(upgraded-complex-part-type 'cons)? 
Author Message
 (upgraded-complex-part-type 'cons)?

What is upgraded-complex-part-type supposed to do when given a
typespec that isn't a subtype of real?

Clisp, ACL, and CMUCL return T for this.  LW barfs.

CLHS doesn't really cover this issue, but the entry for the class
complex says the typespec must be a subtype of real.

I think LW is right since you can't have a complex number with
components which are not subtypes of REAL.

Ray



Wed, 05 Jan 2005 02:23:59 GMT  
 
 [ 1 post ] 

 Relevant Pages 

1. (upgraded-complex-part-type 'rational) ==> ?

2. (cons type type)

3. An Attempt to Formalise Dylan's Type System (Part 2)

4. An Attempt to Formalise Dylan's Type System (Part 1)

5. Telecommuting: Pro's and Con's

6. OREXX - linein('con:')

7. Fortran and Pro's/Con's

8. Strong type checking + Heterogenous con

9. pros and cons of user-defined types

10. using DEFINE to break complex function into simpler parts

11. passing only the real parts of a complex array to a function

12. real and imaginary part from python complex numbers

 

 
Powered by phpBB® Forum Software