Conceptual flaw in pxdom?

[Stefan Behnel]

Paul Boddie wrote:
> On 18 Mai, 08:54, Stefan Behnel wrote:
>> Emanuele D'Arrigo wrote:
>>> I'm looking at pxdom and in particular at its foundation class
>>> DOMObject
>> I didn't know pxdom, but looking at it now I can see that it hasn't been
>> updated since 2006. Not sure if that means that it is complete or that it
>> has been abandoned.
> 
> Maybe the developer is mostly satisfied with it.

Well, hard to tell without asking the author. I'm far from saying that a
"complete" piece of software is a bad thing, but bit-rot is still the death
 of all now-working software.


>> Anyway, seeing that it only provides DOM compliance, without anything
>> further like XPath or whatever, and that it doesn't focus on performance in
>> any way, you might still be better off with ElementTree, which is in the
>> stdlib since Python 2.5 (and available for Py2.2+).
> 
> To put the inquirer's remarks in context, I suggested that he look at
> pxdom specifically as a replacement for minidom and in response to the
> following statement: "I've used etree and lxml successfully before but
> I wanted to understand how close I can get to the W3C DOM standards."
> Maybe you missed that thread, but here's a link to it:
> 
> http://groups.google.com/group/comp.lang.python/browse_frm/thread/d445363b99001ad6

Ah, yes, I missed that thread. I always wonder why people want "DOM
compliance". I don't consider that a value by itself, and I don't see the
advantage that the DOM API has over other XML APIs (which, most of the
time, were designed to make life simpler for DOM suffering developers). I
find it a lot more important to get the stuff working quickly and in a
(somewhat) robust way, which is hard to achieve in DOM code. It's pretty
easy to write unmaintainable code that uses the DOM API, though.

Stefan