0 - now that "tvb_ensure_length_remaining()" ensures that there's at
least one byte of data, we should use "tvb_length_remaining()" for
FT_PROTOCOL fields.
svn path=/trunk/; revision=9440
NOTE: I propose to use packet-MIME-TREE for future media types that will be
added to Ethereal (E.g., packet-image-png.c).
svn path=/trunk/; revision=9437
Windows - the problem is that GTK+ 1.3[.x] and later assume strings
handed to them are UTF-8 strings, not, for example, ISO 8859/x strings.
In packet-radius.c, re-define "isprint()" rather than #ifdeffing its use
(the old code was also incorrectly treating 0x7f as a printable).
svn path=/trunk/; revision=9436
Windows - the problem is that GTK+ 1.3[.x] and later assume strings
handed to them are UTF-8 strings, not, for example, ISO 8859/x strings.
In packet-radius.c, re-define "isprint()" rather than #ifdeffing its use
(the old code was also incorrectly treating 0x7f as a printable).
svn path=/trunk/; revision=9435
there's no data remaining - its callers largely depend on it doing so.
That means that the BEEP dissector doesn't have to check for it
returning 0.
svn path=/trunk/; revision=9433
packet-smpp.c).
Now the higher-level protocols show up without the need of unfolding the SMPP
dissector tree.
Add a new address type AT_STRINGZ, and use it for GSM SMS message reassembly.
Get rid of the different UDH IE subtrees, and replace them with one subtree
for all UDH IEs.
Add debug logging to SMPP and SMPP GSM SMS dissectors.
Move the Short Message reassembly from the SMPP dissector to the SMPP GSM SMS
dissector. Fix duplicated "reassembled in" header fields.
Rename header fields that are now part of SMPP GSM SMS (including the dissector
table name, so it has also been updated in packet-wsp.c and packet-wtp.c).
Add an explicit "if (!tree) return" in the WSP add_headers() method.
NOTE: it would be great if we were able to merge the existing packet-gsm_sms.c
and the SMPP GSM SMS dissector.
svn path=/trunk/; revision=9431
HTTP does.
Note that the RTSP RFC talks about packets with no content length,
although it also says they shouldn't exist.
Update a comment now that we *do* handle data that crosses TCP segment
boundaries.
Handle the payload length a bit differently, to make it a bit clearer
what's going on.
svn path=/trunk/; revision=9425
currently-selected packet.
Use "cfile.count" as the total number of packets in the capture.
Use "plurality" to display "packet" vs. "packets".
svn path=/trunk/; revision=9423
In the loop that processes TCP segment data, loop over the full reported
length of the segment, so that we throw an exception if the frame was
short.
svn path=/trunk/; revision=9420
the WTP length after processing the TPI.
Update reassembly so dissection of reassembled data is only enabled in the last
reassembled packet. This way, we alo see the WTP PDUs.
svn path=/trunk/; revision=9410
after it, treat it as the only fragment" stuff for
"fragment_add_seq_next()" as well, so we handle non-fragmented packets.
svn path=/trunk/; revision=9405
code.
Add the "Reassembled in" header field, and show it if the packet has been
reassembled, but not in the given packet. This also allows to use the "Jump to
the corresponding frame" feature in Ethereal.
svn path=/trunk/; revision=9401