Hi,
Use the undo_fix_852.patch to, well, undo the the fix.
Thanx,
Jaap
On Wed, 1 Nov 2006, Small, James wrote:
> Thank you Steve, I believe you are right.
>
> Jaap/Ulf - I know you are busy and this does not appear to be a high
> priority bug. Is there any work around to "disable" the bug 852 fix so
> that if you want to easily display TCP stream text and are willing to
> except the crash risk you can?
>
> Thanks,
> --Jim
>
> -----Original Message-----
> > Except--when I follow the TCP stream with Ethereal 0.99, this works
> > great. However, when I do the same thing with Wireshark 0.99.3/4
> > (I've tried 0.99.3 and just uninstalled/re-installed 0.99.4), the
> > password does not appear in the ASCII/Raw decoding screens.
>
> This appears to be related to bug #1043:
>
> http://bugs.wireshark.org/bugzilla/show_bug.cgi?id=1043
>
> The work-around has caused the last character of each packet to be
> dropped, which is where the username and password characters are in a
> telnet session.
>
>
> Steve
>
> _______________________________________________
> Wireshark-users mailing list
> Wireshark-users@xxxxxxxxxxxxx
> http://www.wireshark.org/mailman/listinfo/wireshark-users
>
>
Index: gtk/follow_dlg.c
===================================================================
--- gtk/follow_dlg.c (revision 18001)
+++ gtk/follow_dlg.c (revision 18000)
@@ -985,7 +985,6 @@
#endif
#if GTK_MAJOR_VERSION >= 2 || GTK_MINOR_VERSION >= 3
- gboolean line_break = FALSE;
/* While our isprint() hack is in place, we
* have to use convert some chars to '.' in order
* to be able to see the data we *should* see
@@ -995,21 +994,11 @@
for (i = 0; i < nchars; i++) {
if (buffer[i] == '\n' || buffer[i] == '\r')
- {
- line_break = TRUE;
continue;
- }
if (! isprint(buffer[i])) {
buffer[i] = '.';
}
}
-
- /* XXX - workaround for bug 852
- * Force a line break so that the text view
- * doesn't blow up on excessive long lines.
- */
- if (line_break == FALSE)
- buffer[--i] = '\n';
#endif
#if GTK_MAJOR_VERSION < 2