Re: [evolution-patches] gtkhtml, evolution crash when compose a html message

[Eric Zhao <Eric Zhao Sun COM>]

Thanks for your comment.

Acctually,  the corresponding HTML object of a11y object should not be 
NULL.  When a HTML object is destroyed, the relation between this HTML 
object and its a11y object is cut off, and a signal is emitted to notify 
the AT tools to destory the a11y object. But I found this notification 
has no effect on gnopernicus(gnopernicus still tries to access this a11y 
object).  So I use g_return_if_fail as the last resort.

Regards,
Eric

Not Zed wrote:

> Just a comment, you shouldn't use g_return_if_fail if it is a possibly 
> normal/valid condition.  Only if it is an error.  e.g. it might be ok 
> in the code for that pointer to be NULL.
>
> The code should still work and not crash without any of those 
> assertions present.
>
> On Wed, 2004-08-04 at 09:38 +0800, Eric Zhao wrote:
>
> > Hi,
> >  The attachment is a patch for bug #62223 .( 
> > http://bugs.ximian.com/show_bug.cgi?id=62223 ).
> >
> >   Could you spend some time to give it a review? Thanks!
> >
> > Regards,
> > Eric
> > _______________________________________________
> > evolution-patches mailing list
> > evolution-patches lists ximian com <mailto:evolution-patches lists ximian com>
> > http://lists.ximian.com/mailman/listinfo/evolution-patches
> >    
> --
> 	*Michael Zucchi* <notzed ximian com <mailto:notzed ximian com>>
> /"born to die, live to work, it's all downhill from here"/
> Novell's Evolution <http://codeblogs.ximian.com/blogs/evolution/> and 
> Free Software <http://www.gnu.org/philosophy/free-sw.html> Developer