Re: [O] bug in lisp code blocks

[Eric Schulte]

Hi Mark,

Thanks for the example, I've just pushed up a fix.

Best,
Eric

Mark Scala <address@hidden> writes:

> Thanks, Bastien, for helping me report this bug more effectively.
> This email summarizes all the information I have about this bug and
> how to reproduce it. The problem is that a comment which appears as
> the _last_ line of a (common) lisp code block causes an error when
> that block is evaluated with C-c C-c. Here's how to reproduce it:
>
> 1. You will need to have SBCL and Slime
>    installed. Your distribution of linux probably has a good enough
>    SBCL (I'm using 1.2.0 in Arch linux). The easiest way to get the
>    latest version of Slime is with Quicklisp, which you can find here
>    (together with instructions to install):
>    http://www.quicklisp.org/beta/.
>
>    Once you have Quicklisp installed and loaded, installing Slime is a
>    matter of calling (ql:quickload "swank") at the sbcl repl.
>
>   Note: I am also able to reproduce the bug using clisp:
>     -  GNU CLISP 2.49 (2010-07-07) (built on
> var-lib-archbuild-extra-x86_64-barthalion)
>
> 2. Start emacs with the following as the entire init.el (of course,
>    you'll probably have to modify the paths).
>
> #+BEGIN_SRC emacs-lisp
>
> (add-to-list 'load-path (expand-file-name "~/opt/org-mode/lisp"))
>                                                             (require
> 'org)
>
> (org-babel-do-load-languages
>
>   'org-babel-load-languages
>
>       '((lisp . t)))
>
>           (setq inferior-lisp-program "/usr/bin/sbcl")
>
>         (add-to-list 'load-path
> "~/quicklisp/dists/quicklisp/software/slime-2.8")
>                                   (require 'slime)
> #+END_SRC
>
> Note: I've also just reproduced the bug using the version of org-mode
> that ships with Emacs 24:
>
>    - Org-mode version 7.9.3f (release_7.9.3f-17-g7524ef @
> /usr/share/emacs/24.3/lisp/org/)
>
> 3. Finally, create a new org file and insert this block:
>
> #+BEGIN_SRC lisp :results output
>
> (format t "Hello world.")
>   ;; comment after, causes crash when block is evaluated with C-c C-c
> #+END_SRC
>
>
> If you now evaluate the block with C-c C-c, you should get an EOF
> error. Here is a piece of the debugger report:
>
> #+BEGIN_QUOTE
> end of file on #<SB-IMPL::STRING-INPUT-STREAM {1003F862D3}>
>    [Condition of type END-OF-FILE]
> #+END_QUOTE
>
> It appears that the bug is caused by a comment string that is the last
> element of the block. Comments elsewhere do not cause this error.
>
> Here are the versions strings of the software I'm running:
>
>   - GNU Emacs 24.3.1 (x86_64-unknown-linux-gnu, GTK+ Version 3.12.2)
> of 2014-06-11 on var-lib-archbuild-staging-x86_64-jgc
>   - Org-mode version 8.3beta (release_8.3beta-51-g02f2d5 @
> /home/mark/opt/org-mode/lisp/)
>   - SBCL 1.2.0
>
> On Tue, Jul 29, 2014 at 9:13 AM, Bastien <address@hidden> wrote:
>> Hi Mark,
>>
>> you forgot to CC the list -- please do so, so that anyone can help
>> fixing the problem.
>>
>> All this new information is important.  A complete recipe would take
>> the user from a bare emacs -Q to an emacs with a sbcl connection and
>> slime, and explain the steps for launching sbcl, connecting via slime,
>> etc.
>>
>> Assume the ones who may fix the problem (like me and other maintainers)
>> are not 100% familiar with Emacs/slime/sbcl, even though they can follow
>> good recipes easily.
>>
>> Thanks for your efforts!
>>
>> PS: The hint about the last commented line triggered the bug is
>> important and will certainly lead to a quick fix.
>>
>> --
>>  Bastien
>

-- 
Eric Schulte
https://cs.unm.edu/~eschulte
PGP: 0x614CA05D (see https://u.fsf.org/yw)