Re: [Python-markdown-discuss] python-markdown, and wikis.

[Waylan L. <wa...@gm...>]

On Feb 5, 2008 7:59 AM, Blake Winton <bw...@la...> wrote:
>
> >> Oh, and do you handle definition lists?  The examples I saw were all of
> >> the form:
> >> definition term
> >> :  definition description
> >> If you don't handle them, how hard do you think it would be to create a
> >> plugin that would handle them?
> >
> > Unfortunately, that's the one thing I haven't tried to tackle yet. I
> > find the whole 'description gets the markup (:), not the term' a
> > little hard to wrap my mind around. I keep getting the feeling that
> > we'll need to make some changes to the core to get that one to work,
> > although I could be wrong. If you have any suggestions, I'm game.
>
> It seems to me, (after a total of 5 minutes of thinking about it ;) that
>   you could use a combination of pre- and post-processors.  The
> pre-processor would turn a line that preceeds a line that starts with a
> colon (should be easy enough to find, just get lines[index-1]) into
> something more recognizable for the parser, and the post processor could
> turn those things into dt/dd sets.  Actually, I think you could just
> turn them into lines of the form "<dt>...</dt>" and "<dd>...</dd>".
>
Don't forget that you can associate more than one term with a
definition. Second, python-markdown doesn't just substitute text for
html. Rather the text is inserted into a dom object, which is later
output as html when processing is complete. To do that, you need to be
within the core (after preprocessors), although, I suppose you could
set them aside and use a postprocessor for that.

But, then what happens when a definition list is nested in an ordered
or unordered list, or a blockquote? Or what if such blocklevel
elements are nested within a definition? What we need is a easier way
to insert additional blocklevel parsers into the core. This is
something I've been chewing on for a while now and the reason I
haven't tackled definition lists yet. I wanted to work out this
addition extension mechanism (which I've tentatively named
blockparsers) first. If anyone has any suggestions, I'm all ears.



-- 
----
Waylan Limberg
wa...@gm...