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
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
|
package xrand
import (
"context"
"encoding/binary"
"math"
"math/rand"
"testing"
"github.com/bradenaw/juniper/internal/require2"
"github.com/bradenaw/juniper/iterator"
"github.com/bradenaw/juniper/stream"
)
type fuzzRand struct {
t *testing.T
b []byte
}
func (r *fuzzRand) Intn(n int) int {
if len(r.b) < 4 {
return 0
}
x := binary.BigEndian.Uint32(r.b[:4])
r.b = r.b[4:]
return int(x) % n
}
func (r *fuzzRand) Float64() float64 {
if len(r.b) < 8 {
return 0
}
x := binary.BigEndian.Uint64(r.b[:8])
r.b = r.b[8:]
out := float64(x) / math.MaxUint64
if out == 1 {
out = math.Nextafter(out, 0)
}
require2.GreaterOrEqual(r.t, out, float64(0))
require2.Less(r.t, out, float64(1))
r.t.Logf("%f", out)
return out
}
func (r *fuzzRand) Shuffle(int, func(int, int)) {
panic("unimplemented")
}
func FuzzSampleInner(f *testing.F) {
f.Fuzz(func(t *testing.T, b []byte, k int) {
if k <= 0 {
return
}
t.Logf("k %d", k)
r := &fuzzRand{t, b}
samp := newSampler(r, k)
prev := 0
for i := 0; i < 100; i++ {
next, replace := samp.Next()
t.Logf("%d: next %d replace %d", i, next, replace)
if next == math.MaxInt {
break
}
if i < k {
require2.Equal(t, next, i)
require2.Equal(t, replace, i)
} else {
require2.Greater(t, next, prev)
require2.GreaterOrEqual(t, replace, 0)
require2.Less(t, replace, k)
}
prev = next
}
})
}
func stddev(a []int) float64 {
m := mean(a)
sumSquaredDeviation := float64(0)
for i := range a {
deviation := m - float64(a[i])
sumSquaredDeviation += (deviation * deviation)
}
return math.Sqrt(sumSquaredDeviation / float64(len(a)))
}
func mean(a []int) float64 {
sum := 0
for i := range a {
sum += a[i]
}
return float64(sum) / float64(len(a))
}
// f must return the same as Sample(r, 20, 5).
func testSample(t *testing.T, f func(r *rand.Rand) []int) {
r := rand.New(rand.NewSource(0))
counts := make([]int, 20)
for i := 0; i < 10000; i++ {
sample := f(r)
for _, item := range sample {
counts[item]++
}
}
m := mean(counts)
t.Logf("counts %#v", counts)
t.Logf("stddev %#v", stddev(counts))
t.Logf("stddev / mean %#v", stddev(counts)/m)
// There's certainly a better statistical test than this, but I haven't bothered to break out
// the stats book yet.
require2.InDelta(t, 0.02, stddev(counts)/m, 0.01)
}
func TestSample(t *testing.T) {
testSample(t, func(r *rand.Rand) []int {
return RSample(r, 20, 5)
})
}
func TestSampleSlice(t *testing.T) {
a := iterator.Collect(iterator.Counter(20))
testSample(t, func(r *rand.Rand) []int {
return RSampleSlice(r, a, 5)
})
}
func TestSampleIterator(t *testing.T) {
testSample(t, func(r *rand.Rand) []int {
return RSampleIterator(r, iterator.Counter(20), 5)
})
}
func TestSampleStream(t *testing.T) {
testSample(t, func(r *rand.Rand) []int {
out, err := RSampleStream(
context.Background(),
r,
stream.FromIterator(iterator.Counter(20)),
5,
)
require2.NoError(t, err)
return out
})
}
|
