helper routines for packet dissecting, and away from "util.c", which is
now all GUI-related. (Among other things, this makes life more pleasant
for Gilbert Ramirez's "tethereal" stuff, although a lot more separation
of GUI from other stuff needs to be done to make that - or a
"curses"-based variant of Ethereal, or a variant using some other GUI
toolkit - work smoothly.)
svn path=/trunk/; revision=235
proto*() functions. The configure script tries to use ipv6 name resolution if
it knows the type of ipv6 stack the user has (this can be avoided with the
--disable-ipv6 switch) Additionally, the configure script now deals with wiretap
better. If the user doesn't want to compile wiretap, the wiretap is never
visited. A few unnecessary #includes were removed from some wiretap files, and
a CPP macro was moved from bpf.c to wtap.h.
svn path=/trunk/; revision=229
(presumably a Windows version).
Note also that version 2.001 files appear to have microsecond time
stamps, like version 1.1 files.
svn path=/trunk/; revision=228
reference the protocol tree with struct proto_tree and struct proto_item
objects. That way, the packet decoding source code file can be used with
non-gtk packet decoders, like a curses-based ethereal, e.g. I also re-arranged
some of the information in packet.h to more appropriate places (like other
packet-*.[ch] files).
svn path=/trunk/; revision=223
In the detailed expansion of an RR, give a more detailed description of
the RR type, as per Peter Hawkins' suggestion, but leave the record type
in the summary line for the RR, along the lines of John McDermott's
suggestion.
Decode PTR and CNAME RRs.
svn path=/trunk/; revision=222
"dissect_data"; if you call "dissect_data()" with an argument list of
the type expected by a "capture_XXX()" routine, it won't do the right
thing (and may do the wrong thing, e.g. crash).
Have "sap_capture_func()" (and "sap_dissect_func()") return function
pointers rather than "void *"s, so that I don't make a mistake like the
above in the future....
svn path=/trunk/; revision=219
This assumes that the time stamps are still in units of microseconds; I
don't yet have a text decode of the version-2.001 file from the program
that decoded it, so I can't check the time stamps.
svn path=/trunk/; revision=218
This assumes that the time stamps are still in units of microseconds; I
don't yet have a text decode of the version-2.001 file from the program
that decoded it, so I can't check the time stamps.
svn path=/trunk/; revision=217
It seems that a stable version of the library received a new function. This
should help RedHat folks, since they seem to have glib-1.0.1.
svn path=/trunk/; revision=216
NCP is still not decoded much, but the infrastructure for doing so is now in
place, including a hashtable to record the NCP type of each request so that we
now how to parse the response.
svn path=/trunk/; revision=215
and use them to extract stuff in "bpf_mk_bytecmp()", so as to avoid core
dumps on processors that require strict alignment.
svn path=/trunk/; revision=214
they're not used outside this file.
Compute their sizes with the standard "number of elements in an array" C
idiom, rather than hardcoding them as numbers.
Add the "information request" and "information reply" ICMP packet types.
svn path=/trunk/; revision=212
appears to be the UNIX "time_t" when the capture started, so use that to
figure out the time when a packet was captured.
svn path=/trunk/; revision=204
by Network General (subsequently merged with McAfee Associates into
Network Associates), called "Sniffer Basic".
A similar format appears to be used by the Windows Sniffer Pro.
svn path=/trunk/; revision=194
Guy Harris