[lxml-dev] fix for another memory leak

[Stefan Behnel]

Narayan Desai schrieb:
> I updated to the current svn, and it looks like this chunk didn't get
> committed. Is this intentional?

Oh, thanks for checking! :)

It was by accident. It looked so much like the patch I had committed before
the last one that I just thought you had copied it. I'll check it in as soon
as I get to it.

Thank you for taking care,
Stefan


> Index: src/lxml/etree.pyx
> ===================================================================
> --- src/lxml/etree.pyx  (revision 23382)
> +++ src/lxml/etree.pyx  (working copy)
> @@ -775,16 +775,18 @@
>      
>      def __getitem__(self, key):
>          cdef xmlNs* c_ns
> -        cdef char* result
> +        cdef char* cresult
>          ns, tag = _getNsTag(key)
>          if ns is None:
> -            result = tree.xmlGetNoNsProp(self._c_node, tag)
> +            cresult = tree.xmlGetNoNsProp(self._c_node, tag)
>          else:
> -            result = tree.xmlGetNsProp(self._c_node, tag, ns)
> -        if result is NULL:
> +            cresult = tree.xmlGetNsProp(self._c_node, tag, ns)
> +        if cresult is NULL:
>              # XXX free namespace that is not in use..?
>              raise KeyError, key
> -        return funicode(result)
> +        result = funicode(cresult)
> +        tree.xmlFree(cresult)
> +        return result
>  
>      def __len__(self):
>          cdef int c