Cleanup for 1.0.50

[I'm resending this because I'm fairly certain the original (from 
last night) got lost in a mail glitch on my end.  My apologies if I'm 
wrong and two copies get to the list. -RMS]


In light of the current push to get stable versions of the various 
GNOME components for a 1.0.50 (1.2? I'm still not sure which is the 
correct designation for the next stable GNOME release...) release, 
I've temporarily stopped my attempts to add snooze functionality to 
gnomecal (mentioned in an earlier message to the list).  I was 
getting mysterious segfaults, so I figured it was safer to save the 
diff from my working directory, check out a fresh copy of the 
sources, and get back to it after the stable release.

But since I need something to divert my mind from my first few weeks 
of law school, I figured I'd go through the gnome-pim bugs in the BTS 
and try to fix any simple ones which were lying around and ought to 
be fixed before the release.

Several notes/questions that come to mind.  Warning: many of these 
are entirely unrelated to anything other than the fact that they came 
up during my hacking/debugging session today:

* Someone more knowledgable than me should probably go through the 
bugs and evaluate which ones need to be fixed before 1.0.50 and which 
don't.  If nobody else can do it, I will, but I'm not sure I'm likely 
to get those determinations correct.

* Bug #1316 contains a patch which the submitter indicates fixes the 
problem (compilation on IRIX 6.x) but which does not seem to have yet 
been applied to CVS.

* Is there a reason that the default button in the Event Editor is 
Cancel, rather than OK?  I suspect that's at least partially to blame 
for bugs like #1472.

* In trying to track down bug #1514, I found a weird oddity in 
eventedit.c...  I was tracing through ee_rp_init_rule(), which 
contains the following code, beginning at line 978:

	/* Default to today */

	week_vector = 1 << tm->tm_wday;
	default_day = tm->tm_mday;
	def_pos = 0;
	def_off = 0;

Now, the code itself looks fine, although something seems to be 
corrupting the value of tm.  Look at the following excerpt from my 
gdb session, before and after I execute line 982 (the default_day 
assignment, above):

(gdb) p default_day
$18 = 0
(gdb) p tm
$19 = (struct tm *) 0x40491910
(gdb) next
982		def_pos = 0;
(gdb) p tm
$20 = (struct tm *) 0x1

I'm at a loss, and I'm not even sure if this has anything to do with 
the reported bug, but it's driving me a bit crazy. Am I missing 
something obvious?

Ok, that's all I've come across so far.  Sorry for the randomness of 
this message. :)

Russell Steinthal		Columbia Law School, Class of 2002
<>		Columbia College, Class of 1999
<>		UNIX System Administrator,

NOTE: My address will stop working soon... Use one
of the current alternatives listed above!

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]