[babel] common lisp / slime evaluation in org-mode

[babel] common lisp / slime evaluation in org-mode

[Erik Iverson]

Hello,

I have started playing around with SLIME and was pleased to find there
was already some support in org-mode for evaluating Common LISP
blocks. The comments in ob-lisp.el makes it clear that it is not
complete support yet. It appears for example that you can only
evaluate one lisp form per code block. Here is an example.

* First, let's try in emacs lisp

The following works how I would expect. The value in the results block
is the value of the last expression in the code block.

#+begin_src emacs-lisp

(defvar test1 "test1 value")
(defvar test2 "test2 value")
test2

#+end_src

#+results:
: test2 value

* Now, let's try common lisp

First, we need to start up a SLIME session, with M-x slime. This
assumes you have SLIME set up correctly. I am using SBCL as my CL
environment in this example.

#+begin_src lisp :session

(defvar test1 "test1 value")
(defvar test2 "test2 value")
test2

#+end_src

#+results:
: TEST1

You see the result in the buffer is the value of the first form, and
if I go into the inferior-lisp buffer, test2 is not bound to any
value. If I evaluate the CL code block above without the :session
argument, I get an error in the *Org-Babel Error Output* buffer.

Looking at the code in ob-lisp.el, I think I see why this is
happening, but don't know enough about SLIME yet to make it work. It
seems that line that calls =eval-slime= would need to be changed to
=eval-slime-buffer= after dumping the code block into a temporary
buffer. However, that didn't quite work since that function
(=slime-eval-buffer=) doesn't appear to actually return the final
value.

So, I write this in case this is an easy fix for someone with more
knowledge of SLIME, and to confirm that this is indeed the current
behavior that others see. I will continue to investigate a solution.

Thanks,
--Erik

Re: [babel] common lisp / slime evaluation in org-mode

[Erik Iverson]

<snip>

> Looking at the code in ob-lisp.el, I think I see why this is
> happening, but don't know enough about SLIME yet to make it work. It
> seems that line that calls =eval-slime= would need to be changed to
> =eval-slime-buffer= after dumping the code block into a temporary
> buffer. However, that didn't quite work since that function
> (=slime-eval-buffer=) doesn't appear to actually return the final
> value.

I of course meant slime-eval and slime-eval-buffer above...

Re: [babel] common lisp / slime evaluation in org-mode

[Eric Schulte]

Hi Erik,

I've been switching from Clojure to Common Lisp myself, and have also
found that there are some issues with ob-lisp.el (although I've been too
busy recently to look closely at them).

ob-clojure.el works well, and uses slime for code evaluation -- and it
has a simpler setup than ob-lisp -- so maybe it would be possible to
port the code in ob-clojure over to common lisp?

I'll be happy to help get ob-lisp.el into shape once I have a free
moment...

Cheers -- Eric

Erik Iverson <eriki@ccbr.umn.edu> writes:

> Hello,
>
> I have started playing around with SLIME and was pleased to find there
> was already some support in org-mode for evaluating Common LISP
> blocks. The comments in ob-lisp.el makes it clear that it is not
> complete support yet. It appears for example that you can only
> evaluate one lisp form per code block. Here is an example.
>
> * First, let's try in emacs lisp
>
> The following works how I would expect. The value in the results block
> is the value of the last expression in the code block.
>
> #+begin_src emacs-lisp
>
> (defvar test1 "test1 value")
> (defvar test2 "test2 value")
> test2
>
> #+end_src
>
> #+results:
> : test2 value
>
> * Now, let's try common lisp
>
> First, we need to start up a SLIME session, with M-x slime. This
> assumes you have SLIME set up correctly. I am using SBCL as my CL
> environment in this example.
>
> #+begin_src lisp :session
>
> (defvar test1 "test1 value")
> (defvar test2 "test2 value")
> test2
>
> #+end_src
>
> #+results:
> : TEST1
>
> You see the result in the buffer is the value of the first form, and
> if I go into the inferior-lisp buffer, test2 is not bound to any
> value. If I evaluate the CL code block above without the :session
> argument, I get an error in the *Org-Babel Error Output* buffer.
>
> Looking at the code in ob-lisp.el, I think I see why this is
> happening, but don't know enough about SLIME yet to make it work. It
> seems that line that calls =eval-slime= would need to be changed to
> =eval-slime-buffer= after dumping the code block into a temporary
> buffer. However, that didn't quite work since that function
> (=slime-eval-buffer=) doesn't appear to actually return the final
> value.
>
> So, I write this in case this is an easy fix for someone with more
> knowledge of SLIME, and to confirm that this is indeed the current
> behavior that others see. I will continue to investigate a solution.
>
> Thanks,
> --Erik
>
> _______________________________________________
> Emacs-orgmode mailing list
> Please use `Reply All' to send replies to the list.
> Emacs-orgmode@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-orgmode

Re: [babel] common lisp / slime evaluation in org-mode

[Erik Iverson]

On 02/09/2011 07:18 PM, Eric Schulte wrote:
> Hi Erik,
>
> I've been switching from Clojure to Common Lisp myself, and have also
> found that there are some issues with ob-lisp.el (although I've been too
> busy recently to look closely at them).
>
> ob-clojure.el works well, and uses slime for code evaluation -- and it
> has a simpler setup than ob-lisp -- so maybe it would be possible to
> port the code in ob-clojure over to common lisp?
>
> I'll be happy to help get ob-lisp.el into shape once I have a free
> moment...

Great, in the mean time, I'll definitely take a stab at it and submit
a patch if I fix my issue. Thanks for the tip!

--Erik