Re: [xml] reporting bug for lixml2.2.6-8

[Daniel Veillard <veillard redhat com>]

On Mon, Aug 30, 2004 at 12:20:37PM -0400, Rich Salz wrote:
> > I would be willing to put up a (reasonable) wager that changing that 
> > statement
> > to
> >    if ((str >= (xmlChar *)&pool->array) && (str <= pool->free))
>
> You'd lose.  The standard says that if "str" is not within the range of 
> the pool, then the results of the above expression are *implementation 
> defined.*

  Okay, then libxml2 won't work on an implementation where this doesn't
work. More precisely  the dictionnaries won't work, which mean parsing speed
cost, and special compilation options. If we end up with a bug report for
such an architecture then we will take the appropriate measures. In the 
meantime the code will stay as it is.

> Now, it happens that on almost every machine you're likely to run 
> across, the "implementation defined" behavior is "we'll do the right, 
> and obvious, thing."  But there is no guarantee.  It would be nice if 
> Insure had an option to ignore that complaint.

  Simply click on the bug in insra, set the ignore toggle, live happy.

> Depending on how 
> standards-compliant you want to be, it might be nice if someone patched 
> in my idea.

  If it so slow that it makes dictionnaries impracticable, better simply
ignoring dictionnaries and mallocing all strings.

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/