I just checked in the code for the new header_field_info
structs. Laurent -- I know you're on a mission to convert
protocols to be filterable, so you'll have to add a couple
of fields to the header_field_info structs you create. You'll
see that it's pretty easy to do so --- I converted all
the existing hfinfo's w/o problem, but that's because I
didn't try to go back and re-work the bitfield stuff in
the dissector (except a little bit in token-ring).
The new method:
1. Allows you to dissect bitfields and add them
to the proto_tree more easily.
2. Allows users to filter on the value of a field,
even if that field comes from a bitfield
where its value is shifted over a few bits.
3. Allows value_strings to have keys which represent
the true values of the bitfield fields, not
their shifted values.
4. Lets you print in hexadecimal w/o reverting to
proto_tree_add_item_format().
I have tried to document the new structure in doc/proto_tree.
packet-sna.c, which I also checked in, is full of examples
of how to use bitfields with the new hfinfo structs.
At http://ethereal.zing.org/~gram/sample.html I have
the file I sent out on ethereal-dev the other week,
sna1.trc.gz. You can use it to see what the SNA dissector
looks like. The SNA dissector is not complete yet; I have
to put in a few more TH types (FID's >= 4).
--gilbert
