C-x C-f RET change

C-x C-f RET change

[Florian Weimer]

In previous versions, C-x C-f RET reread the current buffer from the
disk.  In Emacs 22, it shows the the current directory using dired.
While the new behavior is certainly more intuitive for beginners, the
old behavior is needed more often by more advanced users, especially
if they are not yet so advanced that they do everything from within a
single Emacs process.

At the very least, the NEWS entry

| ** C-x C-f RET, typing nothing in the minibuffer, is no longer a special case.
| 
| Since the default input is the current directory, this has the effect
| of specifying the current directory.  Normally that means to visit the
| directory with Dired.

should mention that the new key sequence is C-x C-f M-n RET.

Re: C-x C-f RET change

[Reiner Steib]

On Wed, Nov 09 2005, Florian Weimer wrote:

> | ** C-x C-f RET, typing nothing in the minibuffer, is no longer a special case.
> | 
> | Since the default input is the current directory, this has the effect
> | of specifying the current directory.  Normally that means to visit the
> | directory with Dired.
>
> should mention that the new key sequence is C-x C-f M-n RET.

Or `C-x C-v RET'.

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: C-x C-f RET change

[Florian Weimer]

* Reiner Steib:

> On Wed, Nov 09 2005, Florian Weimer wrote:
>
>> | ** C-x C-f RET, typing nothing in the minibuffer, is no longer a special case.
>> | 
>> | Since the default input is the current directory, this has the effect
>> | of specifying the current directory.  Normally that means to visit the
>> | directory with Dired.
>>
>> should mention that the new key sequence is C-x C-f M-n RET.
>
> Or `C-x C-v RET'.

This command always destroys undo information (without prompting), and
resets the major mode, so it's slightly different.

Re: C-x C-f RET change

[Andreas Schwab]

Florian Weimer <fw@deneb.enyo.de> writes:

> * Reiner Steib:
>
>> On Wed, Nov 09 2005, Florian Weimer wrote:
>>
>>> | ** C-x C-f RET, typing nothing in the minibuffer, is no longer a special case.
>>> | 
>>> | Since the default input is the current directory, this has the effect
>>> | of specifying the current directory.  Normally that means to visit the
>>> | directory with Dired.
>>>
>>> should mention that the new key sequence is C-x C-f M-n RET.
>>
>> Or `C-x C-v RET'.
>
> This command always destroys undo information (without prompting), and
> resets the major mode, so it's slightly different.

The best substitute is revert-buffer, but that is not bound by default
(except in dired).

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

RE: C-x C-f RET change

[Drew Adams]

    In previous versions, C-x C-f RET reread the current buffer from the
    disk.  In Emacs 22, it shows the the current directory using dired.
    While the new behavior is certainly more intuitive for beginners, the
    old behavior is needed more often by more advanced users, especially
    if they are not yet so advanced that they do everything from within a
    single Emacs process.

    At the very least, the NEWS entry

    | ** C-x C-f RET, typing nothing in the minibuffer, is no
    longer a special case.
    |
    | Since the default input is the current directory, this has the effect
    | of specifying the current directory.  Normally that means to visit the
    | directory with Dired.

    should mention that the new key sequence is C-x C-f M-n RET.

I too prefer the old behavior, especially for `C-x 4 f'.

Re: C-x C-f RET change

[David Kastrup]

"Drew Adams" <drew.adams@oracle.com> writes:

>     In previous versions, C-x C-f RET reread the current buffer from the
>     disk.  In Emacs 22, it shows the the current directory using dired.
>     While the new behavior is certainly more intuitive for beginners, the
>     old behavior is needed more often by more advanced users, especially
>     if they are not yet so advanced that they do everything from within a
>     single Emacs process.
>
>     At the very least, the NEWS entry
>
>     | ** C-x C-f RET, typing nothing in the minibuffer, is no
>     longer a special case.
>     |
>     | Since the default input is the current directory, this has the effect
>     | of specifying the current directory.  Normally that means to visit the
>     | directory with Dired.
>
>     should mention that the new key sequence is C-x C-f M-n RET.
>
> I too prefer the old behavior, especially for `C-x 4 f'.

Uh, why?  C-x 2 is easier.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

RE: C-x C-f RET change

[Drew Adams]

    > I too prefer the old behavior, especially for `C-x 4 f'.

    Uh, why?  C-x 2 is easier.

I do use `C-x 2' to split a window - which I do only rarely.

I use pop-up-frames = t. Each buffer is in its own frame, by default. So, I
use `C-x 4 f' to open the same file in another frame (cloning the frame, in
effect). I do that, for example, to examine different parts of the same file
side by side.

With the new behavior, I must enter the file name (though I can use M-n to
get it, so this is no biggee).

