bug#33331: 26; doc strings of `tabify' and `untabify'

bug#33331: 26; doc strings of `tabify' and `untabify'

[Drew Adams]

[-- Attachment #1: Type: text/plain, Size: 398 bytes --]

Please say why there is an unused optional arg ARG.

 

In GNU Emacs 26.1 (build 1, x86_64-w64-mingw32)

of 2018-05-30

Repository revision: 07f8f9bc5a51f5aa94eb099f3e15fbe0c20ea1ea

Windowing system distributor `Microsoft Corp.', version 10.0.16299

Configured using:

`configure --without-dbus --host=x86_64-w64-mingw32

--without-compress-install 'CFLAGS=-O2 -static -g3''

 

[-- Attachment #2: Type: text/html, Size: 2887 bytes --]

bug#33331: 26; doc strings of `tabify' and `untabify'

[Eli Zaretskii]

> Date: Fri, 9 Nov 2018 21:48:40 -0800 (PST)
> From: Drew Adams <drew.adams@oracle.com>
> 
> Please say why there is an unused optional arg ARG.

??? The below is from the doc strings of both functions:

  If called interactively with prefix ARG, convert for the entire
  buffer.

What am I missing?

bug#33331: 26; doc strings of `tabify' and `untabify'

[Drew Adams]

> > Please say why there is an unused optional arg ARG.
> 
> ??? The below is from the doc strings of both functions:
> 
>   If called interactively with prefix ARG, convert for the entire
>   buffer.
> 
> What am I missing?

Sorry, I missed that somehow.  Might have
been tripped up by the use of `_arg'; dunno.
I'll close the bug.

That said...

AFAICT there's no need to add that optional
arg.  This would have been sufficient, and
it is what I would have expected:

(defun untabify (start end)
  "..."
  (interactive
    (if current-prefix-arg
        (list (point-min) (point-max))
      (list (region-beginning) (region-end))))
  ...)

Am I missing something?  Does adding that arg,
and using it only in `interactive', help?

Seems like when someone added the prefix-arg
behavior (Emacs 24, apparently) they went
overboard a bit.  Not a problem; just a bit
weird, to my eyes.

bug#33331: 26; doc strings of `tabify' and `untabify'

[Eli Zaretskii]

> Date: Sat, 10 Nov 2018 09:08:11 -0800 (PST)
> From: Drew Adams <drew.adams@oracle.com>
> Cc: 33331@debbugs.gnu.org
> 
> AFAICT there's no need to add that optional
> arg.  This would have been sufficient, and
> it is what I would have expected:
> 
> (defun untabify (start end)
>   "..."
>   (interactive
>     (if current-prefix-arg
>         (list (point-min) (point-max))
>       (list (region-beginning) (region-end))))
>   ...)
> 
> Am I missing something?  Does adding that arg,
> and using it only in `interactive', help?

If we are going to change those functions, I'd prefer instead to make
a change that used the 3rd argument in non-interactive calls.

bug#33331: 26; doc strings of `tabify' and `untabify'

[Drew Adams]

> > AFAICT there's no need to add that optional
> > arg.  This would have been sufficient, and
> > it is what I would have expected:
> >
> > (defun untabify (start end)
> >   "..."
> >   (interactive
> >     (if current-prefix-arg
> >         (list (point-min) (point-max))
> >       (list (region-beginning) (region-end))))
> >   ...)
> >
> > Am I missing something?  Does adding that arg,
> > and using it only in `interactive', help?
> 
> If we are going to change those functions, I'd prefer instead to make
> a change that used the 3rd argument in non-interactive calls.

I was going to say, "Sure, go for it".  But what would
that mean in this case?  Non-interactively you already
supply the START and END.  What would be gained by
a Boolean argument that uses `point-min' and `point-max'
and ignores the values of START and END?

Doesn't make sense to me, but maybe I'm missing something.
Just what did you have in mind here?

bug#33331: 26; doc strings of `tabify' and `untabify'

[Eli Zaretskii]

> Date: Sat, 10 Nov 2018 11:36:19 -0800 (PST)
> From: Drew Adams <drew.adams@oracle.com>
> Cc: 33331@debbugs.gnu.org
> 
> > If we are going to change those functions, I'd prefer instead to make
> > a change that used the 3rd argument in non-interactive calls.
> 
> I was going to say, "Sure, go for it".  But what would
> that mean in this case?  Non-interactively you already
> supply the START and END.  What would be gained by
> a Boolean argument that uses `point-min' and `point-max'
> and ignores the values of START and END?

Just consistency with the interactive use.  Not a big deal, obviously,
but then neither is the fact ARG goes on unused.