On Mon, Nov 07, 2005 at 11:16:43PM +0100, Geert Jansen wrote:
> Daniel Veillard wrote:
> >>Any reason why this couldn't or shouldn't be done the same way 
> >>xmlNewTextWriterMemory is done?
> >>Would only require the addition of write and close callbacks and not 
> >>require any of the other new stuff from the patch.
> >>    
> >
> >  Right, while Geert patch is correct, an approach where you reuse the
> >I/O callback to fill up a buffer looks cleaner to me.
> >  
> OK -- please see the attached patch.

  Looks good, just a few comments:

> XMLPUBFUN xmlOutputBufferPtr XMLCALL
>       xmlOutputBufferCreateBuffer     (xmlBufferPtr buffer,
>                                        xmlCharEncodingHandlerPtr encoder);

  A new API but it's not a bad idea, making the link between normal sources
and xmlOutputBuffer / xmlInputBuffer is always confusing new comers to libxml2.

> +static int
> +xmlBufferClose (void * context) {
> +    (void) context;
> +    return(0);
> +}

  That's absically a noop. I think this can be just removed and NULL be passed
as the function pointer, but to be sure I just removed the 
   (void) context;
line and kept the function declaration, but if you could check that just
removing xmlBufferClose() and passing NULL works that would shave a few
bytes off :-)

   Applied and commited after a bit of cleanup, but this need a bit more testing
before the release.

     Thanks a lot !

Daniel

-- 
Daniel Veillard      | Red Hat 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

Reply via email to