bug#73986: 31.0.50; Allow suppressing message passed to set-transient-map

[Stephen Berman via "Bug reports for GNU Emacs, the Swiss army knife of text editors" <bug-gnu-emacs@gnu.org>]

On Thu, 24 Oct 2024 20:12:20 +0300 Juri Linkov <juri@linkov.net> wrote:

>> If a caller of set-transient-map passes a suitable format string, this
>> is displayed as a message on activating the transient map.  An example
>> is text-scale-adjust, which displays the message "Use +, =, -, 0, C-+,
>> C-=, C--, C-0 for further adjustment".  So if after typing `C-x C-+' you
>> immediately type `+ + + - - -', the text size increases three times and
>> then decreases three times.  But on each key press the message is
>> redisplayed, which is slighty annoying (it actually looks like it stays
>> in the echo area but there is a brief flicker on each key press); the
>> message is helpful the first time, if you're unfamiliar with the
>> command, but on repeated uses, it becomes visual noise.  So I propose a
>> user option to suppress the message, as in the attached patch.
>
> Instead of adding a myriad of user options for every command
> that displays a message, we decided to create a single point
> of customization, so you could just set these options:
>
>   (add-to-list 'set-message-functions 'inhibit-message)
>   (add-to-list 'inhibit-message-regexps "for further adjustment")

Ah, thanks, I now vaguely recall seeing these but forgot about them.
They indeed do work just as well as my proposed change to
set-transient-map for text-scale-adjust as well as two private functions
I use that pass a message string matching the same regexp, so that
satisfies my immediate wish.  Still, there could be any number of
callers of set-transient-map that pass message strings users may want to
suppress but that require a different regexp, in which case the burden
on the user is more than with my patch.  But since that's AFAIK
currently only a theoretical possibility, I won't pursue the proposal.
I'll wait a few days in case someone does chime in favor of my patch,
but otherwise I'll then close the bug.

Steve Berman