Re: defsystem and eval

On Aug 18, 2005, at 10:01 AM, Raymond Toy wrote:

>>>>>> "Sam" == Sam Steingold <[email protected]> writes:
>
>>> * Raymond Toy <[email protected]> [2005-08-16 21:37:09 -0400]:
>>>
>>> Matlisp also uses mk:defsys because there was no asdf back then.
>
>     Sam> could you please try the latest clocc cvs head of defsystem?
>     Sam> defsystem now evaluates some arguments:
>
> Matlisp compiles just fine.  But matlisp doesn't do anything special
> in its use of defsystem.

Matlisp compiles fine be because its use of :FINALLY-DO is a special 
case that is unaffected by the latest changes.

> Perhaps you meant maxima?  Unfortunately maxima uses its own hacked
> version to support gcl.  I can certainly try it with cmucl, though.

The Maxima hacked version just includes a number of missing ANSI 
thingies.  WITH-COMPILATION-UNIT and (unbelievalbly!!!) 
*LOAD-PATHNAME*.

IMO, all in all the change must be reverted.  Otherwise, an 
inconsistency is introduced between top-level DEFSYSTEM specs and 
subcomponent specs.  Sorry Sam :)

What exactly would the change allow that it is not allowed now?

Cheers
--
Marco Antoniotti					http://bioinformatics.nyu.edu
NYU Courant Bioinformatics Group		tel. +1 - 212 - 998 3488
715 Broadway 10th FL				fax. +1 - 212 - 998 3484
New York, NY, 10003, U.S.A.



-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf