Re: [xml] Expat + LibXML DOM vs LibXML

[Luca Padovani <lpadovan cs unibo it>]

On Sat, 2003-03-08 at 17:31, Jerome Pesenti wrote:
> Has anyone tried to compare using Expat SAX + LibXML
> DOM vs LibXML SAX+DOM?

not me, although I've always found the fact that libxml has BOTH an XML
parser AND a DOM-like document representation sort of weird. I wonder if
anybody else has ever thought about separating them in two (or more)
distinct sub-libraries...

> Using simple XML (i.e., element nodes, text and
> attribues only, no comments etc). I get a 20%-30%
> speed improvement (for example, an 8M file takes 500ms
> instead of 630ms).

when you say "using"... what do you mean exactly? Like receiving SAX
events from expat and building a libxml DOM tree? If so, do you think
the speed improvement may have something to do with your definition of
"simple XML"?

> I can post the C code used for testing and some more
> comprehensive results if anyone is interested....

I'm definitely interested and I'd like to have a look at your code. If it is
large, perhaps it is better if you send it to my own email address only.

Thanks,
-- luca