[rfc-i] What the text version is used for (was Re: The <tt> train wreck)

Carsten Bormann cabo at tzi.org
Tue Aug 24 12:52:37 PDT 2021

Hi Paul,

Indeed, dealing with snippets (some of which are *deliberately* incorrect!) is not what I meant here.  Having a 40-page YANG spec in a document that didn’t have the benefit from some tool usage is not an efficient use of everyone’s time.

On 2021-08-24, at 21:28, Paul Kyzivat <pkyzivat at alum.mit.edu> wrote:
> I'm not going to be using a YANG-specific mechanism to do so.

Yes, one of the goals would be to have anchors in the source code the syntax of which is independent of the source code language (they would need to be in a comment of the specific source code language but stand for itself; see PoC below).  For YANG, these could be derived from schema identifiers or SIDs.  For ABNF, the production names are already good referencing anchors (that are probably quite accessible for non-aficionados, too), so it would not need to share that generic fragment identifier anchoring syntax.

> Cutting a a seemingly unique block of the document and pasting it into my review is easier and works for anything.

Reliable copy/paste is indeed one of the most important remaining functions that makes plaintext IETF documents indispensable.

Grüße, Carsten

PoC, assuming the syntax of anchors is “##” 1*DIGIT and YANG’s // comment syntax is being used:

   container system-state {                                //##1720

     container clock {                                     //##1721
       leaf current-datetime {                            
         type date-and-time;

       leaf boot-datetime {                                //##1722
         type date-and-time;

More information about the rfc-interest mailing list