On Fri, Mar 25, 2005 at 05:13:55PM -0500, Rob Richards wrote: > Daniel Veillard wrote: > > > I still think we should xmlFree() the name parameter. The underlying > > reason > >is that it's an API refinement and I prefer an immediate crash immediately > >identified in case someone expected a different behaviour than a slow > >memory > >leak which may only be tracked painfully months later. > > The function eats the name that's expected I think. > > The case of xmlNewNodeEatName the only error is when the program runs out > >of memory, then it will generate a leak, I will fix that, it's broken. > > > > > Hadn't thought of it that way, but I would have to agree with that > point. I tend to think more along the lines that if a function fails its > my job to perform my own cleanup. > > If free'ing the name is the way to go do you want another patch?
If you have the patch handy, yes, sure, otherwise it's really not a big deal :-) > Re-tested the regressions adding xmlFree when called by eatname on the > name in both spots (invalid node type and xmlMalloc failure in the > xmlNewPropInternal function) and everything ran fine without leaking > this time. good :-) Then send the CVs diff to be sure we came to the same conclusion :-) Daniel -- Daniel Veillard | Red Hat Desktop team http://redhat.com/ [EMAIL PROTECTED] | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/ _______________________________________________ xml mailing list, project page http://xmlsoft.org/ [email protected] http://mail.gnome.org/mailman/listinfo/xml
