Re: Lisp code blocks fail

[Eric Schulte <schulte.eric@gmail.com>]

Hi Tom,

Sorry about the slow reply, I've been busy as usual.

I just pushed up a fix for this problem.  As you've discovered, the
slime function returns a two-element list holding any strings written to
STDOUT, and the value of the evaluated code.  In two different places
org-babel-execute:lisp was trying to split this list into the separate
values however by the second time the list had been reduced to "2".

I've just pushed up a fix.

Best,

tsd@tsdye.com (Thomas S. Dye) writes:

> Hi Eric,
>
> I understand this a bit better now.
>
> In org-babel-execute:lisp, the local variable result is getting set to
> ("" "2").
>
> With :results value, the call to read in org-babel-result-cond is being
> passed "2" when it expects a list.  With :results output, it gets "",
> which triggers the same error.
>
> As I noted earlier, slime-eval executes the lisp code properly, but for
> some reason it appears to be returning a string instead of a vector or a
> list.
>
> I'm not sure what I did to make this stop working.  It was fine
> a week ago. I haven't changed slime in six months AFAIK.
>
> Other babel languages work as usual. My problem appears to be confined
> to lisp.
>
> All the best,
> Tom
>
> tsd@tsdye.com (Thomas S. Dye) writes:
>
>> Hi Eric,
>>
>> An edebug adventure :)
>>
>> AFAICT, the error is thrown by this call to a macro:
>> (org-babel-result-cond ("replace") (car result)), where result has been
>> set to "2" by the long (funcall) in (let).
>>
>> Specifically, (org-babel-lisp-vector-to-list "2") appears to trigger the
>> (error)--at least that is the last code line edebug stops at before
>> getting to the (error) line. The modeline says Result: "2" just before
>> the error.
>>
>> Does this help?
>>
>> Tom
>>
>> Eric Schulte <schulte.eric@gmail.com> writes:
>>
>>> Hi Tom,
>>>
>>> That example works for me.  Perhaps you could edebug (C-u C-M-x) the
>>> `org-babel-execute:lisp' function and then run that code block again to
>>> see at which statement the error is thrown.
>>>
>>> I'm not sure what the problem could be.
>>>
>>> Best,
>>>
>>> tsd@tsdye.com (Thomas S. Dye) writes:
>>>
>>>> Hi Eric,
>>>>
>>>> #+begin_src lisp
>>>> (+ 1 1)
>>>> #+end_src
>>>>
>>>> Debugger entered--Lisp error: (wrong-type-argument listp "2")
>>>>   byte-code("\b\211A@)\207" [result x] 2)
>>>>   org-babel-execute:lisp("(+ 1 1)" ((:comments . "") (:shebang . "")
>>>> (:cache . "no") (:padline . "") (:noweb . "yes") (:tangle . "no")
>>>> (:exports . "code") (:results . "replace") (:session . "none")
>>>> (:hlines . "no") (:result-type . value) (:result-params "replace")
>>>> (:rowname-names) (:colname-names)))
>>>>   org-babel-execute-src-block(nil)
>>>>   org-babel-execute-src-block-maybe()
>>>>   org-babel-execute-maybe()
>>>>   org-babel-execute-safely-maybe()
>>>>   run-hook-with-args-until-success(org-babel-execute-safely-maybe)
>>>>   org-ctrl-c-ctrl-c(nil)
>>>>   ad-Orig-call-interactively(org-ctrl-c-ctrl-c nil nil)
>>>>   (with-no-warnings (ad-Orig-call-interactively function record-flag keys))
>>>>   (setq ad-return-value (with-no-warnings
>>>> (ad-Orig-call-interactively function record-flag keys)))
>>>>   (let ((ido-ubiquitous-next-override
>>>> (ido-ubiquitous-get-command-override function))) (setq ad-return-value
>>>> (with-no-warnings (ad-Orig-call-interactively function record-flag
>>>> keys))))
>>>>   (ido-ubiquitous-with-override (ido-ubiquitous-get-command-override
>>>> function) (setq ad-return-value (with-no-warnings
>>>> (ad-Orig-call-interactively function record-flag keys))))
>>>>   (let (ad-return-value) (ido-ubiquitous-with-override
>>>> (ido-ubiquitous-get-command-override function) (setq ad-return-value
>>>> (with-no-warnings (ad-Orig-call-interactively function record-flag
>>>> keys)))) ad-return-value)
>>>>   call-interactively(org-ctrl-c-ctrl-c nil nil)
>>>>
>>>> Org-mode version 8.2.3b (release_8.2.3b-200-gb6522a @
>>>> /Users/dk/.emacs.d/src/org-mode/lisp/)
>>>>
>>>> All the best,
>>>> Tom
>>>>
>>>> Eric Schulte <schulte.eric@gmail.com> writes:
>>>>
>>>>> Hi Tom,
>>>>>
>>>>> Could you provide a minimal example?  I'm unable to debug from the stack
>>>>> trace alone.

-- 
Eric Schulte
https://cs.unm.edu/~eschulte
PGP: 0x614CA05D