9:24 p.m.
Hi there,
who can enlighten me on the meaning and significance of
(define-key global-map 'undo 'undo)
in xemacs/lisp/keydefs.el
please.
Do some keyboards have a key labelled undo?
I would like to make C-/ or C-_ show up in the pcl-cvs menu for undo.
Currently the menu reads (partially):
CVS
...
Commit C
Undo changes undo
Add a
I have looked for the undo key for months until I found C-/ and C-_ to
work also.
Can we do something about this?
--
Adrian Aichner
mailto:adrian＠xemacs.org
http://www.xemacs.org/
9:35 p.m.
Adrian Aichner writes:
Do some keyboards have a key labelled undo?
Yes, this key can be found on Sun keyboards (at least it could 5 years
ago when I last worked on a Sun...)
I have looked for the undo key for months until I found C-/ and C-_
to
work also.
Can we do something about this?
Please. The PCL-CVS key bindings really need some work. You used to
(IIRC) be able to hit "Enter" on a file line and open the file (a la
dired), now you have to click or press 'f'.
9:53 p.m.
>>>> "Charles" == Charles G Waldman
<Charles> writes:
Charles> Adrian Aichner writes:
> Do some keyboards have a key labelled undo?
Charles> Yes, this key can be found on Sun keyboards (at least it
Charles> could 5 years ago when I last worked on a Sun...)
Oh boy, I worked on Suns for a decade. Scarey how one forgets.
I can even remember the Find Cut, and, Paste keys on the left keypad
now -- or am I dreaming already?
> I have looked for the undo key for months until I found C-/ and
C-_ to
> work also.
>
> Can we do something about this?
Charles> Please. The PCL-CVS key bindings really need some work.
Charles> You used to (IIRC) be able to hit "Enter" on a file line
Charles> and open the file (a la dired), now you have to click or
Charles> press 'f'.
Here's my growing pcl-cvs key-binding grip list:
RET runs the command newline (should cvs-mode-find-file)
M-f runs the command cvs-mode-force-command (should be forward-word, bad boy!)
I'd like to move cvs-mode-force-command to !, how about that?
What should be done about the advertized undo key in the CVS menu,
advertise all bindings perhaps (that would be non-standard I guess)?
I guess
(where-is-internal 'undo)
([undo] [(control /)] [(control _)])
could be used to find out.
Any more gripes?
--
Adrian Aichner
mailto:adrian＠xemacs.org
http://www.xemacs.org/
9:57 p.m.
RET runs the command newline (should cvs-mode-find-file)
That's a bug in pcl-cvs-defs.el (someone recently pointed it out to me).
M-f runs the command cvs-mode-force-command (should be forward-word,
bad boy!)
I don't find horizontal cursor movement of any use in *cvs* buffers.
I'd like to move cvs-mode-force-command to !, how about that?
Sounds like a good idea, although it goes against the dired bindings.
Right now PCL-CVS does not follow dired, so maybe it doesn't matter,
but I've heard people request more dired-compatibility.
Stefan
10:35 p.m.
>>>> "Stefan" == Stefan Monnier
<monnier+lists/cvs/pcl(a)RUM.cs.yale.edu> writes:
> RET runs the command newline (should cvs-mode-find-file)
Stefan> That's a bug in pcl-cvs-defs.el (someone recently pointed
Stefan> it out to me).
ah, ok. cvs-mode-find-file might need to become a bit smarter to
handle the case where RET is hit on a line with no file.
Currently I get:
`nil' is not applicable to any of the selected files.
I don't mind it too much, but it doesn't really help a casual user who
has never applied nil to a selected file :-)
> M-f runs the command cvs-mode-force-command (should be
> forward-word, bad boy!)
Stefan> I don't find horizontal cursor movement of any use in
Stefan> *cvs* buffers.
principal of least surprise.
Even though I have a touchpad and wheely-mouse I use the keyboard a lot.
I expect M-b and M-f to do the same conceptual thing in all modes and
they do.
It took me a while to notice that M-f subtly changed cvs behavior.
> I'd like to move cvs-mode-force-command to !, how about that?
Stefan> Sounds like a good idea, although it goes against the
Stefan> dired bindings. Right now PCL-CVS does not follow dired,
Guess what M-f does in dired :-)
M-f runs the command forward-word
Stefan> so maybe it doesn't matter, but I've heard people request
Stefan> more dired-compatibility.
! runs the command dired-do-shell-command
Do you already have a cvs-do-...-command in mind for pcl-cvs?
If not, why not use ! for cvs-mode-force-command.
It has pretty good mmemonic value (force!)
Stefan> Stefan
What do you think about the undo keybinding shown in the CVS menu?
Later,
Adrian
--
Adrian Aichner
mailto:adrian＠xemacs.org
http://www.xemacs.org/
11:29 p.m.
