Re: [xml] Performance xmlXPathEvalExpression

[Daniel Veillard <veillard redhat com>]

On Tue, Aug 14, 2007 at 10:47:32AM +0200, Shirazi Babak wrote:
> I tried this with a 10 MB file:
>
>    *doc = xmlReadFile(ucFilename, NULL, XML_PARSE_COMPACT | XML_PARSE_NOBLANKS);
>
> Without xmlParseFile : over 10 minutes
> With Xml ReadFile : Over 6 minutes
>
> This is an enhancement but not acceptable either.

  I guess it's time at this point that you look at your environment 
and do the debug on your system ! There is only so much we can do for you
and only getting negative feedback for the free tool and support you have
been given starts to get close to impolite.
  Would you mind checking if Microsoft support is as nice and cheap,
I think you need a small reality check at this point, I really really didn't
like your message about MSXML earlier today. Of course you can use something
else, trying to use it to get more attention than you deserve is really an
unpleasant tactic !
Your system is swapping. If you don't understand or don't want to look at it
use something else. If you can do a bit of analysis instead of complaining
you could for example realize that there are streaming parsing APIs in
libxml2 and that data extraction that way could be done in very little 
memory and way more efficiently than with XPath. But if that means we need
to baby step you all the way until you manage to get 'something' we have
no idea about to work, really please consider alternatives.

  thank you,

Daniel

-- 
Red Hat Virtualization group http://redhat.com/virtualization/
Daniel Veillard      | virtualization library  http://libvirt.org/
veillard redhat com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine  http://rpmfind.net/