1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
|
from test_support import TestFailed
from random import random
# XXX need many, many more tests here.
nerrors = 0
def check_close_real(x, y, eps=1e-12):
"""Return true iff floats x and y "are close\""""
# put the one with larger magnitude second
if abs(x) > abs(y):
x, y = y, x
if y == 0:
return abs(x) < eps
if x == 0:
return abs(y) < eps
# check that relative difference < eps
return abs((x-y)/y) < eps
def check_close(x, y, eps=1e-12):
"""Return true iff complexes x and y "are close\""""
return check_close_real(x.real, y.real, eps) and \
check_close_real(x.imag, y.imag, eps)
def test_div(x, y):
"""Compute complex z=x*y, and check that z/x==y and z/y==x."""
global nerrors
z = x * y
if x != 0:
q = z / x
if not check_close(q, y):
nerrors += 1
print `z`, "/", `x`, "==", `q`, "but expected", `y`
if y != 0:
q = z / y
if not check_close(q, x):
nerrors += 1
print `z`, "/", `y`, "==", `q`, "but expected", `x`
simple_real = [float(i) for i in range(-5, 6)]
simple_complex = [complex(x, y) for x in simple_real for y in simple_real]
for x in simple_complex:
for y in simple_complex:
test_div(x, y)
# A naive complex division algorithm (such as in 2.0) is very prone to
# nonsense errors for these (overflows and underflows).
test_div(complex(1e200, 1e200), 1+0j)
test_div(complex(1e-200, 1e-200), 1+0j)
# Just for fun.
for i in range(100):
test_div(complex(random(), random()),
complex(random(), random()))
try:
z = 1.0 / (0+0j)
except ZeroDivisionError:
pass
else:
nerrors += 1
raise TestFailed("Division by complex 0 didn't raise ZeroDivisionError")
if nerrors:
raise TestFailed("%d tests failed" % nerrors)
|
