Today I saw an expert info about an invalied hf_index or somethin like that
on a customer's windows 1.10.5 wireshark in packet-t124.c. Which I can't reproduce
on my machine (suse13.1, svn head).
The only "interesting" change seems to be:
jmayer@egg:~/work/wireshark/svn/trunk/asn1/t124> svn log -c49648 packet-t124-template.c
------------------------------------------------------------------------
r49648 | morriss | 2013-05-31 04:27:40 +0200 (Fr, 31 Mai 2013) | 6 lines
Don't intentionally use -1 as an hfindex: while only naughty before, with the
changes from r49644 you'll get complaints about the hf being unregistered...
Note that this code also intentionally passes -1 as an ett_index. It doesn't
currently complain so I'm leaving it alone for now...
------------------------------------------------------------------------
jmayer@egg:~/work/wireshark/svn/trunk/asn1/t124> svn diff -c49648 packet-t124-template.c
Index: packet-t124-template.c
===================================================================
--- packet-t124-template.c (revision 49647)
+++ packet-t124-template.c (revision 49648)
@@ -163,9 +163,11 @@
* to have a version of dissect_per_sequence() that checks all
* references to the tvbuff before making them and returning "no"
* if they would fail.
+ *
+ * We (ab)use hf_t124_connectGCCPDU here just to give a valid entry...
*/
TRY {
- (void) dissect_per_sequence(tvb, 0, &asn1_ctx, NULL, -1, -1, t124Heur_sequence);
+ (void) dissect_per_sequence(tvb, 0, &asn1_ctx, NULL, hf_t124_connectGCCPDU, -1, t124Heur_sequence);
} CATCH_BOUNDS_ERRORS {
failed = TRUE;
} ENDTRY;
So maybe it should be backported to 1.10?
Ciao
Jörg
--
Joerg Mayer <jmayer@xxxxxxxxx>
We are stuck with technology when what we really want is just stuff that
works. Some say that should read Microsoft instead of technology.