[crossfire] Reminder: Release 1.11 forthcoming

[Kevin R. Bulgrien]

On Monday 21 January 2008 00:53, Kevin R. Bulgrien wrote:
> >   I delayed the 1.11 release a little to give the stringbuffer changes a
> > little time to soak in.  Those look good.
> >
> >   So in all likelihood, I'll be packing up a 1.11 release sometime soon -
> > maybe even as soon as this Sunday (1/20), so just a heads up.
>
> I'd really be interested in getting one more patch into the GTK V2 client
> before the release.  I just made a modification that seems to fix all of
> the problems listed in "[ 1527973 ] bind command does not work"
>
> <
> http://sourceforge.net/tracker/index.php?func=detail&aid=1527973&group_id=1
>3833&atid=113833
>
>
> The only thing is, it is late, and I haven't probably tested it hard
> enough, but need to hit the sack as it is work tomorrow.  I'd like a
> bit more time to bang on it before committing.  Keybinding in the
> client is very broken, so I think this maybe worth waiting on since
> it looks like I have something viable.
>
> I am positive it is an improvement even if it isn't perfect.  So far
> the Keybinding Modifier checkboxes all behave consistently unlike
> the current behavior.
>
> The new code changes the logic of the checkboxes on the keybinding
> dialog.  The existing logic is broken and somewhat inconsistent
> in implementation.  The new behavior is if no Run/Fire/Alt/Meta
> checkboxes are checked, KEYF_NORMAL is assumed (mode N). But,
> if all Run/Fire/Alt/Meta are checked, KEYF_MODIFIERS is assumed
> (mode A).
>
> So far it seems like it works.  I just have not tested interaction
> with the command-line bind/unbind extensively, but enough to think
> I can close #1527973.
>
> I'd also want to check to see if it fixes [ 1839894 ] Keybind editor
> flaw (gtk2, gtkv2)
>
> But even if not, it is a far sight better than the current code.
>
> Kevin

SVN revision 8247 does seem to have also fixed [ 1839894 ] Keybind editor
> flaw (gtk2, gtkv2).

I am marking it as pending.  Testing shows Update Binding does work after
all.  The caveat is that if you have multiple bindings to the same key,
things can behave somewhat unexpectedly if you aren't careful.  Say a key
is bound separately for normal, run, fire, meta, alt, then is bound A.
It will not be possible to get the last binding because one of the others
will trigger first.  In my case, I thought it wasn't working at first
because my keys file had conflicting entries.

Marking this as pending to close the issue technically speaking because the
description is no longer applicable in my opinion.  Whether the keybinding
code is too simple because it is possible to get into situations like this
is another issue, but, considering the keys file is manually editable,
whether it is worth doing any more work on it is debateable.

So, Mark, I believe I'm clear on work to go in before the update if the
libglade client is released instead of the branch revision.  I haven't put
my edits down on branch yet.  I'll check and fix branch - maybe tomorrow?

Gene, and others, feel free to test trunk keybinding and report back.

Kevin Bulgrien