On Sat, Oct 19, 2002 at 11:01:59AM -0500, Ethan Blanton wrote:
> David Odin spake unto us the following wisdom:
> > On Fri, Oct 18, 2002 at 09:57:58PM -0500, Ethan Blanton wrote:
> > > And I also sent a counter-patch that got held up by moderation. :-P
> > Great. I have just a few remarks:
> > - You've put the input entry above the iconbar. Is this on purpose? I
> > tend to prefer the iconbar above the entry.
> Oops! That was a last-minute breakage, probably caused when I put in
> the frame... I certainly did not mean to do it. The patch has been
> updated on the web.
> > - You've replace the 'entry_view' field by 'entry'. I guess it was to
> > reduce the patch size. I'm afraid this could cause some hard-to-find
> > bugs because the 'entry' field can be used with its former meaning
> > (i.e a pointer to a GtkText, not a GtkTextView) by other part of
> > gaim's code.
> As Nathan said, your changing of the variable name guarantees that all
> instances were caught. That said, I'm pretty sure the notify plugin
> is busted with this patch. At one point I had hacked a Makefile not
> to build it, and I don't recall fixing it...
Gtk should issue warning if we use a GtkTextView where a GtkText is
expected anyway. So, these sort of problem should be caught easily.
By the way, I've seen on the gtkspell's official site that the current
version use a GtkTreeView and is ready for gtk+2. What is the policy
there? Grab the source and put them in the gaim tree (after having
informed the gtkspell author of course)? Add a new dependency for gaim
on gtkspell? I would prefer the former.
The biggest mistake you can make is to believe that you are working for