gmane.lisp.slime.devel

Daily ChangeLog diff

Helmut Eller — 2012-05-21T09:44:01

Index: slime/ChangeLog
diff -u slime/ChangeLog:1.2340 slime/ChangeLog:1.2341
--- slime/ChangeLog:1.2340Fri May 11 23:34:47 2012
+++ slime/ChangeLogMon May 21 01:08:07 2012
< at >< at > -1,3 +1,12 < at >< at >
+2012-05-21  Helmut Eller  <heller< at >common-lisp.net>
+
+Reduce dependecy on pre-defined faces.
+
+* slime.el (slime-apropos-symbol, slime-apropos-label): New faces.
+(slime-print-apropos): Use them.
+
+Reported by: Daimrod <daimrod< at >gmail.com>
+
 2012-05-12  Helmut Eller  <heller< at >common-lisp.net>
 
 * swank-sbcl.lisp (stream-force-output): Undo last few changes.

Re: [bug] SLIME apropos

Helmut Eller — 2012-05-21T08:09:59


Thanks for the bug report.  I defined our own faces:
slime-apropos-symbol and slime-apropos-label.  That should work in both
Emacs versions.

Helmut

[bug] SLIME apropos

Daimrod — 2012-05-18T10:17:39


Because of this, slime-apropos-package fails. (apropos-symbol-face and
apropos-label-face are undefined).

For now I've set them to NIL but I hope there is a better solution
compatible with both emacs23 and emacs24.

Re: providing swank-media

Nicolas Martyanoff — 2012-05-17T11:09:36


I didn't have this kind of problem; what kind of error message did you
get ?

Re: providing swank-media

Christophe Rhodes — 2012-05-17T08:18:48


Heh.  I don't know if it's wrong; I wrote slime-media to support R, and
people being adventurous and trying it with CL is new and exciting to
me. :-)  I'll be merging patches for it, probably with some latency.

(It would be good to not have to set emacs to allow remote evaluation
from CL in order to get the goodies; also I have in my memory that
Nikodemus said that there was some issue with the messages which made it
hard to use from CL: probably a package issue to do with the image
symbols.)

Cheers,

Christophe

providing swank-media

Cyrus Harmon — 2012-05-17T04:16:32

So I was trying to Nicolas Martyanoff's example of using slime-media up and running and I ran into a problem where swank-media wasn't providing swank-media. A trivial fix is attached, but perhaps I'm doing something else wrong. Anyone else see this?

thanks,

Cyrus

diff --git i/contrib/swank-media.lisp w/contrib/swank-media.lisp
index 218fe9e..3d5ef7c 100644
--- i/contrib/swank-media.lisp
+++ w/contrib/swank-media.lisp
< at >< at > -21,3 +21,5 < at >< at >
 ;; can only be defined if their specializers already exist; in R's S3
 ;; object system, methods are ordinary functions with a special naming
 ;; convention)
+
+(provide :swank-media)

Slime autocompletion for Common Lisp code

Yuan Luo — 2012-05-16T19:22:28

Hi,
I am trying to use slime-fuzzy-complete-symbol for common lisp code auto
completion. The Lisp dialect I am using is Allegro CL and Emacs is the IDE.
Under its common lisp mode, I use "M-x slime-mode" to enable slime for code
editing. My problem is, for example
(defun foo ()
  (form)
However, after pressing the tab key, it only prompts me (defun name varlist
&rest body) rather than the desired format function. I also note that under
*slime-repl allegro*, tab key generates a fuzzy completion list. What am I
missing here?

Best,
Yuan

Re: cutting and pasting S-expressions from buffer toREPL

Pascal J. Bourguignon — 2012-05-15T00:50:54


It shows that Grant Rettke doesn't know anything about emacs.
If he had only read the tutorial (control-h t), he would know that it's
written C-x C-e.

Re: cutting and pasting S-expressions from buffer toREPL

Grant Rettke — 2012-05-15T01:05:42

On Mon, May 14, 2012 at 7:50 PM, Pascal J. Bourguignon <
pjb< at >informatimago.com> wrote:

I would have read it but someone told that it was:

C:h, t

Re: cutting and pasting S-expressions from buffer to REPL

Stelian Ionescu — 2012-05-14T19:23:12

[...]

Indeed. If you need to modify it, use the first command, otherwise use
the second

Re: cutting and pasting S-expressions from buffer to REPL

Faheem Mitha — 2012-05-14T19:19:10


On Mon, 14 May 2012, Stelian Ionescu wrote:







[snip code]


Nice! That works for me. However, may I clarify/confirm usage?

If I put the cursor on the closing parenthesis of the expression, then the 
first command just copies it to the REPL, the second copies and then 
evaluates it. Is this correct usage?


Well, like I said earlier, in practice, you may have to modify the 
expression before you can run it.

Personally, I think SLIME could do with a command like this.

Thanks for the nice piece of code, Stelian.

