Use Vose's Alias Method to sample the weighted distribution.

The weight generation code also was cleaned up (and now can support
generating distributions that look like what ScrambleSuit does as
a compile time change).

Per: http://www.keithschwarz.com/darts-dice-coins/

1 files changed, 3 insertions, 3 deletions

diff --git a/csrand/csrand.go b/csrand/csrand.go
index a3299aa..b059ed0 100644
--- a/csrand/csrand.go
+++ b/csrand/csrand.go
@@ -68,9 +68,9 @@ func (r csRandSource) Seed(seed int64) {
 	// No-op.
 }
 
-// Int63n returns, as a int64, a pseudo random number in [0, n).
-func Int63n(n int64) int64 {
-	return CsRand.Int63n(n)
+// Intn returns, as a int, a pseudo random number in [0, n).
+func Intn(n int) int {
+	return CsRand.Intn(n)
 }
 
 // Float64 returns, as a float64, a pesudo random number in [0.0,1.0).