The only commands I ever used in pcl-cvs are:
mark file
unmark file
context diff
kill line (or marked files)
revert to checked in version
commit
update
In my humble but correct opinion, those should all be on single
unmodified keys ("m", "u", "d", etc.)
Modes should **never** re-bind control or meta keys (unless they are
customizing motion.) E.g., C-f and M-f should always move the cursor
forward, and the only reason for a mode to rebind those is if they need
to hook into the motion action for some reason. (Though that's probably
better done with a hook anyway.)
Modes should only ever put their bindings under the C-c prefix, or, in
the case of non-editing modes, on keys that would normally have been
bound to self-insert.
Any other way lies madness.
--
Jamie Zawinski
jwz(a)jwz.org http://www.jwz.org/
jwz(a)dnalounge.com http://www.dnalounge.com/
7:14 a.m.
Adrian.Aichner(a)t-online.de (Adrian Aichner) writes:
>>>>> "Stefan" == Stefan Monnier
<monnier+lists/cvs/pcl(a)RUM.cs.yale.edu> writes:
>> RET runs the command newline (should cvs-mode-find-file)
Stefan> That's a bug in pcl-cvs-defs.el (someone recently pointed
Stefan> it out to me).
ah, ok. cvs-mode-find-file might need to become a bit smarter to
handle the case where RET is hit on a line with no file.
Currently I get:
`nil' is not applicable to any of the selected files.
I don't mind it too much, but it doesn't really help a casual user who
has never applied nil to a selected file :-)
Hitting RET on a (sub)directory could eventually expand/colapse the
directory. That could be a nice feature. :-)
Jarl
8:57 a.m.
>>>> "Stefan" == Stefan Monnier
<monnier+lists/cvs/pcl(a)RUM.cs.yale.edu> writes:
> What do you think about the undo keybinding shown in the CVS
menu?
Stefan> For me it's `U' so I don't see any problem with it ;-)
Aha, I can see (cvs annotate of emacs/lisp/pcvs-defs.el):
1.13 (monnier 11-May-01): ;; (undo . cvs-mode-undo)
1.1 (monnier 11-Mar-00): ("?" . cvs-help)
1.1 (monnier 11-Mar-00): ("h" . cvs-help)
Pending approval from XRB, it's outta here too.
Stefan> Stefan
--
Adrian Aichner
mailto:adrian＠xemacs.org
http://www.xemacs.org/
4:47 p.m.
>>>> "Charles" == Charles G Waldman
<cgw(a)alum.mit.edu> writes:
Charles> Adrian Aichner writes:
> Do some keyboards have a key labelled undo?
Charles> Yes, this key can be found on Sun keyboards (at least it could 5 years
Charles> ago when I last worked on a Sun...)
> I have looked for the undo key for months until I found C-/ and
C-_ to
> work also.
>
> Can we do something about this?
Charles> Please. The PCL-CVS key bindings really need some work. You used to
Charles> (IIRC) be able to hit "Enter" on a file line and open the file
(a la
Charles> dired), now you have to click or press 'f'.
And it seems to me that I used to be able to use '=' on all marked
files and it would produce a diff of all the marked files. Now it
just produces a diff on the file the cursor is on. The old behavior
was pretty handy when you wanted to create a selected patch set to
send to someone.
Ray
1:18 p.m.
Raymond Toy wrote:
:
: And it seems to me that I used to be able to use '=' on all marked
: files and it would produce a diff of all the marked files. Now it
: just produces a diff on the file the cursor is on. The old behavior
: was pretty handy when you wanted to create a selected patch set to
: send to someone.
1. M-x customize-group [RET] pcl-cvs [RET]
2. Uncollapse `Invert Ignore Marks'
3. Uncheck `diff'
4. State -> Save for Future Sessions
It's all.
--
Andrew W. Nosenko (awn(a)bcs.zp.ua)
10:34 p.m.
>>>> "Andrew" == Andrew W Nosenko
<awn(a)bcs.zp.ua> writes:
Andrew> Raymond Toy wrote:
Andrew> :
Andrew> : And it seems to me that I used to be able to use '=' on all
marked
Andrew> : files and it would produce a diff of all the marked files. Now it
Andrew> : just produces a diff on the file the cursor is on. The old behavior
Andrew> : was pretty handy when you wanted to create a selected patch set to
Andrew> : send to someone.
Andrew> 1. M-x customize-group [RET] pcl-cvs [RET]
Andrew> 2. Uncollapse `Invert Ignore Marks'
Andrew> 3. Uncheck `diff'
Andrew> 4. State -> Save for Future Sessions
Andrew> It's all.
Thank you! I didn't know the default had changed.
Ray
8126
days inactive
8135
days old
11 comments
8 participants
participants (8)
-
Adrian Aichner -
Adrian Aichner -
Andrew W. Nosenko -
Charles Waldman -
Jamie Zawinski -
Jarl Friis -
Raymond Toy -
Stefan Monnier