bug#74637: [PATCH] Make view-read-only behave like view-file

[Björn Bidar]

Visuwesh <visuweshm@gmail.com> writes:

> [செவ்வாய் டிசம்பர் 03, 2024] Eli Zaretskii wrote:
>
>>> From: Björn Bidar <bjorn.bidar@thaodan.de>
>>> Cc: 74637@debbugs.gnu.org
>>> Date: Tue, 03 Dec 2024 12:06:11 +0200
>>> 
>>> >> > That's an incompatible behavior change.  Is that justified?  How can
>>> >> > we be sure that everyone agrees with your interpretation of this mode?
>>> >> 
>>> >> All other view-file like modes behave like this, you view the file and
>>> >> leave the file with q.
>>> >
>>> > That doesn't change the fact that view-mode didn't behave like that,
>>> > until now.
>>> 
>>> View-mode isn't change outside of the specific situation of opening a
>>> file which isn't writable.
>>
>> It's still a significant change.  Killing a buffer is not a minor
>> think, and restoring it is not always easy, or even possible (e.g.,
>> the file could have been deleted in the meantime).
>>
>> My opinion is that if we install this, we need to provide some way of
>> getting the previous behavior back, for those who may want it.
>>
>> Stefan and Andrea, WDYT?
>
> I am neither Stefan nor Andrea: there have been a number of instances
> where Emacs not eagerly killing buffers has saved accidentally or
> prematurely deleted files on my end.  I would be opposed to a change
> that kills the buffer on exiting view-mode entered due to
> view-read-only=t to avoid potential data loss.

But does apply to instance of view-read-only?  Again it only affects
files which are not writable, i.e. those that could not be deleted by
the user in the first place.

The change is for not writable files plus view-read-only, NOT files which
are visited and then view-mode activated through `view-mode` or
view-read-only through read-only-mode.