REPL

REPL

[Elvin Peterson]

How do I create a REPL for emacs for elisp?

TIA.

Re: REPL

[Maciek Pasternacki]

On Boomtime, The Aftermath 40, 3170 YOLD, Elvin Peterson wrote:

> How do I create a REPL for emacs for elisp?

Just use lisp-interaction-mode, there you can use C-j to evaluate form
before the cursor and result is inserted into the buffer.  The initial
*scratch* buffer is started in this mode.

-- 
__    Maciek Pasternacki <maciekp@japhy.fnord.org> [ http://japhy.fnord.org/ ]
`| _   |_\  / { Hac in hora sine mora corde pulsum tangite;
,|{-}|}| }\/         quod per sortem sternit fortem, mecum omnes plangite!   }
\/   |____/                                           ( Carmina Burana )  -><-

Re: REPL

[Marco Gidde]

Elvin Peterson <elvin_peterson@yahoo.com> writes:

> How do I create a REPL for emacs for elisp?

You could try the command ielm (Interactive Emacs Lisp Mode). It's
more like the 'classical' REPL you might know from other Scheme or
Common Lisp implementations.


-- 
Marco Gidde

Re: REPL

[Elvin Peterson]

Marco Gidde wrote:
> Elvin Peterson <elvin_peterson@yahoo.com> writes:
> 
> 
>>How do I create a REPL for emacs for elisp?
> 
> 
> You could try the command ielm (Interactive Emacs Lisp Mode). It's
> more like the 'classical' REPL you might know from other Scheme or
> Common Lisp implementations.
> 
> 

Thanks.  I get the error -- command cat not found when I try to run it. 
  IIRC, the REPL code for COMMON LISP is just a couple of lines, so I 
was hoping someone would post similar stuff for emacs.

Thanks.

Re: REPL

[Elvin Peterson]

Maciek Pasternacki wrote:
> On Boomtime, The Aftermath 40, 3170 YOLD, Elvin Peterson wrote:
> 
> 
>>How do I create a REPL for emacs for elisp?
> 
> 
> Just use lisp-interaction-mode, there you can use C-j to evaluate form
> before the cursor and result is inserted into the buffer.  The initial
> *scratch* buffer is started in this mode.
> 

That is quite painful to use, especially since I don't get the command 
recall and options to access outputs of previous commands.  Something 
like SLIME would be really nice.

Thanks.

Re: REPL

[Pascal Bourguignon]

Elvin Peterson <elvin_peterson@yahoo.com> writes:

> Maciek Pasternacki wrote:
> > On Boomtime, The Aftermath 40, 3170 YOLD, Elvin Peterson wrote:
> >
> >>How do I create a REPL for emacs for elisp?
> > Just use lisp-interaction-mode, there you can use C-j to evaluate
> > form
> > before the cursor and result is inserted into the buffer.  The initial
> > *scratch* buffer is started in this mode.
> >
> 
> That is quite painful to use, especially since I don't get the command
> recall and options to access outputs of previous commands.  Something
> like SLIME would be really nice.

Yes, some work hard on stuff like ilisp or slime to try to be able to
use other lisp like emacs-lisp, and others are not happy with
emacs-lisp REPL and want the harder lisp REPL...

If you want to get the answer when typing RET instead of C-u C-x C-e, use:

(defun newline-and-eval ()
  (interactive)
  (insert "\n")
  (eval-last-sexp t))

(local-set-key "C-j" 'newline-and-eval)

(and instead of typing RET when you want to insert a new line, type C-q RET)

Or you could use:

(local-set-key [C-return] 'newline-and-eval)

to differenciate a normal new-line (RET) from one when you want to
evaluate (C-return).

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
The world will now reboot; don't bother saving your artefacts.

Re: REPL

[Marco Gidde]

Elvin Peterson <elvin_peterson@yahoo.com> writes:

>> You could try the command ielm (Interactive Emacs Lisp Mode). It's
>> more like the 'classical' REPL you might know from other Scheme or
>> Common Lisp implementations.
>>
>
> Thanks.  I get the error -- command cat not found when I try to run
> it. IIRC, the REPL code for COMMON LISP is just a couple of lines, so
> I was hoping someone would post similar stuff for emacs.

Sorry, I wasn't thinking of Windows NT. But as I saw in another thread
you're just solving your cygwin problems :-)

Re: REPL

[Marco Gidde]

Pascal Bourguignon <spam@mouse-potato.com> writes:

> Yes, some work hard on stuff like ilisp or slime to try to be able to
> use other lisp like emacs-lisp, and others are not happy with
> emacs-lisp REPL and want the harder lisp REPL...

I think both 'modes', the usual emacs/ilisp/slime stuff and the REPL,
have their strong points. The first is most useful for programming,
the second for testing and trying out small snippets of code. If there
were no REPL I would really miss *, ** and *** :-)

-- 
Marco Gidde

Re: REPL

[Kai Grossjohann]

Elvin Peterson <elvin_peterson@yahoo.com> writes:

> Thanks.  I get the error -- command cat not found when I try to run it. 

I think you can tell it to run hexl instead.  This might have been
fixed in the CVS version of Emacs.  So you could search the ChangeLog
file for the relevant entry and patch your version of Emacs.

