make tcp_dissect_pdus() understand how to skip to the next PDU;
make the skip-to-the-next-PDU code stop attempting to decode
segments that are known to be completely within an already seen
PDU.
svn path=/trunk/; revision=10406
WTAP_ENCAP_PRISM_HEADER, WTAP_ENCAP_IEEE_802_11_WLAN_BSD, and
WTAP_ENCAP_IEEE_802_11_WLAN_AVS.
In the seek_read routine, set it for all 802.11 types.
svn path=/trunk/; revision=10404
improve the Info column text for reassembled messages;
register subdissectors for dissecting Q.931 IEs and make the
ISUP dissector call the codeset 0 IE subdissector through a
handle.
svn path=/trunk/; revision=10400
WV-CSP 1.2 specifications will be corrected and in approved state, we can
get rid of those comments (#ifdef ... #endif).
svn path=/trunk/; revision=10396
As this is experimental and does not yet work reliably this feature is
off by default. Enable it by using --with-krb5 during configure.
Please do not yet make use of this feature (i.e. please do not *use* the
defines provided by this option), until it has received further testing
and fixing.
Please run and check what the configure says.
svn path=/trunk/; revision=10395
(a FILE * if zlib isn't used, a gzFile if zlib is used).
Use "size_t" for the amount of data to read in
"eyesdn_check_file_type()", to squelch signed vs. unsigned warnings.
svn path=/trunk/; revision=10392
compile-time optimization; in particular, if you call
GSM_MAP_PARAM_DISPLAY with a Gtag argument of GSM_MAP_P_NONE, the
compiler does figure out that it only need generate code for the first
branch for the first if in that macro, and it sort of seems to
understand that _param_fcn is guaranteed to be null, but it doesn't
optimize out the test and the branch for _param_fcn not being null - it
does, however, generate a call to location 0, which, when you compile as
position-independent code, the assembler refuses to assemble.
Therefore, we put into the test for _param_fcn being null an additional
test for Ga1 being GSM_MAP_P_NONE, which at least manages to get the
code through the compiler.
svn path=/trunk/; revision=10388
Note that Ethereal doesn't work on pre-X Mac OS, in case somebody sees
".macos" and thinks it'll work on OS 9 or earlier versions.
Explain that GLib and GTK+ now need to be shared libraries, and that
the GLib 1.x and GTK+ 1.x release tarballs were built with an older
version of libtool that didn't support building them as shared libraries
on OS X.
svn path=/trunk/; revision=10384
* Correct the version checks (use path expansion
rather than regular expressions, and fail if the
tools are not installed at all).
* Make it possible to specify other names for the
auto* tools to use (e.g., automake-1.6 instead
of automake).
svn path=/trunk/; revision=10383
Read of color filters rewritten to remove possibility of buffer overflow
in read_filters_file() and to fix comment & blank line handling.
This removes any limitation (and stack overflow) on the lengths of color
filter name and filter expression which are now dynamically allocated.
Fix removal/deletion of color filters (avoid segmentation fault when
some color filter is deleted and then the list cleared for instance).
The use of g_slist_remove() within an iterator is simply unsafe if the
code is not prepared to handle this.
Fix a small memory leak in revert_filters().
Remove inappropriate comments in read_filters().
Miscellaneous code cleaning.
svn path=/trunk/; revision=10380
Between v2 and v3 some structures changed their definition.
I arbirtrarily made the preference set V2 as the default.
I have no idea what the most common ndmp version is maybe someone that knows
which is the most popular iversion in the wild will change it to default to that.
svn path=/trunk/; revision=10362
firewall/Symantec Enterprise Firewall. Thanks, Axent/Symantec, for not
asking us for a DLT_ value and not telling us about the link-layer type.
svn path=/trunk/; revision=10361
