[MIT-Scheme-devel] call-with-values requires thunk to use 'values'? Why?

[Daniel Katz]

Hi!

I tried loading Neil Van Dyke's 'testeez' unit testing package (which is 
designed to be R5RS compliant) into MIT-scheme 9.0.1 today, and ran into 
trouble with 'call-with-values'.  In particular, for some test I wrote it turns 
out that a part of Neil's code ends up as the form 

(call-with-values (lambda () 4) list)

which MIT-scheme fails to appreciate: 

> 
> 1 > (call-with-values (lambda () 4) list)
> 
> ;The object 4 is not applicable.
> ;To continue, call RESTART with an option number:
> ; (RESTART 2) => Specify a procedure to use in its place.
> ; (RESTART 1) => Return to read-eval-print level 1.
> 
> 2 error> (restart 1)
> 
> ;Abort!


The MIT-scheme documentation implies that the first function really ought to 
return its values using the 'values' form, which does work:

> 1 > (call-with-values (lambda () (values 4)) list)
> 
> ;Value 3: (4)


but when I look at R5RS, there doesn't seem to be anything in the description 
of call-with-values that would require the use of 'values' to return the values 
from the first argument to call-with-values.  In fact, R5RS even gives an 
example where that's not true 
(http://www.schemers.org/Documents/Standards/R5RS/HTML/r5rs-Z-H-9.html#%_idx_574):

(call-with-values * -) ===> -1

which also doesn't work in MIT-scheme 9.0.1:

> 1 > (call-with-values * -)
> 
> ;The object 1 is not applicable.
> ;To continue, call RESTART with an option number:
> ; (RESTART 2) => Specify a procedure to use in its place.
> ; (RESTART 1) => Return to read-eval-print level 1.
> 
> 2 error> 


Is this just a 'yet to be finished' area of R5RS compliance for MIT-scheme?  Or 
is there some particular reason why the 'values' requirement on 
'call-with-values' has been imposed?

Thanks!

Dan