Re: [STUMP] group-format-map and window-format-map

[Ben Spencer]

On Wed, Mar 16, 2011 at 10:10:00PM -0700, Matt Spear wrote:
> Fixed the 0 index not existing.  Fixed the group map.  Removed the
> print.  Made the #\n default to the X-map-number (and fixed doc).

Thanks, pushed.


> It would be nice to integrate this further, but there are some
> subtelities.  The gselect command allows selecting a group via a
> substring, and in the general case the group-number-map map will be
> non-digits (as I have it now). Maybe the gselect command will default
> to the map index if it is a single character and exists in the
> group-number-map, otherwise will attempt a substring search on the
> groups?

I'd rather not assume that the map index is a single character - in
fact, if we were to change CHAR to ELT in group-map-number it could be
anything, eg:

(setf *group-number-map* #("foo" "bar" "baz"))

or perhaps more usefully:

(setf *group-number-map* #(1 2 3 4 5 6 7 8 9 10))


What if we just check against group-map-number before looking for a
partial match?  If I have groups like this:

a*Default
b+Float
c-Blah
d-Test

And I hit 'd' in gselect, I would probably want the 'Test' group (I'd
use completion if I wanted Default).

Patch attached.  It's probably less efficient than the previous
version but I doubt anybody has enough groups for it to matter.


Ben

0001-Use-group-number-map-in-select-group.patch
Description: Text Data