GNU bug report logs - #66394
29.1; Make register-read-with-preview more useful

Previous Next

Package: emacs;

Reported by: Thierry Volpiatto <thievol <at> posteo.net>

Date: Sat, 7 Oct 2023 19:07:01 UTC

Severity: normal

Found in version 29.1

Done: Eli Zaretskii <eliz <at> gnu.org>

Bug is archived. No further changes may be made.

Full log


Message #174 received at 66394 <at> debbugs.gnu.org (full text, mbox):

From: Eshel Yaron <me <at> eshelyaron.com>
To: Thierry Volpiatto <thievol <at> posteo.net>
Cc: jonas <at> bernoul.li, Michael Heerdegen <michael_heerdegen <at> web.de>,
 Bastien Guerry <bzg <at> gnu.org>, stefankangas <at> gmail.com, 66394 <at> debbugs.gnu.org,
 monnier <at> iro.umontreal.ca
Subject: Re: bug#66394: 29.1; Make register-read-with-preview more useful
Date: Sat, 02 Dec 2023 14:02:52 +0100
Thierry Volpiatto <thievol <at> posteo.net> writes:

> Thierry Volpiatto <thievol <at> posteo.net> writes:
>
>> Hi Bastien and Eshel,
>>
>> I see, you want to go fast.

Thanks, but this not only about speed.
Using a minibuffer to read a key has other disadvantages: it limits
the range of keys that can be easily used as registers since now
specifying the `C-a` register is much harder (as Michael mentioned).
This also makes register commands less convenient or and even
impossible (when `enable-recursive-minibuffers` is nil) to use inside
an existing minibuffer.

>> Using read-key+minibuffer, as I said would be a pain to implement,
>> hopefully (IIUC) we can use only minibuffer and exit immediately and it
>> is simple to implement (3 lines):
>>
>> 1) Set register-use-preview to nil.

Hmm, but this disables the automatic register preview, which I still want
to see, preferably after a short delay, just like before.

>> 2) Apply this patch on register.el:
>> ...
>>
>> Let me know if this is acceptable for you.
>
> Even better is this patch which ask for confirmation when overwriting a
> register but not when jumping, inserting or setting a new register:
>
> ...

That's improves things a bit, but see above.  I can see how confirmation
could be helpful for some people, but not everyone would appreciate the
added friction.  I basically suggest reverting to the previous behavior
(or otherwise restoring full compatibility), and making any involvement
of the minibuffer in reading registers optional, for example by
providing a new user option `register-read-from-minibuffer`.


Best,

Eshel




This bug report was last modified 1 year and 210 days ago.

Previous Next


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