| Re: I for Interactive Programming? (was: widget_control and group_leader) [message #23087] |
Fri, 05 January 2001 08:42 |
Vapuser
Messages: 63
Registered: November 1998
|
Member |
|
|
"Richard G. French" <rfrench@wellesley.edu> writes:
<snip>
> Finally taking the plunge and trying to learn this system. I'm
> using a PC with xterm software on it to connect to my unix
> workstation. Trying to figure out how to use the Alt-TAB command
> inside emacs- Alt-TAB is a windows thing that lets you hop between
> running programs. I tried to use ESC-TAB but it does not seem to do
> the right completion. Instead, I get
>
> Use symbol help mode:
>
> and I am not sure what to do next. I've tried to find a way
> to bind a different key to be my ALT key but I haven't succeeded.
> Any emacs users out there who can help me out here? As should be
> obvious, I have not used emacs before.
>
> Thanks,
> Dick
Let me see if I understand. 'Alt-Tab' is a *windows* command which
has become disabled when the active window is the xterm window in
which you're running emacs and you want the windows functionality
back? Is that the problem? If this is not a correct rephrasing of
your problem, what are you expecting from the ALT-TAB key sequence?
That sequence isn't bound to anything in my (20.4.2) version of
emacs.
I'm going to proceed on the assumption that what you want is the
window's functionality back.
What happens when you make some other window the active window? Does
the ALT-TAB functionality return?
Which version of emacs? And what xserver software are you
running. My guess is that it's the xserver software that's
intercepting the ALT key because that key is significant to many
xservers (it's how the system menus on many X windows are accessed
from the keyboard) For instance, on my system, an SGI, alt-F9 will
minimize the current active window. It's been years since I've run
any such software, but I remember it allowing you to set the
function of various control keys as part of a specific connection
configuration. If I've got your question right, and what you want is
the Windows functionality back, you really want the Windows GUI to
intercept the 'Alt-Tab' so that it never gets sent through the
xserver software to emacs *at all*. Perhaps 'unsetting' it in the
xserver configuration???
Be careful what you wish for, however. There might be some serious
side effects from taking the ALT key away from emacs, not to mention
the xserver. I've never seen any xserver software on windows (or
macs, for that matter) that didn't have some
... ummm... idiosyncracies. If all you're missing is the ability to
ALT-TAB your way around the active Win9x windows, I'd consider that
a fair price.
But that's just me.
William Daffer
--
William Daffer: 818-354-0161: William.Daffer@jpl.nasa.gov
|
|
|
|
| Re: I for Interactive Programming? (was: widget_control and group_leader) [message #23089 is a reply to message #23087] |
Fri, 05 January 2001 07:54  |
Vapuser
Messages: 63
Registered: November 1998
|
Member |
|
|
JD Smith <jdsmith@astro.cornell.edu> writes:
<snip>
> Hear Hear! I've got something even better (from Carsten, of course).
> I'm not sure if this has been put into IDLWAVE yet as a more palatable
> form (I vote for it), but if you have version >4.7b (meaning when 4.8
> comes out, if you're leary of alpha's) you can simply use in your
> .emacs:
>
> (add-hook 'idlwave-shell-mode-hook
> (lambda ()
> (set (make-local-variable 'comint-input-ring-file-name)
> "~/.idlwhist")
> (comint-read-input-ring)))
>
> (add-hook 'idlwave-shell-sentinel-hook 'comint-write-input-ring)
>
> I.e. just set up a "comint input ring filename" to ~/.idlwhist (or
> whatever you like), and ensures it is read in when the shell starts (in
> the idlwave-shell-mode-hook), and written out after it exits (in the
> idlwave-shell-sentinel-hook).
>
> This way, you always get all of your input history (which is actually
> maintained by comint, not IDL), between sessions, etc. Now restarting
> IDL is not nearly so painful. The commands are still right there, at
> your fingertips.
>
> Never wonder where your command history went again. And for the really
> brave, try:
>
> (setq comint-input-ring-size 1024)
>
> outside of your idlwave-shell-mode-hook (since comint runs before
> idlwave-shell). Now you'll have the silly command you tried 3 weeks
> ago, somewhere in there. By the way, try [C-c C-l] to see your entire
> history in another buffer, with ability to scroll through it and click
> on the command you're after (or cut and paste between many such
> commands). .idlwhist is a plain file, which you can search through
> using grep or a pager or however you like, if you don't want to load
> IDL.
>
> Good luck,
>
> JD
Man! I have to put you on my Christmas Card list!
