Re: [xml2rfc] [xml2rfc-dev] Unicode box-drawing for a new --table-borders value?

[Carsten Bormann <[e-mail]>]

On 2021-01-30, at 17:06, Daniel Kahn Gillmor <[e-mail]> wrote:
> 
> Signed PGP part
> On Sat 2021-01-30 07:22:43 +0200, Lars Eggert wrote:
>> I think that would be excellent, but AFAIK Unicode can (still) only be
>> used in contact names (see the recent "Unicode in xml2rfc v3" thread
>> over on rfc-interest@). I really hope we can just get to an agreement
>> to allow Unicode anywhere in the document.
> 
> whoops, i've been using unicode (in particular, the box-drawing
> characters and a few fancy arrows) to pretty good effect (imho) in
> several I-Ds already, for example:
> 
>    https://www.ietf.org/archive/id/draft-dkg-lamps-e2e-mail-guidance-00.html#name-multilayer-cryptographic-en
> 
> That document would be much harder to read and understand (and probably
> much longer) without the compact representations of MIME message
> structure.
> 
> Reading the thread you reference suggests that those characters are
> probably acceptable within an "artwork" or verbatim block, but maybe not
> in the regular text.

That is how xml2rfc is currently programmed.
This is fine for box-drawing, but not so great for α and β, which need to be referenceable in the text.

(A quick fix to get a preview of what will be possible with an updated policy is in http://www.tzi.de/~cabo/contact-hack.sh — you can simply reinstall xml2rfc if that screws up things for you.  This fix is necessary to build Lars’s 8132bis draft.)

> That would mean unicode *can* show up in other places in the .txt
> versions of the RFC, which is the only context where --table-borders
> really matters, afaict.  So it seems like the constraints on unicode in
> the normal text shouldn't be any formal problem with the .txt
> translation.  Does that seem like a plausible analysis?
> 
> Hopefully this change would just be a "simple matter of programming" in
> xml2rfc itself, and not some sort of policy quagmire.

I don’t think there is a change needed to support the artwork in the above-mentioned draft.

Grüße, Carsten