Ah. Thanks. Robby
On Fri, May 28, 2010 at 8:51 PM, Sam Tobin-Hochstadt <sa...@ccs.neu.edu> wrote: > On Fri, May 28, 2010 at 9:21 PM, Robby Findler > <ro...@eecs.northwestern.edu> wrote: >>> - [else (loop (- i 1) >>> + [else (loop (sub1 i) >> >> Okay, I'll bite. Why does (- i 1) do something different than (sub1 >> i)? Is sub1 overloaded in some useful way that - cannot be? Or is it >> something about the constant? > > `sub1' has the type, among others, of: > > (Exact-Positive-Integer -> Exact-Nonnegative-Integer) > > In this example, `i' (the argument to `loop') has argument type > `Exact-Nonnegative-Integer'. But above that line, we've tested that > `i' isn't 0. So `i' is an `Exact-Positive-Integer'. So the `sub1' > call produces exactly the right type for the call to `loop'. `-' > can't have a type that does that, unless we special-case 1. > > -- > sam th > sa...@ccs.neu.edu > _________________________________________________ For list-related administrative tasks: http://list.cs.brown.edu/mailman/listinfo/plt-dev
