bug#71927: 29.4; ibuffer-do-isearch and ibuffer-do-isearch-regexp not prompting for input

[Eli Zaretskii <eliz@gnu.org>]

> From: Stephen Berman <stephen.berman@gmx.net>
> Cc: juri@linkov.net,  me@eshelyaron.com,  kickingvegas@gmail.com,
>   71927@debbugs.gnu.org,  basil@contovou.net,  jpw@gnu.org
> Date: Thu, 04 Jul 2024 20:29:05 +0200
> 
> On Thu, 04 Jul 2024 21:07:36 +0300 Eli Zaretskii <eliz@gnu.org> wrote:
> 
> >> From: Stephen Berman <stephen.berman@gmx.net>
> >> Cc: Eli Zaretskii <eliz@gnu.org>,  Eshel Yaron <me@eshelyaron.com>,
> >>   kickingvegas@gmail.com,  71927@debbugs.gnu.org,  basil@contovou.net,
> >>   jpw@gnu.org
> >> Date: Thu, 04 Jul 2024 19:36:34 +0200
> >>
> >> On Thu, 04 Jul 2024 19:04:42 +0300 Juri Linkov <juri@linkov.net> wrote:
> >>
> >> >>>> FWIW, AFAICT everything is working correctly, it's just that the
> >> >>>> "Operation finished" message hides the prompt.  ibuffer-do-isearch
> >> >>>> should tell define-ibuffer-op not to display that message, somehow.
> >> >>>
> >> >>> I don't see how this could be considered "correct": the "Operation
> >> >>> finished" message is supposed to be shown only after the Isearch is
> >> >>> finished in all the marked buffer, not before.  It looks like we need
> >> >>> a function that will not return until all the buffers where searched,
> >> >>> because that's what define-ibuffer-op expects.  Don't you agree?
> >> >
> >> > It intentionally uses 'no-recursive-edit' set to t, so ibuffer-do-isearch
> >> > correctly exits immediately while leaving isearch-mode enabled.
> >> >
> >> >> The attached patch appears to DTRT, but I only tested it briefly.
> >> >> ...
> >> >>  (define-ibuffer-op ibuffer-do-isearch ()
> >> >>    "Perform a `isearch-forward' in marked buffers."
> >> >>    (:interactive ()
> >> >> -   :opstring "searched in"
> >> >> +   :no-opstring t
> >> >
> >> > Thanks for the patch.  I confirm this is the right thing to do.
> >> > Maybe instead of :no-opstring would be better to use some special value
> >> > like :opstring 'no?  But I'm not sure if this is better than :no-opstring.
> >>
> >> Suppressing the message when :opstring has the value 'no is fine with
> >> me.  If Eli is willing to accept this approach, I can go ahead and
> >> commit it (to master, presumably, since this is a longstanding issue).
> >
> > I already said this didn't sound the right solution here, and I
> > explained why.  I'd be interested in hearing counter-arguments, if
> > there are any.
> 
> I gave a mild counterargument upthread, that making
> ibuffer-do-isearch{-regexp} defuns independent of define-ibuffer-op
> seems like accepting the inadequacy of the latter instead of trying to
> improve it.

It is indeed inadequate for commands that just put Emacs in a special
state and return, as opposed to commands that don't return before they
did the complete job of operating on the marked buffers.

> Also, I am not familiar enough with the ibuffer code to be
> confident that I could implement I could implement the functionality
> without using this macro, but someone else might be in a better position
> to do that.

Fair enough.