good catch, committed with that change, thanks
On Wed, Feb 03, 2010 at 03:38:02PM -0800, Micah Cowan wrote:
> It looks right to me; except that you may want to move the xfree(label)
> back down where it was before (with a NULL check on label), as otherwise
> it'll leak memory if some silly user s
It looks right to me; except that you may want to move the xfree(label)
back down where it was before (with a NULL check on label), as otherwise
it'll leak memory if some silly user specifies both -S _and_ -L.
-mjc
Nicholas Marriott wrote:
> How about this instead? I think it should have the same
How about this instead? I think it should have the same effect.
Index: tmux.c
===
RCS file: /cvs/src/usr.bin/tmux/tmux.c,v
retrieving revision 1.66
diff -u -p -r1.66 tmux.c
--- tmux.c 3 Jan 2010 12:51:05 - 1.66
+++ tmux
Understood. I wasn't sure you would; I just dislike duplicating the same
if ((path = makesockpath(...)) == NULL) {
give_an_error_msg();
exit(1);
}
twice in a row (once in one branch, and again in a branch of an outer
if, IIRC - wherever the xmakesockpath calls are now). And avoiding the
d
Nice idea, will have a better look tonight but I'm not sure I like
the idea of makesockpath calling exit().
On Tue, Feb 02, 2010 at 11:32:26PM -0800, Micah Cowan wrote:
> So, at some point when I was testing some of the recent keybindings I
> introduced into tmux, I ran a separate tmux instance w
So, at some point when I was testing some of the recent keybindings I
introduced into tmux, I ran a separate tmux instance with "-L new" in a
new terminal, so I could test my changes. I typed "tmux list-bindings"
into the shell within this new server's session, and imagine my
confusion when I see a