Dear Log,
My current project that I poke at as I fall asleep each morning/afternoon/night (and I do a lot of wonderful wonderful sleeping, as my brain chemistry is adjusting to Paxil) is writing a new Pod parser module, starting from scratch, to be perlpodspec-compliant. It's stream-based: you feed in a stream of lines of Pod, and it produces either events (probably in SAX format, which seems straightforward enough) or a stream of XML output. So to process Pod, you use the events, or you can send the XML output to the XML module of your choice.
Writing perlpodspec and a Pod parser is a whole lot of work, for which I doubt I will see either fame or fortune (unless you slip me some twenny dallah bills, or get Fox to option my TV series pitch about an open source software developer's zany life.)
But I suppose that my doing both is easier than my doing one and someone else's doing another, since really much of the work of writing perlpodspec was making sure that the ideas I was formalizing would actually be implementable. That meant basically writing Pod parsers in my head. So now it's mainly just getting it all down on disk.
And my doing one from stratch is definitely easier than slogging thru all the existing Pod parsers' code and fixing them. I'm not really good with other people's code.
That sounds like a lovely idea to me
A few things that I'd like to see in an XMLized POD would be 1) a namespace (so that we can mix it freely with other stuff) and 2) the inclusion of the Perl source in the output, mostly so that perldoc -m could be done with simple XSLT.
Re:SAX Toy
TorgoX on 2001-12-21T06:22:36
Those are two excellent ideas!
