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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245
|
package ksuid
import (
"bytes"
"sort"
"strings"
"testing"
)
func TestBase10ToBase62AndBack(t *testing.T) {
number := []byte{1, 2, 3, 4}
encoded := base2base(number, 10, 62)
decoded := base2base(encoded, 62, 10)
if bytes.Compare(number, decoded) != 0 {
t.Fatal(number, " != ", decoded)
}
}
func TestBase256ToBase62AndBack(t *testing.T) {
number := []byte{255, 254, 253, 251}
encoded := base2base(number, 256, 62)
decoded := base2base(encoded, 62, 256)
if bytes.Compare(number, decoded) != 0 {
t.Fatal(number, " != ", decoded)
}
}
func TestEncodeAndDecodeBase62(t *testing.T) {
helloWorld := []byte("hello world")
encoded := encodeBase62(helloWorld)
decoded := decodeBase62(encoded)
if len(encoded) < len(helloWorld) {
t.Fatal("length of encoded base62 string", encoded, "should be >= than raw bytes!")
}
if bytes.Compare(helloWorld, decoded) != 0 {
t.Fatal(decoded, " != ", helloWorld)
}
}
func TestLexographicOrdering(t *testing.T) {
unsortedStrings := make([]string, 256)
for i := 0; i < 256; i++ {
s := string(encodeBase62([]byte{0, byte(i)}))
unsortedStrings[i] = strings.Repeat("0", 2-len(s)) + s
}
if !sort.StringsAreSorted(unsortedStrings) {
sortedStrings := make([]string, len(unsortedStrings))
for i, s := range unsortedStrings {
sortedStrings[i] = s
}
sort.Strings(sortedStrings)
t.Fatal("base62 encoder does not produce lexographically sorted output.",
"expected:", sortedStrings,
"actual:", unsortedStrings)
}
}
func TestBase62Value(t *testing.T) {
s := base62Characters
for i := range s {
v := int(base62Value(s[i]))
if v != i {
t.Error("bad value:")
t.Log("<<<", i)
t.Log(">>>", v)
}
}
}
func TestFastAppendEncodeBase62(t *testing.T) {
for i := 0; i != 1000; i++ {
id := New()
b0 := id[:]
b1 := appendEncodeBase62(nil, b0)
b2 := fastAppendEncodeBase62(nil, b0)
s1 := string(leftpad(b1, '0', stringEncodedLength))
s2 := string(b2)
if s1 != s2 {
t.Error("bad base62 representation of", id)
t.Log("<<<", s1, len(s1))
t.Log(">>>", s2, len(s2))
}
}
}
func TestFastAppendDecodeBase62(t *testing.T) {
for i := 0; i != 1000; i++ {
id := New()
b0 := leftpad(encodeBase62(id[:]), '0', stringEncodedLength)
b1 := appendDecodeBase62(nil, []byte(string(b0))) // because it modifies the input buffer
b2 := fastAppendDecodeBase62(nil, b0)
if !bytes.Equal(leftpad(b1, 0, byteLength), b2) {
t.Error("bad binary representation of", string(b0))
t.Log("<<<", b1)
t.Log(">>>", b2)
}
}
}
func BenchmarkAppendEncodeBase62(b *testing.B) {
a := [stringEncodedLength]byte{}
id := New()
for i := 0; i != b.N; i++ {
appendEncodeBase62(a[:0], id[:])
}
}
func BenchmarkAppendFastEncodeBase62(b *testing.B) {
a := [stringEncodedLength]byte{}
id := New()
for i := 0; i != b.N; i++ {
fastAppendEncodeBase62(a[:0], id[:])
}
}
func BenchmarkAppendDecodeBase62(b *testing.B) {
a := [byteLength]byte{}
id := []byte(New().String())
for i := 0; i != b.N; i++ {
b := [stringEncodedLength]byte{}
copy(b[:], id)
appendDecodeBase62(a[:0], b[:])
}
}
func BenchmarkAppendFastDecodeBase62(b *testing.B) {
a := [byteLength]byte{}
id := []byte(New().String())
for i := 0; i != b.N; i++ {
fastAppendDecodeBase62(a[:0], id)
}
}
// The functions bellow were the initial implementation of the base conversion
// algorithms, they were replaced by optimized versions later on. We keep them
// in the test files as a reference to ensure compatibility between the generic
// and optimized implementations.
func appendBase2Base(dst []byte, src []byte, inBase int, outBase int) []byte {
off := len(dst)
bs := src[:]
bq := [stringEncodedLength]byte{}
for len(bs) > 0 {
length := len(bs)
quotient := bq[:0]
remainder := 0
for i := 0; i != length; i++ {
acc := int(bs[i]) + remainder*inBase
d := acc/outBase | 0
remainder = acc % outBase
if len(quotient) > 0 || d > 0 {
quotient = append(quotient, byte(d))
}
}
// Appends in reverse order, the byte slice gets reversed before it's
// returned by the function.
dst = append(dst, byte(remainder))
bs = quotient
}
reverse(dst[off:])
return dst
}
func base2base(src []byte, inBase int, outBase int) []byte {
return appendBase2Base(nil, src, inBase, outBase)
}
func appendEncodeBase62(dst []byte, src []byte) []byte {
off := len(dst)
dst = appendBase2Base(dst, src, 256, 62)
for i, c := range dst[off:] {
dst[off+i] = base62Characters[c]
}
return dst
}
func encodeBase62(in []byte) []byte {
return appendEncodeBase62(nil, in)
}
func appendDecodeBase62(dst []byte, src []byte) []byte {
// Kind of intrusive, we modify the input buffer... it's OK here, it saves
// a memory allocation in Parse.
for i, b := range src {
// O(1)... technically. Has better real-world perf than a map
src[i] = byte(strings.IndexByte(base62Characters, b))
}
return appendBase2Base(dst, src, 62, 256)
}
func decodeBase62(src []byte) []byte {
return appendDecodeBase62(
make([]byte, 0, len(src)*2),
append(make([]byte, 0, len(src)), src...),
)
}
func reverse(b []byte) {
i := 0
j := len(b) - 1
for i < j {
b[i], b[j] = b[j], b[i]
i++
j--
}
}
func leftpad(b []byte, c byte, n int) []byte {
if n -= len(b); n > 0 {
for i := 0; i != n; i++ {
b = append(b, c)
}
copy(b[n:], b)
for i := 0; i != n; i++ {
b[i] = c
}
}
return b
}
|