Lawrence Bottorff <borgauf@gmail.com> writes:

> I'm using Racket with Geiser and I get this error:
>
> executing Scheme code block...
> => #f
> org-babel-scheme-execute-with-geiser: Invalid read syntax: "#"
>
> when in an org-mode file this code
>
> #+begin_src scheme :exports both :session ch3
> (define (bool-imply2 x y)
>   (or (not x) y))
> #+end_src
>
> is run (C-c-c)
>
> #+BEGIN_SRC scheme :session ch3
> (bool-imply2 #t #f)
> #+END_SRC
>
> In the Racket "ch3" REPL session (bool-imply2 #t #f) works fine. I discovered this problem when I first
> tried a simple export to HTML of the buffer. It seems to not like the second boolean parameter. So
> switching the parameters then complains about #t . What might be going on? BTW, does an export try to
> "run" all the code blocks, i.e., why did I find this when I was exporting to HTML?
>
> LB
>

Aargh - somebody (maybe you?) had run into this a long time ago and I
had suggested a possible fix, but with the demise of the gmane site, I
cannot find the thread - how do people search the ML nowadays?

In any case, there is a basic error in ob-scheme.el, line 176 (at
least in the version that I have):

        ...
        (setq result (if (or (string= result "#<void>")
                             (string= result "#<unspecified>"))
                         nil
                       (read result)))))   ;;;<<<<<<<<<<<<<<<<
        ...

The (read result) is bogus: it tries to use the emacs lisp reader to
parse a string that contains a scheme expression.

IIRC, I suggested changing it to just result:

        ...
        (setq result (if (or (string= result "#<void>")
                             (string= result "#<unspecified>"))
                         nil
                       result))))
        ...

but I didn't (and still don't) know if that breaks anything else.

Plus I'm on a machine that doesn't have geiser so I can't even test the basic "fix",
so I hope I've got it right. I'll try to follow up tonight from a machine that has
geiser installed.

--
Nick