On Fri, Jan 05, 2001 at 02:58:11PM -0600, Jeff Foster wrote:
> 1) I like the concept, however I don't like adding another data structure
> to hold the dissector information. You could (should?) use the conversation
> data structure to implement this functionality.
Yes. The "Dissect As..." function (which should have "..." in the menu
item, as it pops up a dialog box) really means "dissect this
conversation as XXX".
> 2) Saving dissector address under the protocol id is an excellent idea,
> one that has been discussed here before. I would like to see this extended
> to all dissectors.
Yes - see my other response.
> 3) This needs to extend beyond the TCP/UDP realm to other protocols.
Yes - see my other response.
> 4) Change the 'protocol to decode as' to a scrolled list. I fear that this
> list may grow if your idea is accepted.
Yes - for example, the list of protocols to decode a given {TCP,UDP}
conversation as should probably include all protocols that run atop
{TCP,UDP} and that don't have heuristic dissectors (if the heuristic
dissector fails to recognize the packet as one of its own, it's not
going to do so just because you try to force it to do so).
That could be done, in part, by going through the {"tcp","udp"}
dissector table, but you'd also have to handle protocols that don't have
a port assigned to them at all, e.g. RTP.