I'm disappointed that Wireshark doesn't seem to support Unicode or 8-bit ASCII. I can understand Unicode is difficult to parse because each character isn't represented by a fixed number of bytes, but I don't see the excuse for not having 8-bit ASCII. In any case, Unicode is the future, a lot of webpages are using it already and they come up as gibberish in Wireshark. If you follow a TCP stream there is even an option for viewing the now obsolete EBCDIC but no Unicode. asked 02 Oct '10, 18:00  Eventual Zen |
This is a static archive of our old Q&A Site.
Please post any new questions and answers at ask.wireshark.org.
And the question is?
See the first item in
There's a significant amount of work involved in doing it right.
Note also that "Follow TCP Stream" is sort of "dumb by design", in that it just dumps raw TCP payload. To do more than just printable {ASCII,EBCDIC} (note, BTW, that EBCDIC wasn't obsolete when that capability was added, as somebody found that useful enough to implement...), either it would need to hook into subdissectors, so that it would see the payload as more than just a pile of octets, or there would need to be "Follow HTTP conversation" etc. that hook into the subdissectors.