Re: patch to unif.c...

[Gary Houston <ghouston at freewire dot co dot uk>]

| >>>>> "Per" == Per Bothner <per@bothner.com> writes:
| 
|     Per> Clark McGrew <mcgrew@ale.physics.sunysb.edu> writes:
| 
|     >> I keep running into this problem over and over again when I'm
|     >> working with uniform vectors; There isn't a good clean way to
|     >> generate a vector of floats vs doubles.
| 
|     Per> Can I suggest SRFI-4 "Homogeneous numeric vector datatypes"?
|     Per> http://srfi.schemers.org/srfi-4/srfi-4.html
| 
|     Per> I'm a big believer in following standards, as long as the
|     Per> standard is reasonable, and I believe this one is.
|
| I agree in the long run.  The problem as the array support in guile
| currently stands there is no  clean way to generate a vector of floats
| vs doubles.  My patch is the minimal change to handle the problem
| (seriously, it only took 30-40 key stokes).  

The interface is pretty weird.  Wouldn't it work in this case if you
used the canonical templates?:

guile> (make-uniform-vector 3 1/3) 
#i(0.333333333333333 0.333333333333333 0.333333333333333)
guile> (make-uniform-vector 3 1.0)
#s(1.0 1.0 1.0)

There's an optional argument for the initialiser:

guile> (make-uniform-vector 3 1.0 1/3)
#s(0.333333343267441 0.333333343267441 0.333333343267441)