Re: header syntax.

On 21 November 2012 14:29, Shane McCarron <ahby@aptest.com> wrote:
> I don't think there is anything wrong with supporting all of those, although
> there would need to be a precedence order for error recovery.

Which (for the baseline) would make us diverge from more and more implementers?

>
> Why do I think that?  Because one purpose for MD is that the document be not
> just human readable in source form, but also that the human readable version
> be pretty.  # header # is way less pretty than header followed by a line of
> equals signs.

I hope you'll agree that's a subjective judgement.


>
> Also, I would like to propose that in the # header # form, text after the
> trailing # be used as the identifier for the header:
>
> # A Header # myHeaderID
>
> should generate
>
> <h1 id="myHeaderID">A Header</h1> in (X)HTML

Please save that for post baseline input. I'll note it.
Neat idea. CurrentMD id values are ... odd IMHO



>
> As to processing precedence order, I think it should be:
>
> 1. # headerText with optional trailing # and optional ID text
> 2. headerText followed by /^=+/ or /^-+/
>
> If a processor encounters both, the header level determined from (1) above
> is used, and the dividing line is ignored.


Have you seen other implementations which have a defined  precedence?

I am -1 on any form of precedence. All I can see is confusion (users)
and being ignored (implementation).

regards


-- 
Dave Pawson
XSLT XSL-FO FAQ.
Docbook FAQ.
http://www.dpawson.co.uk

Received on Wednesday, 21 November 2012 15:24:37 UTC