Reinhard Speyerer schrieb:
> Christian Gurk wrote:
> 
>> So I wrote the test dissector and it showes up in the Enable Protocols 
>> List (and of course, I activate it). But when I try to use it with 
>> "Decode as", this menu item is grayed out, so I can't use it.
> 
>> Here is the code of my first test:
>>> -- trivial protocol example
>>> -- declare our protocol
>>>
>>> mpic_proto = Proto("mpic485","MPIC 485 bus protocol")
> 
>>> -- local packet_table = DissectorTable.get("packet")
>>> -- packet_table:add(packet.USER15,mpic_proto)
> 
> Hi Christian,
> 
> the following Lua code
> 
> local wtap_encap_table = DissectorTable.get("wtap_encap")
> wtap_encap_table:add(wtap.USER15, mpic_proto)
> 
> combined with text2pcap -l 162 is probably the easiest way
> to use your Lua dissector for your text file packet data.
> 
> Regards,
> Reinhard
> ___________________________________________________________________________
> 
Thanks Reinhard for your answer, but I tried the same already with the 
K12-type text input, which worked fine in principle.
My problem is, that I would like to use an own link-layer type protocol. 
But I played around al little bit and it seems, that I HAVE to enable 
the "ethernet" Protocol do do anything with the data. Is that true?
Is it somehow possible to register an own link-layer protocol as a first 
dissector? I would like to have sometthing like:

link-layer:             pass data to i-layer
internet-layer:         1-4 byte dest and src adresses
transport-layer:        port byte
application-layer:      time,ftp,string,system... (all proprietary)

Hmm, difficult to describe.
Thanks and greetings

-- 
Christian Gurk
Electronics Working Group / Research & Development
Max-Planck-Insitute for Chemistry, Mainz, Germany
+49-6131-305387 / [email protected]

___________________________________________________________________________
Sent via:    Wireshark-dev mailing list <[email protected]>
Archives:    http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
             mailto:[email protected]?subject=unsubscribe

Reply via email to