Re: [PATCH] Allow inhibiting 'auto-save-visited-mode' on a per-buffer basis.

[Philipp Stephani]

Am Do., 2. Apr. 2020 um 11:59 Uhr schrieb Štěpán Němec <address@hidden>:
>
> On Thu, 02 Apr 2020 11:47:09 +0200
> Robert Pluim wrote:
>
> >>>>>> On Thu, 2 Apr 2020 11:40:45 +0200, Philipp Stephani <address@hidden> 
> >>>>>> said:
> >
> >     Philipp> Am Do., 2. Apr. 2020 um 11:25 Uhr schrieb Robert Pluim 
> > <address@hidden>:
> >     >>
> >     >> >>>>> On Wed,  1 Apr 2020 19:57:36 +0200, Philipp Stephani 
> > <address@hidden> said:
> >     >>
> >     Philipp> * lisp/files.el (auto-save-visited-inhibit): New buffer-local
> >     Philipp> variable.
> >     Philipp> (auto-save-visited-mode): Use it.
> >     >>
> >     >> Could you expand on why this is needed? What's stopping you from just
> >     >> turning off auto-save-visited-mode in those buffers, either directly
> >     >> or via local variables?
> >
> >     Philipp> auto-save-visited-mode is a global mode, it can't be turned 
> > off for
> >     Philipp> individual buffers.
> >
> > Youʼre right, I saw the 'define-minor-mode' bit and missed that itʼs a
> > global minor mode.
>
> Still, wouldn't it be better to add the rationale to the commit message?
> IMO there should always be some commentary or explanation, unless it's
> just an obvious or trivial fix. I often find myself wondering "why was
> this change made?" when browsing the commit history, and I'm sure I'm
> not the only one.

Sure, I'll add something like the following:
At least for me, auto-save-visited-mode is very slow and blocks user
interaction for files visited over TRAMP. Therefore, I'd like a
mechanism to disable auto-save-visited-mode for some buffers (namely,
those visiting remote files).