cfrg/draft-irtf-cfrg-vdaf

Clean up use of xml2rfc artwork and sourcecode tags

Closed this issue · 2 comments

divergentdave commented

We currently use a mixture of <artwork> and <sourcecode> tags. These are both generated when we fence off preformatted text with three or more tildes. If an identifier appears after the leading group of tildes, on the same line, then a <sourcecode> element is generated, with its type attribute set to that identifier. See https://datatracker.ietf.org/doc/html/rfc7991#section-2.48 for the xml2rfc tag definition, <sourcecode> is now preferred over <artwork> for source code and formal languages. The document has a mix of ASCII art and source code excerpts, though almost all ~~~ code fences lack an identifier, save for one with state and four with transition. We should change excerpts from the proof of concept to ~~~ python, and change ~~~ state and ~~~ transition to ~~~ pseudocode, which are among the list of preferred values. As a bonus, my editor supports embedded syntax highlighting when I switch a block to ~~~ python. The fences around true ASCII art can stay as they are.

divergentdave commented

There is a living list of preferred values for type at https://www.rfc-editor.org/rpc/wiki/doku.php?id=sourcecode-types, and it incldues tls-presentation. We should use this in the ping-pong topology section.

cjpatton commented

There is a living list of preferred values for type at https://www.rfc-editor.org/rpc/wiki/doku.php?id=sourcecode-types, and it incldues tls-presentation. We should use this in the ping-pong topology section.

We also use TLS representation for specifying some of the message formats in {{prio3}} and {{poplar1}}.