dkg

@cabo if there is a way to do that from kramdown-rfc, that would be great. You mention it as a way to parameterize the outer list itself, which would definitely...

> Will the code need to consider intervening unordered lists? Or can we start at 1 again if that happens? I don't have much of an opinion here -- that...

This works great. Thank you, @cabo! Much appreciated.

fwiw, i'm using PyPDF2 version 2.11.2-1 and Weasyprint 57.1 when i'm doing these tests, on debian unstable. When i look at the intermediate pdf output structure, it looks like there...

yes, i see the upcoming switch to `pypdf`, but that' s [not yet in debian](https://bugs.debian.org/1028559), so transitioning to it in the context of debian or its derivatives is going to...

@jrlevine, i'd prefer to ensure that the software i'm running is also in debian, as that ecosystem has a better history of long-term maintenance and reproducibility than any language-specific library...

This is what the relevant page of the pdf looks like when i render it in evince, fwiw: 

fwiw, i've updated to PyPDF2 2.12.1 and i'm still seeing the same whitespace from `PyPDF2.PdfReader(open('elements.pdf', 'rb'), strict=False).pages[9].extract_text())` -- i haven't tried the 3.x series yet

@MartinThoma wrote: > In short: pypdf does not "inject" whitespaces, but has to guess where which amount of whitespaces need to go. Just like any PDF-to-text tool. It's an inherent...

I used weasyprint 57.1-1 to generate [elements.pdf](https://github.com/ietf-tools/xml2rfc/files/10406590/elements.pdf). It renders without whitespace between the words and the commas in the caption, and tools like `pdf2txt` (from `pdfminer`, debian package version 20220319+dfsg-1)...