// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package strings_test
import (
"bytes"
"fmt"
"io"
"math"
"math/rand"
"reflect"
"strconv"
. "strings"
"testing"
"unicode"
"unicode/utf8"
"unsafe"
)
func eq(a, b []string) bool {
if len(a) != len(b) {
return false
}
for i := 0; i < len(a); i++ {
if a[i] != b[i] {
return false
}
}
return true
}
var abcd = "abcd"
var faces = "☺☻☹"
var commas = "1,2,3,4"
var dots = "1....2....3....4"
type IndexTest struct {
s string
sep string
out int
}
var indexTests = []IndexTest{
{"", "", 0},
{"", "a", -1},
{"", "foo", -1},
{"fo", "foo", -1},
{"foo", "foo", 0},
{"oofofoofooo", "f", 2},
{"oofofoofooo", "foo", 4},
{"barfoobarfoo", "foo", 3},
{"foo", "", 0},
{"foo", "o", 1},
{"abcABCabc", "A", 3},
// cases with one byte strings - test special case in Index()
{"", "a", -1},
{"x", "a", -1},
{"x", "x", 0},
{"abc", "a", 0},
{"abc", "b", 1},
{"abc", "c", 2},
{"abc", "x", -1},
// test special cases in Index() for short strings
{"", "ab", -1},
{"bc", "ab", -1},
{"ab", "ab", 0},
{"xab", "ab", 1},
{"xab"[:2], "ab", -1},
{"", "abc", -1},
{"xbc", "abc", -1},
{"abc", "abc", 0},
{"xabc", "abc", 1},
{"xabc"[:3], "abc", -1},
{"xabxc", "abc", -1},
{"", "abcd", -1},
{"xbcd", "abcd", -1},
{"abcd", "abcd", 0},
{"xabcd", "abcd", 1},
{"xyabcd"[:5], "abcd", -1},
{"xbcqq", "abcqq", -1},
{"abcqq", "abcqq", 0},
{"xabcqq", "abcqq", 1},
{"xyabcqq"[:6], "abcqq", -1},
{"xabxcqq", "abcqq", -1},
{"xabcqxq", "abcqq", -1},
{"", "01234567", -1},
{"32145678", "01234567", -1},
{"01234567", "01234567", 0},
{"x01234567", "01234567", 1},
{"x0123456x01234567", "01234567", 9},
{"xx01234567"[:9], "01234567", -1},
{"", "0123456789", -1},
{"3214567844", "0123456789", -1},
{"0123456789", "0123456789", 0},
{"x0123456789", "0123456789", 1},
{"x012345678x0123456789", "0123456789", 11},
{"xyz0123456789"[:12], "0123456789", -1},
{"x01234567x89", "0123456789", -1},
{"", "0123456789012345", -1},
{"3214567889012345", "0123456789012345", -1},
{"0123456789012345", "0123456789012345", 0},
{"x0123456789012345", "0123456789012345", 1},
{"x012345678901234x0123456789012345", "0123456789012345", 17},
{"", "01234567890123456789", -1},
{"32145678890123456789", "01234567890123456789", -1},
{"01234567890123456789", "01234567890123456789", 0},
{"x01234567890123456789", "01234567890123456789", 1},
{"x0123456789012345678x01234567890123456789", "01234567890123456789", 21},
{"xyz01234567890123456789"[:22], "01234567890123456789", -1},
{"", "0123456789012345678901234567890", -1},
{"321456788901234567890123456789012345678911", "0123456789012345678901234567890", -1},
{"0123456789012345678901234567890", "0123456789012345678901234567890", 0},
{"x0123456789012345678901234567890", "0123456789012345678901234567890", 1},
{"x012345678901234567890123456789x0123456789012345678901234567890", "0123456789012345678901234567890", 32},
{"xyz0123456789012345678901234567890"[:33], "0123456789012345678901234567890", -1},
{"", "01234567890123456789012345678901", -1},
{"32145678890123456789012345678901234567890211", "01234567890123456789012345678901", -1},
{"01234567890123456789012345678901", "01234567890123456789012345678901", 0},
{"x01234567890123456789012345678901", "01234567890123456789012345678901", 1},
{"x0123456789012345678901234567890x01234567890123456789012345678901", "01234567890123456789012345678901", 33},
{"xyz01234567890123456789012345678901"[:34], "01234567890123456789012345678901", -1},
{"xxxxxx012345678901234567890123456789012345678901234567890123456789012", "012345678901234567890123456789012345678901234567890123456789012", 6},
{"", "0123456789012345678901234567890123456789", -1},
{"xx012345678901234567890123456789012345678901234567890123456789012", "0123456789012345678901234567890123456789", 2},
{"xx012345678901234567890123456789012345678901234567890123456789012"[:41], "0123456789012345678901234567890123456789", -1},
{"xx012345678901234567890123456789012345678901234567890123456789012", "0123456789012345678901234567890123456xxx", -1},
{"xx0123456789012345678901234567890123456789012345678901234567890120123456789012345678901234567890123456xxx", "0123456789012345678901234567890123456xxx", 65},
// test fallback to Rabin-Karp.