                                                    Regards, Faheem

Re: cutting and pasting S-expressions from buffer to REPL

Stelian Ionescu — 2012-05-14T18:50:03

This is what I have in my .emacs(wasn't written by me and I don't
remember where I copied it from):

(defun slime-send-dwim (arg)
  "Send the appropriate forms to CL to be evaluated."
  (interactive "P")
  (save-excursion
    (cond
      ;;Region selected - evaluate region
      ((not (equal mark-active nil))
       (copy-region-as-kill (mark) (point)))
      ;; At/before sexp - evaluate next sexp
      ((or (looking-at "(")
           (save-excursion
             (ignore-errors (forward-char 1))
             (looking-at "(")))
       (forward-list 1)
       (let ((end (point))
             (beg (save-excursion
                    (backward-list 1)
                    (point))))
         (copy-region-as-kill beg end)))
      ;; At/after sexp - evaluate last sexp
      ((or (looking-at ")")
           (save-excursion
             (backward-char 1)
             (looking-at ")")))
       (if (looking-at ")")
           (forward-char 1))
       (let ((end (point))
             (beg (save-excursion

Re: cutting and pasting S-expressions from buffer to REPL

Faheem Mitha — 2012-05-14T18:47:27


On Mon, 14 May 2012, Grant Rettke wrote:


Great, thanks.

                                                       Regards, Faheem

Re: cutting and pasting S-expressions from buffer to REPL

Faheem Mitha — 2012-05-14T18:45:49


On Mon, 14 May 2012, Dave Cooper wrote:


No, I'm trying to copy it over to the REPL. The idea is to try to evaluate 
the S-expression on its own. This doesn't work if one tries to evaluate it 
directly inside the function because typically it needs to use stuff 
(variables) that are defined by evaluations outside that expression. This 
works fine for self-contained expressions, but those are the exception.

Anyway, the idea of copying it over to the REPL is so I can modify it 
suitably.

                                                       Regards, Faheem

Re: cutting and pasting S-expressions from buffer toREPL

Grant Rettke — 2012-05-14T18:37:12

Do "C:h, k" and then do "C:x, C:e" an emacs will open a window to describe
what that keybinding does.

It sends the last s-expr to the repl.

On Mon, May 14, 2012 at 1:33 PM, Faheem Mitha <faheem< at >faheem.info> wrote:

Re: cutting and pasting S-expressions from buffer to REPL

Faheem Mitha — 2012-05-14T18:33:18


On Mon, 14 May 2012, Grant Rettke wrote:


What does "C:x, C:e" do?

         Regards, Faheem

Re: cutting and pasting S-expressions from buffer to REPL

Faheem Mitha — 2012-05-14T18:32:03


On Mon, 14 May 2012, Dave Cooper wrote:


Hi Dave,

Wow, that was a fast reply. Thanks for the helpful tips. I tried them, and 
they work.

Yes, I'd like to be able to do it in one shot, but I don't know any Emacs 
Lisp. I wonder how different Emacs Lisp is from Common Lisp.

                                                           Regards, Faheem

Re: cutting and pasting S-expressions from buffer toREPL

Grant Rettke — 2012-05-14T18:27:45

I think Faheem wants "C:x, C:e" with the cursor at the right of the
expression, C:M:f to get it there. I could be wrong, too.

On Mon, May 14, 2012 at 1:07 PM, Dave Cooper <gendl< at >genworks.com> wrote:

Re: cutting and pasting S-expressions from buffer toREPL

Dave Cooper — 2012-05-14T18:07:44

Hi Faheem,

 With the cursor on the _opening_ paren of an s-expression, you can do

   C-M-<space>

 to select the S-expression, then

  M-w

 to copy it.

 Then go over to the repl (I bind a personal shortcut of C-x & for this), then

 C-y

to "yank" (paste) the expression to the repl.

I'm not sure about automatically switching to the repl and pasting in
one go, but I'm sure that would be a few lines of emacs lisp...

And a quick way to get the cursor to the opening paren of an s-expression is

  C-M-b



On Mon, May 14, 2012 at 1:35 PM, Faheem Mitha <faheem< at >faheem.info> wrote:

cutting and pasting S-expressions from buffer to REPL

Faheem Mitha — 2012-05-14T17:35:54

Hello everyone,

I've recently started using SLIME with SBCl on Debian, where SLIME 
defaults to using SBCL, at least when it is installed. So far I am very 
impressed. I haven't been able to take off and fly yet, but I'm on the 
runway.:-)

I'm particularly impressed by the underlining source which has compile 
errors, and doing a popup on mouseover feature. I don't know whose idea 
this was, but it is pretty damn cool. Does anyone know the history of this 
feature?

So, I have a question about SLIME. When debugging, or indeed otherwise, I 
find it useful to copy and paste S-expressions from a buffer to the REPL. 
I've been using the mouse for this, which is decidedly sub-optimal. I hate 
using the mouse. I was thinking that it would be handy to have a command 
which would copy and paste S-expressions from a buffer to the REPL, while 
at the same time switching the buffer to the REPL. The S-expression would 
be chosen by having the cusor to the right of the closing expression. I'm 
not sure if this positio

Re: Daily ChangeLog diff

Helmut Eller — 2012-05-13T21:06:58


If a timeout occurs in the middle of a non-atomic code sequence, some
heap object can be in a non-consistent state.  E.g. the buffer contents
gets sent to Emacs before resetting the buffer index; if the timeout
occurs before updating the index, some output can be sent twice.

Helmut