[XML-SIG] Re: Confused and grasping
Thomas B. Passin
Sat, 25 Sep 1999 13:33:28 -0400
Jeff Blaine wrote:
> I still have this burning desire to do electronic text Right. That's
> the reason *I* want to use XML. I want to be able to write a readable
> semi-marked up README.txt conforming to some DTD and generate HTML,
> Postscript, etc. I want to redo my personal web pages with a common and
> consistent look and feel.
Jeff, what you want to do is get XT from James Clark's web site:
get the XSLT standard from the W3C web site:
and start fooling with XSLT. XSLT is one way to generate all those document types from a master document. It has nothing to do with extracting data and creating the original document, but given an XML document, you will be in business. Soon there will probably be XSL Postscript formatters, and you will need XSLT to use them.
There is very little written about using XSLT, and most of what there is, is out of date and virually worthless. The is because there have been MAJOR changes to the standards since December 1998, but most published work refers to the April or August 1998 versions. Nothing earlier than the Spring 1999 version will do.
You should plan on doing a lot of experimenting (XSLT is hard), but it may be worth it. I generate a number of different web pages for my own site from XML, awk, and XT. You don't need DTDs to learn XSLT.
Another thing you might want is to convert an HTML page to an XML one. There is a very nice tool called Tidy, linked to on the W3C site, that can do this job. And Tidy has been incorporated into a Windows GUI-style editor called WinTidy, put out by Chami (sorry, I don't have the URL here) (free). Excellent tools! Tidy/WInTidy also can fix up many common errors in HTML.
Now, what about Python in all this? Once you are up to speed with the tools above, you may well want to start creating your master xml documents with the help of Python programs, and even to use Python to do some format conversions. It would be easier than writing for Java, C, or (perhaps) perl.