bug#81441: [PATCH] Have 'toggle-rot13-mode' behave buffer locally

Ping!  How should we make further progress in this matter?

> Cc: [email protected]
> Date: Tue, 21 Jul 2026 14:01:20 +0300
> From: Eli Zaretskii <[email protected]>
> 
> > From: Philip Kaludercic <[email protected]>
> > Cc: [email protected]
> > Date: Mon, 20 Jul 2026 19:10:29 +0000
> > 
> > Eli Zaretskii <[email protected]> writes:
> > 
> > >> From: Philip Kaludercic <[email protected]>
> > >> Date: Sun, 19 Jul 2026 21:23:25 +0000
> > >> 
> > >> I was recently surprised that this command changes the display table of
> > >> the window, which if confusing if you switch between buffers or move
> > >> them around between windows.  I would suggest just changing the
> > >> behavior outright, without a replacement as the function remains simple
> > >> enough to write oneself if a per-window workflow is that important
> > >> (which is doubtful considering that we are talking about rot13 here).
> > >
> > > This has a clear downside: using a buffer display table means you
> > > cannot have this buffer shown in another window as human-readable
> > > text.
> > 
> > Do you think/know that that is a common use-case?
> 
> I don't know for sure, but that's my strong impression both from the
> documentation and from what the command does.
> 
> Also note that this is email-related feature, and it is unusual to
> have the email reading buffer be displayed in more than a single
> window.  At least when using Rmail, there's usually just one window
> which shows the message.
> 
> > > So I don't think we should make such a change, except as an opt-in
> > > behavior.  The documentation of this feature clearly hints that the
> > > rot13 display is ephemeral, so IMO window-specific display table is a
> > > better default.
> > 
> > I see your argument, but I don't think that either is strictly
> > preferable.  Would it make sense to then instead add a proper buffer
> > local minor mode `rot13-buffer-mode' as an alternative to existing
> > command?
> 
> I don't see how it could be an alternative, because a mode by
> definition affects the buffer in any window.  We could have such a
> mode _in_addition_ to the existing command, though.
> 
> 
> 
>