a related note, we plan to move to fonttools: #418

I'm not sure this will solve this initial request here, but I'm going to close this issue for now