What happened to wdired.el?

What happened to wdired.el?

[Dan Nicolaescu]

There was some discussion about wdired.el on this list in 2001.
(see
http://mail.gnu.org/archive/html/emacs-devel/2001-11/msg00342.html)

What happened to it? 
IMHO wdired.el is an excellent package, so it would be great to have
it included in Emacs. I have nothing to do with wdired.el, I am just
a happy user. 

Re: What happened to wdired.el?

[Stefan Monnier]

> There was some discussion about wdired.el on this list in 2001.

Installed a couple days ago,


        Stefan

Re: What happened to wdired.el?

[Dan Nicolaescu]

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

  > > There was some discussion about wdired.el on this list in 2001.
  > 
  > Installed a couple days ago,

Great! What a coincidence... 

Do you also plan to add a key binding / menu entry to enable wdired
from dired? 

Re: What happened to wdired.el?

[Juri Linkov]

Dan Nicolaescu <dann@godzilla.ics.uci.edu> writes:
> Great! What a coincidence... 
>
> Do you also plan to add a key binding / menu entry to enable wdired
> from dired? 

I see no reason not to add the binding suggested in the commentary
in wdired.el:

(define-key dired-mode-map "r" 'wdired-change-to-wdired-mode)

BTW, wdired deactivates [return] keys by an empty function.
Since dired doesn't works well with newlines in file names,
these keys could be bound to `wdired-finish-edit'.  While
using C-c C-c to finish wdired mode might be more correct,
typing RET looks more natural to finish renaming one file.

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

Re: What happened to wdired.el?

[Stefan Monnier]

> using C-c C-c to finish wdired mode might be more correct,
> typing RET looks more natural to finish renaming one file.

The few times I've tried wdired, I frequently needed up in undesirable
(tho temporarily desired) states, and hitting RET is something very natural
that can happen by mistake, so I feel like it's a bit too dangerous.

I agree it makes sense if you're only changing one file (which is what
dired-efap.el is for), but wdired does more than that.


        Stefan

Re: What happened to wdired.el?

[Stefan Monnier]

> Do you also plan to add a key binding / menu entry to enable wdired
> from dired? 

Why not check out the current state of the CVS archive?
There's already a menu entry for it.
I haven't added a keybinding yet because the "r" suggested by wdired.el
is rather inconvenient in my opinion.  I'd rather use C-x C-q but I haven't
done the work to see if/how that would work.


        Stefan

Re: What happened to wdired.el?

[Juri Linkov]

Stefan Monnier <monnier@iro.umontreal.ca> writes:
> I'd rather use C-x C-q but I haven't done the work to see if/how
> that would work.

I think C-x C-q makes sense for wdired mode.  And the same key could
be used to start and finish wdired mode.

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

Re: What happened to wdired.el?

[Miles Bader]

On Fri, Apr 30, 2004 at 03:47:28AM +0300, Juri Linkov wrote:
> I think C-x C-q makes sense for wdired mode.  And the same key could
> be used to start and finish wdired mode.

Instead of binding C-x C-q, can it be made to use conditionalized bindings
contingent on the `buffer-read-only' variable, as modes like diff-mode do?

That would make it work seamlessly when toggle-read-only has been rebound.

-Miles
-- 
"Though they may have different meanings, the cries of 'Yeeeee-haw!' and
 'Allahu akbar!' are, in spirit, not actually all that different."

Re: What happened to wdired.el?

[Stefan Monnier]

>> I think C-x C-q makes sense for wdired mode.  And the same key could
>> be used to start and finish wdired mode.

Since switching back to dired has irreversible side effects, I'm a bit
nervous about doing it that way.  It would work great tho if wdired was
changed so it doesn't perform the file operations but just records them for
dired so a dired-execute later actually performs them.

> Instead of binding C-x C-q, can it be made to use conditionalized bindings
> contingent on the `buffer-read-only' variable, as modes like diff-mode do?

I don't think we can do that with the current code: when switching to
wdired, we store the "current state" so that at the end we can see what has
changed and apply those changes.  Switching only bindings is not enough.
But if we change dired to constantly keep the "current state", then we
could do that.


        Stefan

Re: What happened to wdired.el?

[Juri Linkov]

Stefan Monnier <monnier@iro.umontreal.ca> writes:
>>> I think C-x C-q makes sense for wdired mode.  And the same key could
>>> be used to start and finish wdired mode.
>
> Since switching back to dired has irreversible side effects, I'm a bit
> nervous about doing it that way.  It would work great tho if wdired was
> changed so it doesn't perform the file operations but just records them for
> dired so a dired-execute later actually performs them.

The user can forget to call dired-execute later.  But what about
C-x C-q asking yes/no question about accepting the changes,
and an option to disable such a question?  There is already
a function `wdired-check-kill-buffer' asking a question,
so a similar function `wdired-check-finish-buffer' could be
added to finish wdired by C-x C-q safely.

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

Re: What happened to wdired.el?

[Richard Stallman]

    The user can forget to call dired-execute later.  But what about
    C-x C-q asking yes/no question about accepting the changes,
    and an option to disable such a question?

What are the changes in question, and why would they be lost?
Can we find a solution whereby they are not lost?

Re: What happened to wdired.el?

[Juri Linkov]

Richard Stallman <rms@gnu.org> writes:
>     The user can forget to call dired-execute later.  But what about
>     C-x C-q asking yes/no question about accepting the changes,
>     and an option to disable such a question?
>
> What are the changes in question, and why would they be lost?

wdired provides one user-level function `wdired-change-to-wdired-mode'
to start editing the file names and two functions to exit wdired-mode:
`wdired-finish-edit' and `wdired-abort-changes' with the former
committing the changes and the latter aborting them.  Both return
wdired to dired mode.

I proposed to add one user-level function which depending on the
user's answer to the question about accepting changes would call
either `wdired-finish-edit' or `wdired-abort-changes'.  And an option
which will allow to assume that the user always wants to accept the
changes by answering `yes' and calling `wdired-finish-edit'.

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

Re: What happened to wdired.el?

[Richard Stallman]

    wdired provides one user-level function `wdired-change-to-wdired-mode'
    to start editing the file names and two functions to exit wdired-mode:
    `wdired-finish-edit' and `wdired-abort-changes' with the former
    committing the changes and the latter aborting them.  Both return
    wdired to dired mode.

I think that is the right way to do it.
The Rmail Edit feature uses C-c C-c to finish the edit
and C-c C-] to abort.  It would be a good idea for wdired to
use the same bindings.  (Maybe it already does.)

Re: What happened to wdired.el?

[Stefan Monnier]

> I think that is the right way to do it.
> The Rmail Edit feature uses C-c C-c to finish the edit
> and C-c C-] to abort.  It would be a good idea for wdired to
> use the same bindings.  (Maybe it already does.)

It indeed already does.


        Stefan