BTW - I didn't mean to suggest, by my post, that Emacs should necessarily
return to the old behavior; I meant only to report one (no doubt atypical)
user's preference. I should have mentioned that explicitly, along with
mentioning pop-up-frames = t. Thanks for prompting me to be clearer.

Re: C-x C-f RET change

[Andreas Schwab]

"Drew Adams" <drew.adams@oracle.com> writes:

> I use pop-up-frames = t. Each buffer is in its own frame, by default. So, I
> use `C-x 4 f' to open the same file in another frame (cloning the frame, in
> effect).

What's wrong with `C-x 5 2'?

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

RE: C-x C-f RET change

[Drew Adams]

    > I use pop-up-frames = t. Each buffer is in its own frame, by
    > default. So, I use `C-x 4 f' to open the same file in another
    > frame (cloning the frame, in effect).

    What's wrong with `C-x 5 2'?

Nothing.  Good point.

I guess I never think of the `C-x 5' prefix, because I use pop-up-frames =
t. `C-x 4' generally does the same thing as `C-x 5' in that case, except for
a few exceptions like this (there is no `C-x 4 2').  I'm just not in the
habit of using the `C-x 5' commands.

I also use a redefinition of `delete-window' (`C-x 0') that deletes a
one-window frame, so I end up just using the window commands to operate on
frames.  I forget about the difference, in general, and about the
frame-specific commands, in particular.  (As I said, "atypical".)

Re: C-x C-f RET change

[David Kastrup]

"Drew Adams" <drew.adams@oracle.com> writes:

>     > I use pop-up-frames = t. Each buffer is in its own frame, by
>     > default. So, I use `C-x 4 f' to open the same file in another
>     > frame (cloning the frame, in effect).
>
>     What's wrong with `C-x 5 2'?
>
> Nothing.  Good point.
>
> I guess I never think of the `C-x 5' prefix, because I use pop-up-frames =
> t. `C-x 4' generally does the same thing as `C-x 5' in that case, except for
> a few exceptions like this (there is no `C-x 4 2').

There is C-x 2 though.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

RE: C-x C-f RET change

[Drew Adams]

    >     > I use pop-up-frames = t. Each buffer is in its own frame, by
    >     > default. So, I use `C-x 4 f' to open the same file in another
    >     > frame (cloning the frame, in effect).
    >
    >     What's wrong with `C-x 5 2'?
    >
    > Nothing.  Good point.
    >
    > I guess I never think of the `C-x 5' prefix, because I use
    > pop-up-frames = t. `C-x 4' generally does the same thing as
    > `C-x 5' in that case, except for a few exceptions like this
    > (there is no `C-x 4 2').

    There is C-x 2 though.

I guess I miss your point.  Even with pop-up-frames = t, `C-x 2' does not
open the file in a separate frame (and it shouldn't).

Re: C-x C-f RET change

[David Kastrup]

"Drew Adams" <drew.adams@oracle.com> writes:

>     >     > I use pop-up-frames = t. Each buffer is in its own frame, by
>     >     > default. So, I use `C-x 4 f' to open the same file in another
>     >     > frame (cloning the frame, in effect).
>     >
>     >     What's wrong with `C-x 5 2'?
>     >
>     > Nothing.  Good point.
>     >
>     > I guess I never think of the `C-x 5' prefix, because I use
>     > pop-up-frames = t. `C-x 4' generally does the same thing as
>     > `C-x 5' in that case, except for a few exceptions like this
>     > (there is no `C-x 4 2').
>
>     There is C-x 2 though.
>
> I guess I miss your point.  Even with pop-up-frames = t, `C-x 2' does not
> open the file in a separate frame (and it shouldn't).

Then we probably should have a separate C-x 4 2 for symmetry that
obeys the setting of pop-up-frames.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

RE: C-x C-f RET change

[Drew Adams]

    >     >     > I use pop-up-frames = t. Each buffer is in its
    own frame, by
    >     >     > default. So, I use `C-x 4 f' to open the same
    file in another
    >     >     > frame (cloning the frame, in effect).
    >     >
    >     >     What's wrong with `C-x 5 2'?
    >     >
    >     > Nothing.  Good point.
    >     >
    >     > I guess I never think of the `C-x 5' prefix, because I use
    >     > pop-up-frames = t. `C-x 4' generally does the same thing as
    >     > `C-x 5' in that case, except for a few exceptions like this
    >     > (there is no `C-x 4 2').
    >
    >     There is C-x 2 though.
    >
    > I guess I miss your point.  Even with pop-up-frames = t,
    > `C-x 2' does not open the file in a separate frame (and it
    > shouldn't).

    Then we probably should have a separate C-x 4 2 for symmetry that
    obeys the setting of pop-up-frames.

