Add support for ETV Data processing & simple MPEG DSM-CC handling.
Witha a change of the name of dissect() in packet-etv.c to dissect-etv_common().
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6978
svn path=/trunk/; revision=41735
The attached (trivial) patch adds some missing breaks in switch-case blocks.
This fixes coverity defects #445, #446, #1316 and #1380.
svn path=/trunk/; revision=41727
Fix
Coverity has 3 complaints about this dissector, reported in CID's 1450-1451
CONSTANT_EXPRESSION_RESULT:
Event result_independent_of_operands: (crid & 3) >> 2 is 0 regardless of the
values of its operands [non-specific operand of assignment].
1385 crid_type = (crid &
MPEG_DESCR_CONTENT_IDENTIFIER_CRID_LOCATION_MASK) >> 2;
Event result_independent_of_operands: flags & 0x1000 is always 0 regardless of
the values of its operands (logical operand of if).
1512 if (flags & MPEG_DESCR_LOGON_INITIALIZE_CONNECTIVITY_MASK) {
Event result_independent_of_operands: flags & 0x1000 is always 0 regardless of
the values of its operands (logical second operand of '&&').
1544 if (offset < end && (flags &
MPEG_DESCR_LOGON_INITIALIZE_CONNECTIVITY_MASK)) {
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6960
svn path=/trunk/; revision=41722
are present. However, still only create the graph for the first/only
one.
LTE MAC or RLC frames often contain multiple SDUs that are segments of
the same TCP conversation - this avoids the need to find a frame with
only one SDU.
svn path=/trunk/; revision=41721
ISAKMP : VendorID CheckPoint : Malformed Packet
Wrong minimum length when there is not CheckPoint "Data" after VendorID
svn path=/trunk/; revision=41718
IKE: Life-Duration wrong displayed in LTV format
The Fix commit in Revision 38387 is wrong
Get the following error message : ISAKMP dissector bug: failed assertion proto.c:2957
use a HF by type (uint32, uint64, bytes...) for life duration
svn path=/trunk/; revision=41717
Guards the calledOnce boolean, against malformed traps, using register_frame_end_routine.
Fixes openSAFETY - Malformed packets disable the dissector for the WS session
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6950
svn path=/trunk/; revision=41674
I'm contributing a new dissector for the HART/IP protocol. This
protocol is specified by the HART Conformance Foundation (HCF). It is
a standard protocol used in the process control industry. It
essential wraps the multip-drop serial HART packets in TCP or UDP
packets. The standard has been approved by the HCF and has been
assigned UDP/TCP port 5094 by IANA.
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6961
--This line, and those below,
will be ignored--
M AUTHORS
M epan/CMakeLists.txt
M epan/dissectors/Makefile.common
AM epan/dissectors/packet-hartip.c
M ui/gtk/main_menubar.c
svn path=/trunk/; revision=41644
Repeat r41359 in another location:
Don't try to display tone entries if there are none (but there is length for
them). Could/should be an expert_info.
Also fix a typo: Nomber->Number.
svn path=/trunk/; revision=41639
Don't use a proto_item as the tree in proto_tree_add_text(). (There's code
here to hang a subtree off the item but it's commented out.)
svn path=/trunk/; revision=41638
tidlist is registered as a FT_UINT16 so don't add an array of 16-bit entities
(tids) using it. Instead add each TID individually.
svn path=/trunk/; revision=41637
Conversation address generation during processing of endpoint descriptors
is not being done in a manner consistent with that of general USB
packet dissection. I am not aware of any observable manifestation of this
bug.
svn path=/trunk/; revision=41634
cover the entire MCS field, in the hopes of avoiding the need to ask
questions such as
http://ask.wireshark.org/questions/9461
svn path=/trunk/; revision=41633
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6949 :
Don't use a guint8 for() loop index when the end of the loop value is stored
in a guint: the loop index may never reach the end-of-loop value.
Also remove a shadow'd variable.
svn path=/trunk/; revision=41625
Stig Bjørlykke