Re: [xml] XPath question w/ libxml2

[Daniel Veillard <veillard redhat com>]

On Fri, Apr 16, 2004 at 11:50:16AM -0600, Mark_Vakoc peoplesoft com wrote:
> First of all, I want a better compiler....
> > > I'm afraid you might be seeing some weird floating point troubles.
>
> You have no idea...

  Actually, the whole NaN and Inf xpath support has been hell to get running
cleanly on a variety of compilers and platform, so I have an idea :-)

> C:\>test
>   Object is a string : NaN
>   Object is a number : NaN
>   Object is a number : 1
> obj1(1.#QNAN0) is equal to obj2(1.000000) : 1
> C:\>
>
> Seems bizarre to me.

  Nice ! Seems the Free Download of the MSC compiler now reflect better the
value of the tool :-)

> The attached patch fixes this specific problem
> with msvc, but I wonder where else this might bite.

  That should be safe, it's just costly. So to minimize the impact,
I swapped to make the atomic test first, and second used conditional
compilation to avoid the penalty on platforms which should not suffer from
it:

#ifdef WIN32
            return(((res->floatval == ctxt->context->proximityPosition) &&
                   (!xmlXPathIsNaN(res->floatval)) /* MSC pbm Mark Vadoc !*/);
#else
            return(res->floatval == ctxt->context->proximityPosition);
#endif

  That should go in the next release,

    thanks,

Daniel

-- 
Daniel Veillard      | Red Hat Desktop team http://redhat.com/
veillard redhat com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/