Re: [Cellar] [Tools-discuss] expressing both svg graphics and ascii-art fallback within an rfc2xml version 3 file?

[Dave Rice <[e-mail]>]

Hi Henrik, Julian,

> On Sep 1, 2019, at 12:46 PM, Henrik Levkowetz <[e-mail]> wrote:
> 
> Hi David,
> 
> Response at the bottom:
> 
> On 2019-09-01 18:29, Michael Richardson wrote:
>> 
>> {reposting to tools-discuss}
>> 
>> Hello cellar,
>> 
>> The draft of FFV1 now uses xml2rfc version 3 as defined at
>> https://tools.ietf.org/html/rfc7991
>> <https://tools.ietf.org/html/rfc7991>. Presently in the markdown source of
>> the FFV1 draft, the math is maintained in both LaTeX expressions (which are
>> converted to SVG for inclusion in the RFC XML) as well as an Ascii Art
>> representation (heavily reliant on functions defined within the draft). For
>> example: 
>> 
>> vs
>> 
>> r_{i} = floor( ( R_{i} * S_{i,C_{i}} ) / 2^8 )
>> 
>> Whether in SVG or ascii art, this data is contained by the <artwork> node in
>> rfc2xml version 3. In https://tools.ietf.org/html/rfc7991#section-2.5
>> <https://tools.ietf.org/html/rfc7991#section-2.5> it states, 
>> 
>>   Alternatively, the "src" attribute allows referencing an external
>>   graphics file, such as a vector drawing in SVG or a bitmap graphic
>>   file, using a URI.  In this case, the textual content acts as a
>>   fallback for output representations that do not support graphics;
>>   thus, it ought to contain either (1) a "line art" variant of the
>>   graphics or (2) prose that describes the included image in sufficient
>>   detail.
>> 
>> A bit later it says that SVG data may be encoded and stored within the
>> <artwork>’s @src attribute such as  
>> 
>>   o  As a data: URI, such as: <artwork type="svg" src="data:image/
>>      svg+xml,%3Csvg%20xmlns%3D%22http%3A%2F%2Fwww.w3…">
>> 
>> So from this I was hoping we could work on one draft rfc2xml version 3 file
>> that could use the SVG image data or use the ascii art as a fallback; however
>> when I try to craft the <artwork> node with both such as: 
>> 
>>   <artwork type="svg" src="data:image/svg+xml,ENCODED_SVG_MATH_IMAGE_HERE”>FALLBACK_ASCII_ART_HERE</artwork>
>> 
>> then the output “v3-plaintext” output of the conversion tool at
>> https://xml2rfc.tools.ietf.org/experimental.html
>> <https://xml2rfc.tools.ietf.org/experimental.html>, does not use the textual
>> content of the <artwork> node as a fallback but instead places a warning such
>> as 
>> 
>>   (Artwork only available as svg: No external link available, see
>>   draft-ietf-cellar-ffv1-08.html for artwork.)
>> 
>> Is it feasible (recommended?) to have one rfc2xml XML document that could be
>> a source for both plaintext formats (using an ascii art fallback) and
>> non-plaintext formats such as HTML and PDF which would use the embedded SVG
>> data?
> 
> Yes.  The specification of how to handle the text fallback in the RFC 799x
> documents was in part contradictory, and also not as extensible and easily
> editable as one could wish.  It currently looks like rfc7991bis largely will
> follow the proposal discussed on the xml2rfc-dev list earlier this year, and
> captured in the xml2rfc implementation notes draft here:
> 
> https://tools.ietf.org/html/draft-levkowetz-xml2rfc-v3-implementation-notes-09#section-3.1.1 <https://tools.ietf.org/html/draft-levkowetz-xml2rfc-v3-implementation-notes-09#section-3.1.1>
> 
> which introduces an <artset> element which can contain multiple <artwork>
> elements with different type attributes, for instance (artwork and some
> details omitted:
> 
>  <artset>
>    <artwork type="svg" src="..."/>
>    <artwork type="ascii-art" ...>
> 	# ascii art
>    </artwork>
>  </artset>
> 
> For more on this, and for a light introduction to the v3 schema, check:
> http://tools.ietf.org/src/xml2rfc/trunk/cli/doc/xml2rfc3.html#artwork-alternatives <http://tools.ietf.org/src/xml2rfc/trunk/cli/doc/xml2rfc3.html#artwork-alternatives>

Thanks, this was exactly the clarification that I needed! I see that with the <artset> use described here, that the v3-plaintext and v3-html outputs of https://xml2rfc.tools.ietf.org/experimental.html <https://xml2rfc.tools.ietf.org/experimental.html> use the appropriate <artwork> for the output. If you’re interested, my work to ‘build’ this XML from out markdown source is in https://github.com/FFmpeg/FFV1/pull/160 <https://github.com/FFmpeg/FFV1/pull/160>, and an example of <artset> in the output of this PR can be seen at https://gist.github.com/dericed/3063446119a6e5971ecad29027977073#file-draft-ietf-cellar-ffv1-08-xml-L489-L502 <https://gist.github.com/dericed/3063446119a6e5971ecad29027977073#file-draft-ietf-cellar-ffv1-08-xml-L489-L502>.

Thanks so much,
Dave Rice