bug#34765: 26.1; with-temp-buffer should not run buffer-list-update-hook

[Eli Zaretskii <eliz@gnu.org>]

> From: martin rudalics <rudalics@gmx.at>
> Cc: Eli Zaretskii <eliz@gnu.org>, 34765@debbugs.gnu.org, alexanderm@web.de,
>  monnier@IRO.UMontreal.CA
> Date: Tue, 21 May 2019 12:04:46 +0200
> 
> I have not managed to understand the Vcode_conversion_workbuf_name
> vs Vcode_conversion_reused_workbuf rigmarole yet.

The latter is a fixed buffer, created once and never killed.  So the
buffer hooks never run for it, and never affect Emacs.  By contrast,
the former is a buffer created when the reused buffer is busy and
cannot be reused.  We then kill Vcode_conversion_workbuf_name when we
no longer need it.  Thus, buffer hooks run for these work buffers all
the time, and for a code-conversion intensive code they could slow
down Emacs, specially if the list of buffer hooks is long.

> > The possibilities for the buffer creation subroutine are either to act
> > specially on certain buffer name prefixes, or to accept an extra
> > argument indicating what to do, no?  Are there any others?  There was
> > mention of exposing a buffer-local variable to Elisp, but IIRC setting
> > that after creating the buffer would already be too late.
> 
> So far there is no extra argument, the entire analysis is based on
> examining the proposed name argument.

Since we currently keep this special treatment limited to a small
number of buffer names, I'm not sure there's a need to expose this
facility to Lisp.

> > Buffer names starting with spaces are already special in some contexts,
> > so extending that idea for inhibiting buffer hooks doesn't sound too
> > bad,
> 
> Eli thinks that "this is too drastic a measure".

Yes.  No one said buffer hooks must _never_ run for temporary buffers.
There could be legitimate Lisp programs which want those hooks to run
in that case.