{"oxoxoxoxoxoxoxoxoxoxoxoy", "oy", 22},
{"oxoxoxoxoxoxoxoxoxoxoxox", "oy", -1},
}
var lastIndexTests = []IndexTest{
{"", "", 0},
{"", "a", -1},
{"", "foo", -1},
{"fo", "foo", -1},
{"foo", "foo", 0},
{"foo", "f", 0},
{"oofofoofooo", "f", 7},
{"oofofoofooo", "foo", 7},
{"barfoobarfoo", "foo", 9},
{"foo", "", 3},
{"foo", "o", 2},
{"abcABCabc", "A", 3},
{"abcABCabc", "a", 6},
}
var indexAnyTests = []IndexTest{
{"", "", -1},
{"", "a", -1},
{"", "abc", -1},
{"a", "", -1},
{"a", "a", 0},
{"\x80", "\xffb", 0},
{"aaa", "a", 0},
{"abc", "xyz", -1},
{"abc", "xcz", 2},
{"ab☺c", "x☺yz", 2},
{"a☺b☻c☹d", "cx", len("a☺b☻")},
{"a☺b☻c☹d", "uvw☻xyz", len("a☺b")},
{"aRegExp*", ".(|)*+?^$[]", 7},
{dots + dots + dots, " ", -1},
{"012abcba210", "\xffb", 4},
{"012\x80bcb\x80210", "\xffb", 3},
{"0123456\xcf\x80abc", "\xcfb\x80", 10},
}
var lastIndexAnyTests = []IndexTest{
{"", "", -1},
{"", "a", -1},
{"", "abc", -1},
{"a", "", -1},
{"a", "a", 0},
{"\x80", "\xffb", 0},
{"aaa", "a", 2},
{"abc", "xyz", -1},
{"abc", "ab", 1},
{"ab☺c", "x☺yz", 2},
{"a☺b☻c☹d", "cx", len("a☺b☻")},
{"a☺b☻c☹d", "uvw☻xyz", len("a☺b")},
{"a.RegExp*", ".(|)*+?^$[]", 8},
{dots + dots + dots, " ", -1},
{"012abcba210", "\xffb", 6},
{"012\x80bcb\x80210", "\xffb", 7},
{"0123456\xcf\x80abc", "\xcfb\x80", 10},
}
// Execute f on each test case. funcName should be the name of f; it's used
// in failure reports.
func runIndexTests(t *testing.T, f func(s, sep string) int, funcName string, testCases []IndexTest) {
for _, test := range testCases {
actual := f(test.s, test.sep)
if actual != test.out {
t.Errorf("%s(%q,%q) = %v; want %v", funcName, test.s, test.sep, actual, test.out)
}
}
}
func TestIndex(t *testing.T) { runIndexTests(t, Index, "Index", indexTests) }
func TestLastIndex(t *testing.T) { runIndexTests(t, LastIndex, "LastIndex", lastIndexTests) }
func TestIndexAny(t *testing.T) { runIndexTests(t, IndexAny, "IndexAny", indexAnyTests) }
func TestLastIndexAny(t *testing.T) {
runIndexTests(t, LastIndexAny, "LastIndexAny", lastIndexAnyTests)
}
func TestIndexByte(t *testing.T) {
for _, tt := range indexTests {
if len(tt.sep) != 1 {
continue
}
pos := IndexByte(tt.s, tt.sep[0])
if pos != tt.out {
t.Errorf(`IndexByte(%q, %q) = %v; want %v`, tt.s, tt.sep[0], pos, tt.out)
}
}
}
func TestLastIndexByte(t *testing.T) {
testCases := []IndexTest{
{"", "q", -1},
{"abcdef", "q", -1},
{"abcdefabcdef", "a", len("abcdef")}, // something in the middle
{"abcdefabcdef", "f", len("abcdefabcde")}, // last byte
{"zabcdefabcdef", "z", 0}, // first byte
{"a☺b☻c☹d", "b", len("a☺")}, // non-ascii
}
for _, test := range testCases {
actual := LastIndexByte(test.s, test.sep[0])
if actual != test.out {
t.Errorf("LastIndexByte(%q,%c) = %v; want %v", test.s, test.sep[0], actual, test.out)
}
}
}
func simpleIndex(s, sep string) int {
n := len(sep)
