Why is xml.dom.minidom so slow?

[Paul Boddie]

martin at v.loewis.de (Martin v. Löwis wrote in message news:<mailman.1041548525.2273.python-list at python.org>...
> 
> What *has* been demonstrated to be a speed-up over minidom is to use
> 4Suite's cDomlette.

I can definitely agree with this. While the XML work I've been doing
has also involved DOM operations (parsing isn't interesting on its
own, after all), I've seen the performance of different packages in
order of decreasing speed to be: cDomlette, minidom, 4DOM. However,
I'm deliberately not using the extra functionality of 4DOM.

[...]

> When completed, it still gives you a Python-conforming DOM tree. That
> DOM tree misses some of the DOM functionality, though, that's why they
> call it a Domlette.

Still, cDomlette does have things like importNode which are missing
from minidom (which at least was missing from PyXML a couple of
releases ago).

Paul