I don't know about that, but we could. It wouldn't hurt.

It might be somewhat confusing when pop-up-frames = nil. In that case, `C-x
4' would (I imagine) do the same thing as `C-x 2'. (I don't mind.)

`C-x 2' and `C-x 5 2' are not strictly parallel, because window commands
have two prefixes: `C-x' for "this" window and `C-x 4' for "other" window.
Frame commands do not have two prefixes - they have only `C-x 5'.

`C-x 2' says to split _this_ window in two; it doesn't say to do anything
about another window, according to the conventional interpretation. It could
alternatively be thought of, however, as `make-another-window' instead of
`split-window', in which case it would make sense as `C-x 4 2'.

So, it might be worth creating a separate `make-another-window' (or
`make-window-command', in analogy to `make-frame-command'), bound to `C-x 4
2'. With pop-up-frames = nil, this would do the same thing as
`split-window'. With pop-up-frames = t, this would do the same thing as
`make-frame-command'. That would keep the conventions and terminology
consistent.

In any case, `C-x 4 2' is by no means an important requirement.

Re: C-x C-f RET change

[Juri Linkov]

> `C-x 2' says to split _this_ window in two; it doesn't say to do anything
> about another window, according to the conventional interpretation. It could
> alternatively be thought of, however, as `make-another-window' instead of
> `split-window', in which case it would make sense as `C-x 4 2'.
>
> So, it might be worth creating a separate `make-another-window' (or
> `make-window-command', in analogy to `make-frame-command'), bound to `C-x 4
> 2'. With pop-up-frames = nil, this would do the same thing as
> `split-window'. With pop-up-frames = t, this would do the same thing as
> `make-frame-command'. That would keep the conventions and terminology
> consistent.

Since `C-x 2' is not the exact equivalent of `C-x 4 f M-n RET'
(the difference is where point lands after the command: in the first case
it is in the initial window, in the second case it is in a new window).
The same difference makes sense for `C-x 2' and new `make-window-command':
`make-window-command' with pop-up-frames=nil would leave point in a new
window, like `make-frame-command' leaves point in a new frame.

-- 
Juri Linkov
http://www.jurta.org/emacs/

RE: C-x C-f RET change

[Drew Adams]

    > `C-x 2' says to split _this_ window in two; it doesn't say to 
    > do anything
    > about another window, according to the conventional 
    > interpretation. It could
    > alternatively be thought of, however, as 
    > `make-another-window' instead of
    > `split-window', in which case it would make sense as `C-x 4 2'.
    >
    > So, it might be worth creating a separate `make-another-window' (or
    > `make-window-command', in analogy to `make-frame-command'), 
    > bound to `C-x 4
    > 2'. With pop-up-frames = nil, this would do the same thing as
    > `split-window'. With pop-up-frames = t, this would do the 
    > same thing as
    > `make-frame-command'. That would keep the conventions and terminology
    > consistent.
    
    Since `C-x 2' is not the exact equivalent of `C-x 4 f M-n RET'
    (the difference is where point lands after the command: in the 
    first case
    it is in the initial window, in the second case it is in a new window).
    The same difference makes sense for `C-x 2' and new 
    `make-window-command':
    `make-window-command' with pop-up-frames=nil would leave point in a new
    window, like `make-frame-command' leaves point in a new frame.
    
Yes. Good.

Re: C-x C-f RET change

[Eli Zaretskii]

> From: "Drew Adams" <drew.adams@oracle.com>
> Date: Wed, 9 Nov 2005 07:29:04 -0800
> 
> I too prefer the old behavior, especially for `C-x 4 f'.

It was infinitely more confusing for lots of users.  Even I sometimes
fell into that trap.

Re: C-x C-f RET change

[Edward O'Connor]

Florian Weimer wrote:

> In previous versions, C-x C-f RET reread the current buffer from the
> disk.  In Emacs 22, it shows the the current directory using dired.
> While the new behavior is certainly more intuitive for beginners, the
> old behavior is needed more often by more advanced users, especially
> if they are not yet so advanced that they do everything from within a
> single Emacs process.

As a "more advanced user", I can honestly say that I never cared for the
old behavior, and infinitely prefer the newer behavior. In fact, I'm
still in the habit of typing C-x C-f DEL RET precisely because the old
behavior got in the way of what I intended.

Actually, until your email I never even knew what the old behavior was,
exactly. It seemed like C-x C-f RET was a no-op.

I've always used M-x revert-buffer RET when wanting Emacs to reread the
buffer from disk.


Ted

-- 
Edward O'Connor
hober0@gmail.com

Ense petit placidam sub libertate quietem.

Re: C-x C-f RET change

[Richard M. Stallman]

    should mention that the new key sequence is C-x C-f M-n RET.

Ok, thanks.