is greater than 3 we can get detailed interface information. This should
limit the amount of work we have to do when new WinPcap releases come out.
svn path=/trunk/; revision=20542
pinfo.curr_proto appeared as pinfo.match
pinfo.columns in the refman was pinfo.cols in lua (now it is both in lua and in refman)
svn path=/trunk/; revision=20539
exception rather than aborting the program; using it means that
dissector bugs show up as such rather than as malformed packets.
svn path=/trunk/; revision=20532
scsi sense information potentially followed by iscsi event data.
this is used by a target to tell an initiator (among other things) that the lun configuration has changed and now is a good time to rescan the target for lun changes.
svn path=/trunk/; revision=20531
"The Help ! Contents and the Help ! User's Guide both display the User's
Guide."
Remove the duplicate Help!/User's Guide item
svn path=/trunk/; revision=20523
as an argument, and, on an error, if they have an error message, have
them set that "gchar *" to point to a g_malloc()ed string containing the
error message, rather than taking a pointer to a buffer for that message
as an argument. That's more like what's done in Wiretap, and doesn't
impose an upper limit on the lengths of those error messages. If that
pointer is null, don't allocate the message string and return it.
Have that error message already have the "cant_get" processing applied
to it, so nobody other than those routines need to call the "cant_get"
routines to process the error messages.
Have get_airpcap_interface_list() explicitly set "*err" to the
appropriate error code.
Clean up indentation.
svn path=/trunk/; revision=20521
I found a rare situation in which the BSSAP dissector seems to wrongly
assume a packet.
When a RANAP DirectTransfer message contains the GSM Supplementary
Service 'Call Confirmed' this seems to yield a message that the BSSAP
dissector recognizes as a BSSMAP BLOCK message (and from the perspective
of BSSAP, this is perfectly correct).
My patch includes code that checks this very special case.
svn path=/trunk/; revision=20520
for consistency. The frame.pkt_len filter is now deprecated, but still
supported as a hidden field for an easy transition. The new field name is
frame.len.
svn path=/trunk/; revision=20519
patch against a crash of wireshark with gtk1.2 when using packet
length statistics window. The crash was due to a difference of
implementation of g_strsplit between gtk1.2 and gtk2.
svn path=/trunk/; revision=20518
default font of Sans 10 instead of the setting in Wireshark (such as
Monospace 10). On some systems this bug was causing the contents of columns
such as the command line specified (CLS) time to be cut off.
svn path=/trunk/; revision=20512
A user asked what file format the flow graph is saved as, which isn't
apparent by looking at the dialog. I had to try saving a file and check
its type -- plain text file. Could someone please apply the attached
patch or similar change to graph_analysis.c. My patch updates the title
of the dialog box from "Wireshark: Save graph to file" to "Wireshark:
Save graph to text file".
However, I used the term "plain text file" instead ;-)
svn path=/trunk/; revision=20510
property, which selects the entire contents of a label when it comes
into focus. This property annoyingly defaults to TRUE, which meant that
the labels in simple dialogs and the about box would unexpectedly show
up selected.
Work around this by setting the focus on the "OK" button in the about
dialog and the first button specified in simple dialogs.
svn path=/trunk/; revision=20507
This patch adds support for the IPv4 Commercial IP Security Option (CIPSO) as
defined in the IETF draft, draft-ietf-cipso-ipsecurity-01.txt. While this
draft has long since expired, it has become a de-facto standard for labeled
networking with support from several commercial Multi-Level Security (MLS)
operating systems such as HP-UX CMW and Trusted Solaris; in addition, Linux
Kernels 2.6.19 and later provide support for CIPSO in conjunction with
SELinux.
Copies of the expired CIPSO draft can be found at the NetLabel project page:
* http://netlabel.sf.net
svn path=/trunk/; revision=20506
having been reassembled.
Fix the comments in reassembly.c and reassembly.h regarding what the reassembly
routines actually return in the 802.11 and no-sequence-number cases when they
are given the first and last packet (that is, a non-segmented packet): in
particular the routines return a pointer to a list containing just the one
fragment.
svn path=/trunk/; revision=20505
code in each function is consistent (previously there were some sections of
code indented, oh, about 10 indentations too far). At the same time, remove
trailing white space and change spaces to tabs.
Remove one duplicated call to sccp_assoc() and s/isup_apm/sccp/g in the hf[]
array (apparently when the XUDT reassembly was put in it was copied from ISUP
but the filter names weren't changed).
svn path=/trunk/; revision=20502
A clean_setup is done first now to ensure that the target dirs don't contain old files - probably remaining from a previous setup run.
explicit removing of the gtk2 dir is no longer required therefore
svn path=/trunk/; revision=20497
I've added a call to dos2unix to convert check_svn_version.sh that way
Is it safe to assume that dos2unix is available on a common UNIX developer machine?!?
svn path=/trunk/; revision=20491
Stephen Fisher