Fix type instabilities in extirpolate

Do not assert lengths of X and Y are the same, this is already done in
`_press_rybicki'; make N mandatory and do not change its value; make
`denominator' always an AbstractFloat.

src/extirpolation.jl

@@ -30,13 +30,9 @@ end
 
 function extirpolate{RE<:Real,NU<:Number}(X::AbstractVector{RE},
                                           Y::AbstractVector{NU},
-                                          N::Integer=0, M::Integer=4)
-    @assert length(X) == length(Y)
-    x, y = collect(X), collect(Y)
-    if N <= 0
-        # Get the maximum of "X", `maximum' has a faster method for ranges.
-        N = trunc(Int, maximum(X) + 0.5*M + 1)
-    end
+                                          N::Integer, M::Integer=4)
+    x = collect(X)
+    y = copy(Y)
     # Now use legendre polynomial weights to populate the results array; This is
     # an efficient recursive implementation (See Press et al. 1989)
     result = zeros(NU, N)
@@ -50,7 +46,7 @@ function extirpolate{RE<:Real,NU<:Number}(X::AbstractVector{RE},
     # the limits are within the range 0...N
     ilo = clamp(trunc(Int, x - div(M, 2)), 0, N - M)
     numerator = y .* [prod(x[j] - ilo[j] - (0:(M - 1))) for j in eachindex(x)]
-    denominator = factorial(M - 1)
+    denominator = float(factorial(M - 1))
     @inbounds for j in 0:(M - 1)
         if j > 0
             denominator *= j/(j - M)

test/runtests.jl

@@ -129,8 +129,8 @@ LombScargle.add_at!(a, [3, 1, 3, 1, 2], 1:5)
 # Test extirpolation function
 x = linspace(0, 10)
 y = sin(x)
-@test_approx_eq LombScargle.extirpolate(x, y) [0.39537718210649553,3.979484140636793,4.833090108345013,0.506805556164743,-3.828112427525919,-4.748341359084166,-1.3022050566901917,3.3367666084342256,5.070478111668922,1.291245296032218,-0.8264466821981216,0.0,0.0]
-@test_approx_eq LombScargle.extirpolate(x, y, 11) LombScargle.extirpolate(x, y)[1:11]
+@test_approx_eq LombScargle.extirpolate(x, y, 13) [0.39537718210649553,3.979484140636793,4.833090108345013,0.506805556164743,-3.828112427525919,-4.748341359084166,-1.3022050566901917,3.3367666084342256,5.070478111668922,1.291245296032218,-0.8264466821981216,0.0,0.0]
+@test_approx_eq LombScargle.extirpolate(x, y, 11) LombScargle.extirpolate(x, y, 13)[1:11]
 
 # Test trig_sum
 C, S = LombScargle.trig_sum(x, y, 1, 10)