RE: [PHP-CVS] cvs: php4 /ext/qtdom qtdom.c qtdom.h

> It seems like GNOME dom parser uses twice the amount of memory. I > parsed a > document containing: > <doc name="test"> > </doc> Just to speak from a practical point of view, the GNOME parser generally doesn't deal with documents like the above too intuitively. Mainly because it treats all the information outside of tags as a node. If I'm parsing an XML data string, I generally run it through this first: $xml = trim(preg_replace('/>\s+</m', '><', $xml)); .. which cuts the size of the resulting tree in half. Of course, you have to be sure that the newlines are meaningless in your data, since: <doc name="test"> </doc> Is technically different than: <doc name="test"></doc> ... but you get the idea. Whether this is a bug or feature of the GNOME parser, someone else can elaborate. :) - Colin