[cells-devel] Re: Where's the MOP?

ktilton at nyc.rr.com ktilton at nyc.rr.com
Wed Jun 22 15:21:37 UTC 2005



----- Original Message -----
From: Luke Crook <luke at balooga.com>
Date: Tuesday, June 21, 2005 8:56 pm
Subject: Re: Where's the MOP?

> ktilton at nyc.rr.com writes: 
> 
> > 
> > Does this help? Or, does it not work?: 
> > 
> > (defpackage :cells
> >...
> >      #+cormanlisp #:common-lisp
> >... 
> > 
> >      #-(or allegro clisp cmu cormanlisp lispworks mcl sbcl)
> >...
> 
> 
> Making progress, thanks. 
> 
> (pushnew #p"C:/Lisp/libs/" asdf-util:*source-dirs* :test #'equal)
> (#P"C:\Lisp\libs\")
> (asdf:oos 'asdf:load-op :cells)
> ; loading system definition from #P"C:\Lisp\libs\cells_2.0\cells.asd"
> ; into #<PACKAGE "ASDF7512">
> ; registering #<SYSTEM :CELLS #xF21D38> as CELLS
> ;;; Warning: the following function(s) are called from forms
> ;;; in file "#P"C:\Lisp\libs\cells_2.0\cell-types.lisp"" but have 
> not yet 
> been defined:
> ;;;     C-BREAK
> ;;;     C-OPTIMIZE-AWAY?!
> ;;;     MD-STATE
> ;;;     C-CURRENTP
> ;;; An error occurred in function MACROEXPAND:
> ;;; Error: There is no SETF expansion method defined for (MD-SLOT-
> CELL-TYPE 
> CLASS-NAME SLOT-NAME)

This...:

(defun (setf md-slot-cell-type) (new-type class-name slot-name)
  (let ((entry (assoc slot-name (get class-name :cell-types))))
    (if entry
        (setf (cdr entry) new-type)
      (push (cons slot-name new-type) (get class-name :cell-types)))))

...is found in model-object.lisp. Possibly there is a missing dependency
in cells.asd. Just make sure whichever file is failing to compile has a
dependency directly ot indirectly on model-object.lisp.

FYI, this happens because I do not use ASDF in my regular work, so the
cells.asd does not get heavy use.

Mind you, I am only guessing. Possibly Corman Lisp has a problem with
(defun (setf ....)). You should check to see if other symbols from
model-object.lisp exist, or its FASL, or in some way determine if
model-object has been both compileed and loaded.

If so, and if the setter still is not acknowledged, there might be a
problem with Corman.

kt




More information about the cells-devel mailing list