RE: [xsl] XSLT match with regex what's the best current solution?

[Thomas Winkler <t dot winkler at itcampus dot de>]

Am Mit, 2002-01-16 um 23.21 schrieb Marc Portier:
> now this _is_ a use case, have to admit far higher then what we've been
> testing it on for now, if you don't mind discussing features and reporting
> bugs while we're going (and maybe waiting for their implementation/fix) this
> could be something for you (if you now Java you might want to step in
> eventually)

i guess i wouldn't mind discussing and testing and i'd still be
interested.  the word html problem is only one use case i could think
of, even if there might be some different and more specific solutions to
this one. i think i'll give the POI project, steven mentioned, a try
too.  

> being cocoon addicts ourselves, we have the cocoon-generator version built
> in the package
> (one of the ant-deploy-y-targets in our build is to push the thing onto your
> tomcat under the automount of cocoon so you can start of with some examples.
> Yours would be great to have in there :-))

sounds great for me.

so, if you think you are ready to have someone test regexslt on a
specific use case (as maybe word html) tell me.
 
> > are there more information available?
> Steven wrote a sort of a requirements doc and draft of our little syntax for
> ourselves (there is also a xsd), current development has somewhat
> sidetracked from that, since we're still discovering the problem domain
> while solving it... dunno when/how Steven is planning on syncing again and
> completing the documentation. (it will be up with the rest, hopefully soon)
> 
> Biggest item on our current todolist is moving our cvs server to linux, and
> finally apply the new look of our website, we were planning a zip-release of
> regexslt together with that launch.
>
> > whats the state of the project?
> pre .00 as Steven stated...
> plenty of todo's and moving as we go allong, but we always have a working
> version in cvs... hard to catch the state in alpha or beta terms, since
> knowing were we will end up is as much a journey for us as it will be to
> anyone ever using it, having at all stages in that process a
> clean/working/functional/usefull thing is far more important I guess...
> 
> getting stuff like this in some official XSLT 2 or later is a different
> thread of execution, this mail-thread has kinda balanced on both
> 
> more on status/stability we use jaxp and jakarta oro, those are pretty
> stable, our thingy should be a skinny layer on top, big idea currently is to
> have regexslt induce needed markup, but then use xslt afterwards to pollish
> it up. (easy in cocoon in every case)
> 
> as you might of have read in this list earlier: there are _known_
> limitations of the regex approach an sich, we still think however that keen
> xslt work (the kinda sniff stuff you can catch on this list) should be able
> to come around most of them...
> 
> return question: what's the timeframe to solving your word-docbook issue?

right now i am just testing without a close time frame and thinking of a
java/xsl (cocoon preferred) based solution. so there is no real hurry on
this one. 

maybe you could keep me informed concerning the state of your project...

thank you,
thomas



 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list