[CM] slime cm-2.10 and ASDF

[Torsten Anders]

I meanwhile found out that loading CM into the free PWGL application causes a stack overflow. The stack overflow does not happen directly when loading CM, but after that whenever some computation is attempted, which results in the perceived slow-down of the system. 

Not sure whether this can be overcome by actually compiling CM first... 

Best wishes,
Torsten

--
Dr Torsten Anders
Course Leader, Music Technology
University of Bedfordshire
Park Square, Room A315
http://www.torsten-anders.de



On 6 Aug 2013, at 18:18, Torsten Anders <torsten.anders at beds.ac.uk> wrote:

>> I could load this file into Lisp Works, more specifically as a library into the free PWGL application 
> 
> I meant, I loaded CM2 with this file ...
> 
> Best wishes,
> Torsten
> 
> --
> Dr Torsten Anders
> Course Leader, Music Technology
> University of Bedfordshire
> Park Square, Room A315
> http://www.torsten-anders.de
> 
> 
> 
> On 6 Aug 2013, at 18:17, Torsten Anders <torsten.anders at beds.ac.uk> wrote:
> 
>> Dear Ralf,
>> 
>> Thanks for sharing your cm.asd file. 
>> 
>> I could load this file into Lisp Works, more specifically as a library into the free PWGL application (http://www2.siba.fi/pwgl/downloads.html). So, in principle this would allow to use PWGL (e.g., its score editors and break-point functions etc.) alongside CM2, which would be great. 
>> 
>> However, after loading I noticed that PWGL became unusably slow. Usually, whenever PWGL loads a library all sources are automatically compiled (resulting in *. clufasl files). For some reason, this is not the case after loading your cm.asd file. So, I had a look at that file and noticed that it contains much much more than the *.asd files I saw before (which usually consist of a single asdf:defsystem call. So, it is a bit difficult to figure out why the sources are not compiled. 
>> 
>> However, I noticed that in the asdf:defsystem call, most files are called with :scheme set to T. Why is that? After all, there are *.lisp versions of these files as well?
>> 
>> Thanks! 
>> 
>> For completeness, here are two further minor changes I had to make before I could load CM as a PWGL library. 
>> - Renaming the file cm.asd into cm2.asd (the name of the system defined with asdf:defsystem).
>> - The next error then said "Cannot load this library because Scheme source <PathToCM2>/src/gluplot.scm not found". The  CM tar ball version 2.10.0 I just downloaded from http://sourceforge.net/projects/commonmusic/files/cm/ does not contain this file. Luckily I had this file still somewhere else, but perhaps the tar ball should be updated? 
>> 
>> Best wishes,
>> Torsten
>> 
>> On 6 Aug 2013, at 10:03, R. Mattes <rm at mh-freiburg.de> wrote:
>>> On Tue, 06 Aug 2013 14:53:36 +1200, adam wrote
>>>> 
>>>> I'm sorry to be a pest, gentlemen.
>>> 
>>> You aren't. That's what this list is for.
>>> 
>>>> After setting up sbcl with Quicklisp, and setting the
>>>> environment variable,
>>>> $ export CM_RUNTIME=sbcl
>>>> 
>>>> Running  cm.sh  of CM-2.10 fails with the error below.
>>>> 
>>>> I can supply further information if required. While I can't seem
>>>> to find the sbcl Implementation Notes, I have established that
>>>> UNIX-FILE-KIND is not mentioned in the sbcl manual. Indeed it
>>>> seems to be a reoccurring problem for sbcl.
>>>> 
>>>> If anyone can recommend the best fix or patch for this, then
>>>> much appreciated.
>>> 
>>> As a quick fix, just replace the file cm.asd in your cm-directory
>>> with the one in the attachment.
>>> BTW, I find it way more convenient to use CM like this:
>>> 
>>> - create a symboli link from you cm-directory into ~/quicklisp/local-projects [1]
>>> 
>>> - start your lisp of choice and enter:
>>> 
>>> (ql:quickload :cm)
>>> (cm)
>>> 
>>> HTH RalfD
>>> 
>>> [1] on Linux, do:
>>> 
>>> $ ln -s /where/you/installed/cm/ ~/quicklisp/local-projects/
>>> 
>>> 
>>> --
>>> R. Mattes -
>>> Hochschule fuer Musik Freiburg
>>> rm at inm.mh-freiburg.de
>>> 
>>> <cm.asd>_______________________________________________
>>> Cmdist mailing list
>>> Cmdist at ccrma.stanford.edu
>>> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist
>> 
> 
> 
> _______________________________________________
> Cmdist mailing list
> Cmdist at ccrma.stanford.edu
> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist