mirror of https://github.com/exaloop/codon
120 lines
3.3 KiB
Python
120 lines
3.3 KiB
Python
import random as R
|
|
import time
|
|
import sys
|
|
from copy import copy
|
|
|
|
seed = int(time.time())
|
|
# sys.stderr.write('seed: ' + str(seed) + '\n')
|
|
R.seed(seed)
|
|
|
|
|
|
@test
|
|
def test_rnd_result(name, results, invariant):
|
|
print(name)
|
|
for a in results:
|
|
assert invariant(a)
|
|
|
|
|
|
test_rnd_result(
|
|
"randrange", [R.randrange(10) for _ in range(100)], range(10).__contains__
|
|
)
|
|
test_rnd_result(
|
|
"randrange", [R.randrange(5, 20) for _ in range(100)], range(5, 20).__contains__
|
|
)
|
|
test_rnd_result(
|
|
"randrange",
|
|
[R.randrange(9, 99, 3) for _ in range(100)],
|
|
range(9, 99, 3).__contains__,
|
|
)
|
|
test_rnd_result(
|
|
"randint", [R.randint(5, 20) for _ in range(100)], range(5, 20 + 1).__contains__
|
|
)
|
|
|
|
population = list("ABCDEFGHIJKLMNOP")
|
|
test_rnd_result(
|
|
"choice", [R.choice(population) for _ in range(100)], population.__contains__
|
|
)
|
|
test_rnd_result(
|
|
"choice", [R.choice(population) for _ in range(100)], population.__contains__
|
|
)
|
|
test_rnd_result("choices", R.choices(population), population.__contains__)
|
|
test_rnd_result("choices", R.choices(population, k=100), population.__contains__)
|
|
|
|
|
|
@test
|
|
def test_shuffle(v):
|
|
s = copy(v)
|
|
R.shuffle(s)
|
|
assert sorted(v) == sorted(s)
|
|
|
|
|
|
test_shuffle(list(range(100)))
|
|
|
|
|
|
@test
|
|
def test_sample(n: int, k: int):
|
|
s = R.sample(list(range(n)), k=k)
|
|
assert len(s) == k
|
|
assert len(set(s)) == len(s)
|
|
for a in s:
|
|
assert a in range(n)
|
|
|
|
|
|
test_sample(100, 5)
|
|
test_sample(100, 100)
|
|
test_sample(100, 0)
|
|
|
|
|
|
from python import random as Rpy
|
|
|
|
@test
|
|
def test_vs_python(*args, seed, method: Static[str], T: type = float):
|
|
print(seed, method, args)
|
|
R1 = R.Random(seed)
|
|
R2 = Rpy.Random(seed)
|
|
|
|
N = 50
|
|
A1 = [T(getattr(R1, method)(*args)) for _ in range(N)]
|
|
A2 = [T.__from_py__(getattr(R2, method)(*args).p) for _ in range(N)]
|
|
assert A1 == A2
|
|
|
|
test_vs_python(-10, 10, seed=22, method='randrange', T=int)
|
|
test_vs_python(-10, 10, 3, seed=33, method='randrange', T=int)
|
|
test_vs_python(-10, 10, seed=44, method='randint', T=int)
|
|
test_vs_python(20, seed=55, method='getrandbits', T=int)
|
|
test_vs_python(32, seed=55, method='getrandbits', T=int)
|
|
test_vs_python(40, seed=55, method='getrandbits', T=int)
|
|
test_vs_python(63, seed=55, method='getrandbits', T=int)
|
|
test_vs_python(seed=0, method='random')
|
|
test_vs_python(-12.5, 101.2, seed=1, method='uniform')
|
|
test_vs_python(-13, 5.5, 0, seed=2, method='triangular')
|
|
#test_vs_python(1.0, 2, seed=3, method='betavariate') # different in older Python versions
|
|
test_vs_python(0.3, seed=4, method='expovariate')
|
|
#test_vs_python(1.0, 2, seed=5, method='gammavariate') # different in older Python versions
|
|
test_vs_python(1.0, 2.0, seed=-101, method='gauss')
|
|
test_vs_python(1.0, 2.0, seed=-102, method='lognormvariate')
|
|
test_vs_python(1.0, 2.0, seed=-103, method='normalvariate')
|
|
test_vs_python(1.0, 2.0, seed=0xffffffff, method='vonmisesvariate')
|
|
test_vs_python(1.0, seed=0xffffffff-1, method='paretovariate')
|
|
test_vs_python(1.0, 2.0, seed=0, method='weibullvariate')
|
|
|
|
|
|
@test
|
|
def test_state():
|
|
r = R.Random(1234)
|
|
state = r.getstate()
|
|
N = 100
|
|
A1 = [r.random() for _ in range(N)]
|
|
B1 = [r.gauss() for _ in range(N)]
|
|
r.setstate(state)
|
|
A2 = [r.random() for _ in range(N)]
|
|
B2 = [r.gauss() for _ in range(N)]
|
|
r.seed(1234)
|
|
A3 = [r.random() for _ in range(N)]
|
|
B3 = [r.gauss() for _ in range(N)]
|
|
|
|
assert A1 == A2 == A3
|
|
assert B1 == B2 == B3
|
|
|
|
test_state()
|