Hi, Martin v. Löwis wrote: > I don't think Python is or should be the premier language to do XML > processing.
I object! > If you have an application that is entirely about XML > processing, use Java. I highly object! Performance is one reason: http://www.xml.com/pub/a/2007/05/09/xml-parser-benchmarks-part-1.html http://www.xml.com/pub/a/2007/05/16/xml-parser-benchmarks-part-2.html http://effbot.org/zone/celementtree.htm#benchmarks Simplicity is another. Python wins that contest, for XML and most other topics you may choose. > If you have an application that integrates a lot of different things > (or perhaps just two or three of them), and XML processing is one, > you should consider Python. Then you should analyze your processing > needs, and pick a Python library that meets these needs. Absolutely. > If you found that validation is a processing need, I strongly recommend > that you re-evaluate your processing needs (whether you use Python > or not). IMHO, validation is much over-rated and over-used. It's very handy, though. You can validate on the way in (right in the parser) and be sure that the structure you get is as expected, without adding tons of "is this valid input" checks to your code. That one is about simplicity, too. I think validation is somewhat comparable to assertions that you put into your code. > As for the flagship library to do XML processing: I still think that's > the standard library. It has always met my own processing needs, and > it comes as an included battery. Most applications of PyXML should > easily port to the standard library. Again, depends on your needs. But I think we agree that the stdlib should include the batteries for the majority of processing needs, and leave special needs to external packages. And I would agree that it meets this goal for XML. Stefan _______________________________________________ XML-SIG maillist - XML-SIG@python.org http://mail.python.org/mailman/listinfo/xml-sig
