Closed DamienCassou closed 6 years ago
I have the impression that aw-dispatch-alist is just an half-backed hydra. What about rewriting that in terms of an hydra?
The packages do share these ideas, I often use ideas in one package to prototype the other.
But aw-dispatch-alist
is a bit different: it's an interface that takes (lambda (window) ...)
, while defhydra
takes (lambda () (interactive) ...)
.
Thanks to the interface, the dispatch works similarly on windows to how vi does it with text objects. So it's rather more a half-baked evil.el rather than hydra.el.
But it could also be useful to add ace-window-hydra
with similar functionality. The reason why it's not there by default is that people (myself included) don't like gratuitous dependencies.
For references see how ivy-hydra.el
(a separate package) rebinds M-o in ivy from ivy-dispatching-done
to ivy-dispatching-done-hydra
.
Another reference is hydra-lispy-x
from lispy.el: the ? binding gradually shows more help.
I have the impression that
aw-dispatch-alist
is just an half-backed hydra. What about rewriting that in terms of an hydra?