Re: inhibit-message variant that doesn't even log to *Messages*

[Eli Zaretskii <eliz@gnu.org>]

On January 31, 2019 10:08:29 AM GMT+02:00, Gergely Risko <gergely@risko.hu> wrote:
> Hi,
> 
> I use more and more packages that provide some functionality by doing
> something periodically.
> 
> A recent example is sync-recentf:
> https://github.com/ffevotte/sync-recentf
> 
> When sync-recentf decides to periodically save the recent file, a lot
> of
> messages get written with the `message' function.
> 
> I worked around most if this with:
> 
>     (defadvice recentf-cleanup (around nce/recentf-cleanup activate)
>       (let ((inhibit-message t))
>         ad-do-it))
> 
> My problem is that, these messages still get logged to *Messages*, so
> if
> I set the frequency to every 15 seconds, and try to debug something
> unrelated in Elisp, then these messages get to be annoying a bit.
> 
> Would there be any downside to providing an additional variable (under
> the name `inhibit-message-fundamentally' or such) that would inhibit
> message printing altogether (echo area AND *Messages*)?
> 
> Would there be willingness to merge this if I were to work on this?
> 
> flet, letf, cl-letf and the like will not work for me, because some of
> the messages I'm interested in temporarily disabling are coming from C
> code.
> 
> Cheers,
> Gergely

Is there any problem in binding message-log-max to a nil value?