Proper behavior of C-u 3 C-v

Proper behavior of C-u 3 C-v

[rgrant]

Is this supposed to scroll down by three lines, or is it supposed to 
scroll down by three screen-fulls?

Re: Proper behavior of C-u 3 C-v

[Eli Zaretskii]

> Date: Sat, 15 Aug 2020 01:50:48 +0200
> From: rgrant@posteo.net
> 
> Is this supposed to scroll down by three lines, or is it supposed to 
> scroll down by three screen-fulls?

The former.  The doc string says it clearly:

  Scroll text of selected window upward ARG lines

Re: Proper behavior of C-u 3 C-v

[Amin Bandali]

[-- Attachment #1: Type: text/plain, Size: 681 bytes --]

Eli Zaretskii writes:

>> Date: Sat, 15 Aug 2020 01:50:48 +0200
>> From: rgrant@posteo.net
>> 
>> Is this supposed to scroll down by three lines, or is it supposed to 
>> scroll down by three screen-fulls?
>
> The former.  The doc string says it clearly:
>
>   Scroll text of selected window upward ARG lines

The doc string is indeed clear, but I personally wouldn't have expected
that behaviour.  I'm curious if there's a particular reason why C-v
deviates from the convention of other similar movement commands, where
specifying a numerical ARG `n' would have the effect of performing that
command's action `n' times (e.g. M-3 M-f to move forward three words).

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 857 bytes --]

Re: Proper behavior of C-u 3 C-v

[Emanuel Berg via Users list for the GNU Emacs text editor]

Amin Bandali wrote:

> The doc string is indeed clear, but I personally
> wouldn't have expected that behaviour. I'm curious
> if there's a particular reason why C-v deviates
> from the convention of other similar movement
> commands, where specifying a numerical ARG `n'
> would have the effect of performing that command's
> action `n' times (e.g. M-3 M-f to move forward
> three words).

Another inconsistency is

  (scroll-down 1)
  (scroll-up   1)

where one expects the byte-compiler to say, use
(scroll 1) and (scroll -1) instead, but it doesn't,
and for good reason because there isn't
a `scroll', even.

And another inconsistency is:

  (scroll-other-window-down 1)
  (scroll-other-window      1)

compared to scroll-up/scroll-down, where -up
is explicit.

But actually, I think that looks better and what
I see nothing stops you from using
(scroll-other-window -X) instead of
(scroll-other-window-down X). It is also more clear
since people have different ideas what up and down,
but just "scroll" everyone can agree means "read on".

BTW civilized scrolling, one step at a time,
including horizontally:

  https://dataswamp.org/~incal/emacs-init/scroll.el

-- 
underground experts united
http://user.it.uu.se/~embe8573
https://dataswamp.org/~incal

Re: Proper behavior of C-u 3 C-v

[Eli Zaretskii]

> From: Amin Bandali <bandali@gnu.org>
> Date: Sun, 16 Aug 2020 00:15:53 -0400
> 
> >   Scroll text of selected window upward ARG lines
> 
> The doc string is indeed clear, but I personally wouldn't have expected
> that behaviour.  I'm curious if there's a particular reason why C-v
> deviates from the convention of other similar movement commands, where
> specifying a numerical ARG `n' would have the effect of performing that
> command's action `n' times (e.g. M-3 M-f to move forward three words).

Because what we do is more useful, I guess.  I find this so useful
that I have a special command to scroll by 1 line bound to a key since
time immemoriam.

But anyway, this behavior is so old that the person to ask is probably
RMS himself.