Re: simplifying windmove-frame-edges

[Eli Zaretskii]

> From: Juanma Barranquero <address@hidden>
> Date: Fri, 25 Oct 2019 14:17:23 +0200
> Cc: martin rudalics <address@hidden>, Emacs developers <address@hidden>
> 
> > We cannot just delete them, we need to obsolete them first.  They were
> > with us for far too long.
> 
> I find the argument less than compelling, given the fact that the functions 
> have been known not to work well in
> some cases for the past six years or so.

>From bitter experience, we have no good ways of knowing which APIs are
used by unbundled packages and user code out there.  Emacs is a stable
package, so users rightfully expect it not to break their code by
backward-incompatible changes.

> --- a/lisp/windmove.el
> +++ b/lisp/windmove.el
> @@ -473,15 +473,19 @@ windmove-other-window-loc
>  ;; Rewritten on 2013-12-13 using `window-in-direction'.  After the
>  ;; pixelwise change the old approach didn't work any more.  martin
>  (defun windmove-find-other-window (dir &optional arg window)
> -  "Return the window object in direction DIR.
> -DIR, ARG, and WINDOW are handled as by `windmove-other-window-loc'."
> +  "Return the window object in direction DIR as seen from WINDOW.
> +DIR is one of `left', `up', `right', or `down'.
> +Optional ARG, if negative, means to use the right or bottom edge of
> +WINDOW as reference position, instead of `window-point'; if positive,
> +use the left or top edge of WINDOW as reference point.
> +WINDOW must be a live window and defaults to the selected one."
>    (window-in-direction dir window nil arg windmove-wrap-around t))

The last sentence should be the 2nd, because you mention WINDOW in the
first sentence of the doc string.

> From 1e7b2fe06a4fce4dc2ce52b037145190d86176ca Mon Sep 17 00:00:00 2001
> From: Juanma Barranquero <address@hidden>
> Date: Fri, 25 Oct 2019 14:02:22 +0200
> Subject: [PATCH 3/3] windmove.el: Remove comments about the old implementation

Are these comments really no longer relevant?

Thanks.