["Adam Doppelt" <amd@xxxxxxxxxxxxxxxx>]

> > I wrote it as a heuristic dissector within the "http" subdissector
> > list. The
> > dissector checks for "<xml" at the start of the body and returns FALSE
> > if it
> > doesn't match. Is there a better approach?
>
> That's one way of doing it.
>
> Another way would be to check the Content-Type: value, if present; that
> would require that we add a string dissector table for Content-Type.
> That would also mean that only those types registered by the XML
> dissector would be dissected as XML.

That approach is better. I wanted to check the content-type but it wasn't
possible with the current http dissector. Perhaps I should send in the
plugin and we can revisit when the content-type dissector gets implemented.

Adam