bug#12570: 24.2.50; C-u does not work in read-passwd

bug#12570: 24.2.50; C-u does not work in read-passwd

[Michael Welsh Duggan]

When read-passwd was changed in revision 107874
(monnier@iro.umontreal.ca-20120411181320-zd3wwgbom1229q74) to use
read-string, it no longer allows C-u to reset the currently entered
string.

In GNU Emacs 24.2.50.1 (i686-pc-linux-gnu, GTK+ Version 2.10.4)
 of 2012-09-25 on waterbuck.yellow.cert.org
Bzr revision: 110196 eliz@gnu.org-20120925124413-us1fkny5urer83c3
Windowing system distributor `The X.Org Foundation', version 11.0.70101000
System Description:	Red Hat Enterprise Linux Server release 5.8 (Tikanga)

-- 
Michael Welsh Duggan
(mwd@cert.org)

bug#12570: 24.2.50; C-u does not work in read-passwd

[Stefan Monnier]

> When read-passwd was changed in revision 107874
> (monnier@iro.umontreal.ca-20120411181320-zd3wwgbom1229q74) to use
> read-string, it no longer allows C-u to reset the currently entered
> string.

You can use C-a C-k, instead, like in any other minibuffer.


        Stefan

bug#12570: 24.2.50; C-u does not work in read-passwd

[Michael Welsh Duggan]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> When read-passwd was changed in revision 107874
>> (monnier@iro.umontreal.ca-20120411181320-zd3wwgbom1229q74) to use
>> read-string, it no longer allows C-u to reset the currently entered
>> string.
>
> You can use C-a C-k, instead, like in any other minibuffer.

If you truly think that this is TRT, so be it.  I will adjust
(eventually).  But it does sit uneasily in my mind.  I can type C-u to
cancel typing in passwords for login shells, ssh-add, and just about any
other program that accepts a password.  It feels odd to switch my
password typing habits just in Emacs.  When I fumble-finger a password,
I barely pause in typing before typing C-u and continuing to type.  The
change is consistent with the rest of Emacs, but I do not know that I
can declare it a wise consistency in this case.

-- 
Michael Welsh Duggan
(mwd@cert.org)

bug#12570: 24.2.50; C-u does not work in read-passwd

[Richard Stallman]

I agree with you.  I think consistency with other programs that read
passwords is more important than consistency with the minibuffer.

-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use Ekiga or an ordinary phone call

bug#12570: 24.2.50; C-u does not work in read-passwd

[Andrew W. Nosenko]

On Thu, Oct 4, 2012 at 11:19 PM, Richard Stallman <rms@gnu.org> wrote:
> I agree with you.  I think consistency with other programs that read
> passwords is more important than consistency with the minibuffer.
>

Excuse me, but proposed "consistence" is not a consistence with other
password-reading programs but a consistence with teminals or teminal
emulators.

For example, for me the most used password entries are GDM (login),
gnome-screensaver (session lock) and pinentry for GnuPG and SSH.  All
of them use GTK+, and "reset previous typed passwd" is Shift-Home
followed by Del or Backspace.

Should I propose Emacs to behave the same?..
Rhetorical question.

-- 
Andrew W. Nosenko <andrew.w.nosenko@gmail.com>

bug#12570: 24.2.50; C-u does not work in read-passwd

[Richard Stallman]

    Excuse me, but proposed "consistence" is not a consistence with other
    password-reading programs but a consistence with teminals or teminal
    emulators.

I normally use terminals.

Perhaps read-passwd should implement C-u on terminals
and Shift-HOME DEL on a window system?

-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use Ekiga or an ordinary phone call

bug#12570: 24.2.50; C-u does not work in read-passwd

[Andrew W. Nosenko]

On Fri, Oct 5, 2012 at 4:35 AM, Richard Stallman <rms@gnu.org> wrote:
>     Excuse me, but proposed "consistence" is not a consistence with other
>     password-reading programs but a consistence with teminals or teminal
>     emulators.
>
> I normally use terminals.

Then, for consistency with terminals, you need not C-u, but what the
"kill" control-character is set to (see KILL control character in
termios(4) and "kill" control character in stty(1)).  Yes, it set to
^U (C-u in Emacs notation) by default, but may be redefined by user at
the any time using stty(1), for example.

>
> Perhaps read-passwd should implement C-u on terminals
> and Shift-HOME DEL on a window system?

Oh, no, please!  I hate CUA and prefer Emacs way!

-- 
Andrew W. Nosenko <andrew.w.nosenko@gmail.com>

bug#12570: 24.2.50; C-u does not work in read-passwd

[Stefan Monnier]

> If you truly think that this is TRT, so be it.  I will adjust
> (eventually).  But it does sit uneasily in my mind.  I can type C-u to
> cancel typing in passwords for login shells, ssh-add, and just about any
> other program that accepts a password.  It feels odd to switch my
> password typing habits just in Emacs.  When I fumble-finger a password,

We can add a C-u binding as well.  I don't have time right now, tho, so
if someone else wants to do it...


        Stefan

bug#12570: 24.2.50; C-u does not work in read-passwd

[Glenn Morris]

Stefan Monnier wrote:

> We can add a C-u binding as well.  I don't have time right now, tho, so
> if someone else wants to do it...

Done; hopefully to your satisfaction.