https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6438
--- Comment #7 from Stephen Fisher <steve@xxxxxxxxxxxxxxxxxx> 2011-10-20 10:45:39 MDT ---
This comment in gtk/follow_[tcp|udp|ssl].c sounds like the issue reported in
this bug, specifically the end of the comment:
/*
* XXX - the routine pointed to by "print_line_fcn_p" doesn't get handed lines,
* it gets handed bufferfuls. That's fine for "follow_write_raw()"
* and "follow_add_to_gtk_text()", but, as "follow_print_text()" calls
* the "print_line()" routine from "print.c", and as that routine might
* genuinely expect to be handed a line (if, for example, it's using
* some OS or desktop environment's printing API, and that API expects
* to be handed lines), "follow_print_text()" should probably accumulate
* lines in a buffer and hand them "print_line()". (If there's a
* complete line in a buffer - i.e., there's nothing of the line in
* the previous buffer or the next buffer - it can just hand that to
* "print_line()" after filtering out non-printables, as an
* optimization.)
*
* This might or might not be the reason why C arrays display
* correctly but get extra blank lines very other line when printed.
*/
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
