Wow. That is quite a tool chain. I hope we can end up with something simpler
for users.
Regards/Ngā mihi
Brian
On 29-Jan-26 04:28, Michael Richardson wrote:
<#secure method=pgpmime mode=sign>
From: Dave Rice <[email protected]>
Subject: Re: [Cellar] math in RFCs
Date: Wed, 28 Jan 2026 10:20:34 -0500
Cc: [email protected]
To: Michael Richardson <[email protected]>
Hi Michael,
Our FFV1 specification uses a lot of math,
https://datatracker.ietf.org/doc/draft-ietf-cellar-ffv1/20/.
The representation was managed for two types of displays for plain text and
formats that permit more complex mathematical expression (see sections 3.8 in
the HTML and txt, for instance).
- in HTML
https://www.ietf.org/archive/id/draft-ietf-cellar-ffv1-20.html#name-coding-of-the-sample-differ
- in plain text, see section 3.8 of
https://www.ietf.org/archive/id/draft-ietf-cellar-ffv1-20.txt
To do this will integrated the math into our markdown file with a line prefix
“SVGI:” and “SVGC", see the same section 3.8,
https://github.com/FFmpeg/FFV1/blob/master/ffv1.md#range-coding-mode.
We then have a script called ‘makesvg’ at
https://github.com/FFmpeg/FFV1/blob/master/makesvg which parses out our math
expressions and uses pandoc, pdfcrop, pdf2svg, and xmlstarlet to generate an
SVG XML that is acceptable to xml2rfc. Our makefile at
https://github.com/FFmpeg/FFV1/blob/master/Makefile then uses another one of
our scripts called svg2src (https://github.com/FFmpeg/FFV1/blob/master/svg2src)
to insert the SVG XML into the right places to generate the full XML input for
xml2rfc. Then the result is math that looks reasonably nice whether in HTML or
plain text.
Overall the CELLAR working group uses a makefile approach, where we manage
code-like drafts of a specification and then run a ‘make’ process to generate
our RFC submissions. That permits us to take approaches like this rather than
trying to manage changes to SVG directly in a repository.
Kind Regards,
Dave
On Jan 28, 2026, at 3:14 AM, Michael Richardson <[email protected]> wrote:
Some of the authors in CELLAR might have opinions or advice about what the
best way to represent equations in RFCs.
See:
https://mailarchive.ietf.org/arch/msg/rswg/htkywpv1IyNVbKUmOXWMnDM23mw
and related messages in that thread.
--
Michael Richardson <[email protected]>, Sandelman Software Works
-= IPv6 IoT consulting =- *I*LIKE*TRAINS*
_______________________________________________
Cellar mailing list -- [email protected]
To unsubscribe send an email to [email protected]
--
rswg mailing list -- [email protected]
To unsubscribe send an email to [email protected]
