bug-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#34656: 27.0.50; Directional window swap states


From: Juri Linkov
Subject: bug#34656: 27.0.50; Directional window swap states
Date: Wed, 27 Feb 2019 23:12:16 +0200
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/27.0.50 (x86_64-pc-linux-gnu)

> +  (let ((other-window (window-in-direction dir nil nil nil
> +                                           windmove-wrap-around t)))
>
> If 'windmove-wrap-around' is nil, this may swap the selected window's
> state with that of the minibuffer window.  Or am I misinterpreting?

Thanks for the review.  Now I fixed the minibuffer case and closed this request.

Before this fix, trying to swap states with the minibuffer failed with the 
error:

  (error "Window #<window 4 on  *Minibuf-0*> too small to accommodate state")

Now after the fix, it doesn't consider the minibuffer as a candidate to swap 
states.

BTW, before proposing the implementation in windmove, the first idea was
to improve the logic of finding the second window in window-swap-states like:

diff --git a/lisp/window.el b/lisp/window.el
index 80828bb35c..3961ec3975 100644
--- a/lisp/window.el
+++ b/lisp/window.el
@@ -6018,7 +6018,10 @@ window-swap-states
   (if window-2
       (unless (window-live-p window-2)
         (error "%s is not a live window" window-2))
-    (setq window-2 (next-window window-1 'nomini 'visible)))
+    (setq window-2 (or (get-mru-window 'visible t t)
+                       (get-mru-window 0 t t)
+                       (get-mru-window t t t)
+                       (next-window window-1 'nomini 'visible))))
   (unless (eq window-1 window-2)
     (let* ((height (memq size '(t height)))
            (width (memq size '(t width)))

This logic is like finding the second window in
`compare-windows-get-recent-window'.

But since directional swapping in windmove is more convenient,
it seems the above change is not needed.





reply via email to

[Prev in Thread] Current Thread [Next in Thread]