Kai

Re: REPL

[Elvin Peterson]

Pascal Bourguignon wrote:
> Elvin Peterson <elvin_peterson@yahoo.com> writes:
> 
> 
>>Maciek Pasternacki wrote:
>>
>>>On Boomtime, The Aftermath 40, 3170 YOLD, Elvin Peterson wrote:
>>>
>>>
>>>>How do I create a REPL for emacs for elisp?
>>>
>>>Just use lisp-interaction-mode, there you can use C-j to evaluate
>>>form
>>>before the cursor and result is inserted into the buffer.  The initial
>>>*scratch* buffer is started in this mode.
>>>
>>
>>That is quite painful to use, especially since I don't get the command
>>recall and options to access outputs of previous commands.  Something
>>like SLIME would be really nice.
> 
> 
> Yes, some work hard on stuff like ilisp or slime to try to be able to
> use other lisp like emacs-lisp, and others are not happy with
> emacs-lisp REPL and want the harder lisp REPL...

I was thinking more in terms of teaching new users rather than 
application development.  Downloading, installing, and configuring a 
standalone lisp seems more involved than just running emacs.

Thanks for the reply.

Re: REPL

[Alan Mackenzie]

Elvin Peterson <elvin_peterson@yahoo.com> wrote on Sun, 28 Nov 2004
00:25:18 +0530:
> How do I create a REPL for emacs for elisp?

A _what_?

A Regular Expression Parsing Language?
A Redundant Elisp Parenthesis Lopper?
A Railway Engine Passing Loop?

???

> TIA.

-- 
Alan Mackenzie (Munich, Germany)
Email: aacm@muuc.dee; to decode, wherever there is a repeated letter
(like "aa"), remove half of them (leaving, say, "a").

Re: REPL

[Joe Corneli]

   Elvin Peterson <elvin_peterson@yahoo.com> wrote on Sun, 28 Nov 2004
   00:25:18 +0530:
   > How do I create a REPL for emacs for elisp?

   A _what_?

GIYF.

http://www.cliki.net/REPL

   A Redundant Elisp Parenthesis Lopper?

Basically correct.

Re: REPL

[Friedrich Dominicus]

Elvin Peterson <elvin_peterson@yahoo.com> writes:

> How do I create a REPL for emacs for elisp?
use M-x ielm

Regards
Friedrich

-- 
Please remove just-for-news- to reply via e-mail.

Re: REPL

[Oliver Scholz]

[Since M-x ielm has already been pointed to by others ...]

Elvin Peterson <elvin_peterson@yahoo.com> writes:
[...]
> IIRC, the REPL code for COMMON LISP is just a couple of lines, so I
> was hoping someone would post similar stuff for emacs.

Sure, to get a very basic REPL in Emacs is simple:

;; Because we use `loop' for the looks of it:
(require 'cl)

(let* ((buffer (generate-new-buffer "*tmp*"))
       (standard-output buffer))
  (switch-to-buffer buffer)
  ;; The REPL:
  (loop (print (eval (read)))))

The extra code in ielm is to get something a tiny little bit more
comfortable.

    Oliver
-- 
14 Frimaire an 213 de la Révolution
Liberté, Egalité, Fraternité!

Re: REPL

[Pascal Bourguignon]

Oliver Scholz <alkibiades@gmx.de> writes:

> [Since M-x ielm has already been pointed to by others ...]
> 
> Elvin Peterson <elvin_peterson@yahoo.com> writes:
> [...]
> > IIRC, the REPL code for COMMON LISP is just a couple of lines, so I
> > was hoping someone would post similar stuff for emacs.
> 
> Sure, to get a very basic REPL in Emacs is simple:
> 
> ;; Because we use `loop' for the looks of it:
> (require 'cl)
> 
> (let* ((buffer (generate-new-buffer "*tmp*"))
>        (standard-output buffer))
>   (switch-to-buffer buffer)
>   ;; The REPL:
>   (loop (print (eval (read)))))
> 
> The extra code in ielm is to get something a tiny little bit more
> comfortable.

Good idea. It could be useful to handle errors, history variables and
termination:


(defvar +++ nil)
(defvar ++  nil)
(defvar +   nil)
(defvar /// nil)
(defvar //  nil)
(defvar /   nil)
(defvar *** nil)
(defvar **  nil)
(defvar *   nil)
(defvar -   nil)

(defun repl ()
  (interactive)
  ;; I prefer to do it in the current buffer.
  (let ((standard-output  (current-buffer)))
    (switch-to-buffer standard-output)
    (block nil
      (loop for hist from 1 do
            (setf +++ ++   
                  ++  + 
                  +   -
                  -   (read))
            (insert (format  "[%d]> %S\n" hist -))
            (when (or (member - '((quit)(exit)(continue))))
              (insert "Good bye.\n")
              (return))
            (setf /// //   
                  //  /
                  /   (list (condition-case G61912 (eval -) (error G61912)))
                  *** **
                  **  *
                  *   (first /))
            (insert (format (format "%%%ds %%S\n\n" 
                              (length (format "[%d]>" hist))) "-->" *))))))

M-x repl RET

[1]> (dotimes (i 5) (if (oddp i) (princ "odd: %d
" i) (princ "even: %d
" i)))
 --> (invalid-function 0)

[2]> (princ "toto")
toto --> "toto"

[3]> (dotimes (i 5) (princ (format (if (oddp i) "odd:  %d
" "even: %d
") i)))
even: 0
odd:  1
even: 2
odd:  3
even: 4
 --> nil

[4]> (/ 5550690 30)
 --> 185023

[5]> (quit)
Good bye.

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
The world will now reboot; don't bother saving your artefacts.