Thanks!
whd
--
William Daffer: 818-354-0161: William.Daffer@jpl.nasa.gov
|
|
|
|
| Re: I for Interactive Programming? (was: widget_control and group_leader) [message #23100 is a reply to message #23089] |
Thu, 04 January 2001 20:00 |
Richard French
Messages: 173
Registered: December 2000
|
Senior Member |
|
|
JD Smith wrote:
> Hear Hear! I've got something even better (from Carsten, of course).
> I'm not sure if this has been put into IDLWAVE yet as a more palatable
> form (I vote for it), but if you have version >4.7b (meaning when 4.8
> comes out, if you're leary of alpha's) you can simply use in your
> .emacs:
>
> (add-hook 'idlwave-shell-mode-hook
> (lambda ()
> (set (make-local-variable 'comint-input-ring-file-name)
> "~/.idlwhist")
> (comint-read-input-ring)))
>
> (add-hook 'idlwave-shell-sentinel-hook 'comint-write-input-ring)
>
Finally taking the plunge and trying to learn this system.
I'm using a PC with xterm software on it to connect to my
unix workstation. Trying to figure out how to use the Alt-TAB
command inside emacs- Alt-TAB is a windows thing that lets
you hop between running programs. I tried to use ESC-TAB
but it does not seem to do the right completion. Instead, I
get
Use symbol help mode:
and I am not sure what to do next. I've tried to find a way
to bind a different key to be my ALT key but I haven't succeeded.
Any emacs users out there who can help me out here? As should be
obvious, I have not used emacs before.
Thanks,
Dick
|
|
|
|
| Re: I for Interactive Programming? (was: widget_control and group_leader) [message #23105 is a reply to message #23100] |
Thu, 04 January 2001 16:22 |
John-David T. Smith
Messages: 384
Registered: January 2000
|
Senior Member |
|
|
Vapuser wrote:
>
> For those willing to learn emacs and Carsten Dominik's marvelous
> idlwave_shell mode, you would have no more problems with saving the
> results of your 'tappling' at the keyboard. Just start up an instance
> of idlwave_shell in emacs and tapple to your hearts delight, then
> write the idlwave-shell buffer out to a file. Or, kill/yank
> (i.e. cut/paste) it to another buffer, edit it and write that out to a
> file.
>
> This is almost the *only* way I develop code anymore, I try stuff
> out in emacs/idlwave-shell and then cut/paste it to the 'work in
> progress' file. (doesn't work for widgets programming, of course, but
> in this area it's the debugging capabilities, which are totally
> awesome *d00d* that recommend it)
>
> No more need to remember to turn journalling on or to set
> variables to assure that you save enough of your work.
>
> It does have a steep learning curve, especially if you're not used to
> emacs or you're on a Windows/Mac system. If that's the case,
> journalling is probably the best bet. But if you're moderately used to
> emacs and on a unix system, emacs/idlwave-shell mode simply can't be
> beat (IMHO) and it's well worth the effort!
>
> Kudo's to Carsten for his fine work!
Hear Hear! I've got something even better (from Carsten, of course).
I'm not sure if this has been put into IDLWAVE yet as a more palatable
form (I vote for it), but if you have version >4.7b (meaning when 4.8
comes out, if you're leary of alpha's) you can simply use in your
.emacs:
(add-hook 'idlwave-shell-mode-hook
(lambda ()
(set (make-local-variable 'comint-input-ring-file-name)
"~/.idlwhist")
(comint-read-input-ring)))
(add-hook 'idlwave-shell-sentinel-hook 'comint-write-input-ring)
I.e. just set up a "comint input ring filename" to ~/.idlwhist (or
whatever you like), and ensures it is read in when the shell starts (in
the idlwave-shell-mode-hook), and written out after it exits (in the
idlwave-shell-sentinel-hook).
This way, you always get all of your input history (which is actually
maintained by comint, not IDL), between sessions, etc. Now restarting
IDL is not nearly so painful. The commands are still right there, at
your fingertips.
Never wonder where your command history went again. And for the really
brave, try:
(setq comint-input-ring-size 1024)
outside of your idlwave-shell-mode-hook (since comint runs before
idlwave-shell). Now you'll have the silly command you tried 3 weeks
ago, somewhere in there. By the way, try [C-c C-l] to see your entire
history in another buffer, with ability to scroll through it and click
on the command you're after (or cut and paste between many such
commands). .idlwhist is a plain file, which you can search through
using grep or a pager or however you like, if you don't want to load
IDL.
Good luck,
JD
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
