Mike Orr wrote: > > On Fri, Apr 19, 2002 at 12:14:15PM -0400, Jeff Johnson wrote: > > > On Fri, Apr 19, 2002 at 12:29:23PM -0000, [EMAIL PROTECTED] wrote: > > > > I just wanted nicer ways of accessing XML information through the > > > > DOM API but without all the usual messing about handling > > XML > > > data > > > > which isn't interesting - spurious text sections and so on. One > > problem > > > with > > > > the DOM API is that to use it effectively, there's a lot of > > explaining > > > to do. > > > > > > Have you considered pyRXP? I haven't used it yet but it looks very > > > intriguing. It converts the document to a native Python > > > tuple/list/dictionary structure rather than DOM, making it far easier > > to > > > go through and grab whatever data you want. > > > > > > http://www.reportlab.com/xml/pyrxp.html > > > > I haven't checked out pyrxp yet but have you all tried XPath? It's > > really easy to access XML info with it. > > No, what's xpath? ... Hmm, Parnassus lists three xpath packages (xml_indexer, > a 4Suite tool, and PyXPath). Which do you recommend? > http://www.vex.net/parnassus/apyllo.py?so=d&find=xpath
Think of XPath as perhaps something like the find command for XML. Except they didn't include case-sensitive handling and regular expressions as part of the spec, though there are some math manipulations & other wildcard stuff to work with. And then you have to figure out an XML storage system to serve your XPath, and now you're getting into heavy stuff. (though the eXist.sf.net project makes it pretty easy to start up an XML store and it extends XPath to handle regex & case). No, I like the reportlab/pyrxp concept that "hey, maybe this is really all you need for XML, screw the rest". Give it a try and see if it does what you need. -- Bill Eldridge Radio Free Asia [EMAIL PROTECTED] _______________________________________________ Webware-discuss mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/webware-discuss
