Re: Wiki conversions

On 5/4/05, Nikolas 'Atrus' Coukouma <lists@atrus.org> wrote: > Justin Patrin wrote: > > >On 5/4/05, Nikolas 'Atrus' Coukouma <lists@atrus.org> wrote: > > > > > >>Justin Patrin wrote: > >> > >> > >>>On 5/4/05, Nikolas 'Atrus' Coukouma <lists@atrus.org> wrote: > >>> > >>> > >>>...and they are. How do you think I did it? I did mention that this is > >>>done using Text_Wiki. I copied the default Parse objects and altered > >>>them for Tiki and copied the Xhtml render objects and altered them for > >>>DokuWiki, then copied those for CoWiki. If you want to see the > >>>rulesets: > >>>http://pear.reversefold.com/Text/Wiki/ > >>> > >>> > >>I'm not sure what you're asking, then ("Should I bother creating > >>...?") > >>if you've already created it. Are you asking if they should be part of > >>Text_Wiki? A subpackage? in your own channel? > >> > >>The bundling with Text_Wiki seems like it should be up to the maintainer > >>(Paul Jones). I'm not sure if it's worth making a subpackage and putting > >>it in PEAR. I'd definitely suggest packaging it and putting it in your > >>own channel to make it easy to install. > >> > >> > > > >I only made a TIki Parser and CoWiki and DokuWIki renderers. I'm > >asking if anyone would find it useful to have the rest (Tiki renderer, > >CoWiki and DokuWiki parsers). > > > >Also, I just want to know if anyone else would use any of this. > > > >I could put it in a channel of my own...but I don't really have a good > >place to host it right now. We'll see.... > > > > > Ahh, okay. I can see this being handy, but I don't have a use for it > myself (yet?). I think it would be really nice to have this sort of > functionality as a (sub)package (Text_Wiki_Conversion?) that could also > convert the other outputs (e.g. XHTML) into Wiki text. That's the thing, there's no need for a subpackage. Just write an XHtml parser and you can convert to any of the Renderer types. > > To digress a bit, I'd also like to see Text_Wiki (and by extension > Conversion) use a real lexer and parser. > Yeah....it would make some sense.... The way it currently works seems good to me, though... As long as you write your regexes well and make sure you don't include the deliminator in your matching you're fine. But if you'd like to make it use a lexer and parser... ;-) I just don't know how easy those are to set up in real life. I've never actually worked with a real lexer, just played with grammars in College. -- Justin Patrin