GNU bug report logs - #46469
27.1; `isearch-del-char' should move point further back

Previous Next

Full log

Message #23 received at 46469 <at> debbugs.gnu.org (full text, mbox):

From: Augusto Stoffel <arstoffel <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 46469 <at> debbugs.gnu.org
Subject: Re: bug#46469: 27.1; `isearch-del-char' should move point further back
Date: Sat, 13 Feb 2021 10:53:44 +0100

On Sat, 13 Feb 2021 at 10:59, Eli Zaretskii <eliz <at> gnu.org> wrote:

> If we are talking about personal preferences, then I suggest to make
> your desired behavior an opt-in one by introducing a defcustom.  We
> cannot change the behavior of isearch-del-char unconditionally if the
> existing behavior is not a bug, but just something some users may not
> like.

In this case, the distinction between being buggy or just weird and not
very useful is a bit blurry.  Let me note, however (assuming here that
DEL is bound to ‘isearch-del-char’):

1) The patched `isearch-del-char' serves the purpose of undoing
something you typed by mistake.  The result of ‘C-s y DEL x’ is
equivalent to the result of ‘C-s x’.  Just like, in normal editing,
typing ‘x DEL y’ is equivalent to typing ‘y’.

2) The current `isearch-del-char' serves a different purpose altogether.
‘C-s y DEL x’ is equivalent to ‘C-s y RET C-s x’, i.e., you are
searching for an ‘x’ occurring after an ‘y’.

As to this being just a personal preference: I'm sure you'll find people
attached to the current behavior if you poll the mailing list, but let
me note that, in other sensible search UIs, DEL works like the patched
‘isearch-del-char’.  I've just tested Swiper, gedit and zile.  They're
all like the patched version.  I see no reason for a defcustom here.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.