Hi,
Some random ideas that some one with unlimited time on their hands could dig into :-)
Seriously - I thought I'd just scribble down some things I've idly been thinking about to see if it spurs any interest in implementing them or why it's a bad idea.
If any one is interested in any of the items listed and wants to continue the discussion break that out in a separate email. Feel free to add to the list.
Reduce Memory usage:
- In frame_data.h it should be possible to cut down on the nstime_t structures by only having abs_ts and the others could be offsets to that or something similar thus saving a couple of bytes per packet.
- In the reassembly routines If I remember correctly, I might be wrong I think we may waste memory for TCP. Fragments and reassembled fragments?
- It might be possible to store file pointer and length rather than the fragment data and read in that data when needed.
Filtering:
- We have a string with per packet protocols, could that string be saved and used in filtering - don't dissect the packet if it does not contain the filtered protocol.
Pcap-ng
- The defined blocks are capture oriented should we define some analysis re-saving oriented ones.
- UDP/TCP/SCTP... port map similar to the NRB (think decode as)
- Read filter used ( save filtered trace)
- File history ( saved file A as B (using read filter X) ...)
- ...
Features
- Ability to edit conversations protocol data, make it possible to edit or add data to a conversation after own analysis.
:
Best regards
Anders