for i := n; i <= len(s); i++ {
if s[i-n:i] == sep {
return i - n
}
}
return -1
}
func TestIndexRandom(t *testing.T) {
const chars = "abcdefghijklmnopqrstuvwxyz0123456789"
for times := 0; times < 10; times++ {
for strLen := 5 + rand.Intn(5); strLen < 140; strLen += 10 { // Arbitrary
s1 := make([]byte, strLen)
for i := range s1 {
s1[i] = chars[rand.Intn(len(chars))]
}
s := string(s1)
for i := 0; i < 50; i++ {
begin := rand.Intn(len(s) + 1)
end := begin + rand.Intn(len(s)+1-begin)
sep := s[begin:end]
if i%4 == 0 {
pos := rand.Intn(len(sep) + 1)
sep = sep[:pos] + "A" + sep[pos:]
}
want := simpleIndex(s, sep)
res := Index(s, sep)
if res != want {
t.Errorf("Index(%s,%s) = %d; want %d", s, sep, res, want)
}
}
}
}
}
func TestIndexRune(t *testing.T) {
tests := []struct {
in string
rune rune
want int
}{
{"", 'a', -1},
{"", '☺', -1},
{"foo", '☹', -1},
{"foo", 'o', 1},
{"foo☺bar", '☺', 3},
{"foo☺☻☹bar", '☹', 9},
{"a A x", 'A', 2},
{"some_text=some_value", '=', 9},
{"☺a", 'a', 3},
{"a☻☺b", '☺', 4},
// RuneError should match any invalid UTF-8 byte sequence.
{"�", '�', 0},
{"\xff", '�', 0},
{"☻x�", '�', len("☻x")},
{"☻x\xe2\x98", '�', len("☻x")},
{"☻x\xe2\x98�", '�', len("☻x")},
{"☻x\xe2\x98x", '�', len("☻x")},
// Invalid rune values should never match.
{"a☺b☻c☹d\xe2\x98�\xff�\xed\xa0\x80", -1, -1},
{"a☺b☻c☹d\xe2\x98�\xff�\xed\xa0\x80", 0xD800, -1}, // Surrogate pair
{"a☺b☻c☹d\xe2\x98�\xff�\xed\xa0\x80", utf8.MaxRune + 1, -1},
}
for _, tt := range tests {
if got := IndexRune(tt.in, tt.rune); got != tt.want {
t.Errorf("IndexRune(%q, %d) = %v; want %v", tt.in, tt.rune, got, tt.want)
}
}
haystack := "test世界"
allocs := testing.AllocsPerRun(1000, func() {
if i := IndexRune(haystack, 's'); i != 2 {
t.Fatalf("'s' at %d; want 2", i)
}
if i := IndexRune(haystack, '世'); i != 4 {
t.Fatalf("'世' at %d; want 4", i)
}
})
if allocs != 0 && testing.CoverMode() == "" {
t.Errorf("expected no allocations, got %f", allocs)
}
}
const benchmarkString = "some_text=some☺value"
func BenchmarkIndexRune(b *testing.B) {
if got := IndexRune(benchmarkString, '☺'); got != 14 {
b.Fatalf("wrong index: expected 14, got=%d", got)
}
for i := 0; i < b.N; i++ {
IndexRune(benchmarkString, '☺')
}
}
var benchmarkLongString = Repeat(" ", 100) + benchmarkString
func BenchmarkIndexRuneLongString(b *testing.B) {
if got := IndexRune(benchmarkLongString, '☺'); got != 114 {
b.Fatalf("wrong index: expected 114, got=%d", got)
}
for i := 0; i < b.N; i++ {
IndexRune(benchmarkLongString, '☺')
}
}
func BenchmarkIndexRuneFastPath(b *testing.B) {
if got := IndexRune(benchmarkString, 'v'); got != 17 {
b.Fatalf("wrong index: expected 17, got=%d", got)
}
for i := 0; i < b.N; i++ {
IndexRune(benchmarkString, 'v')
}
}
func BenchmarkIndex(b *testing.B) {
if got := Index(benchmarkString, "v"); got != 17 {
b.Fatalf("wrong index: expected 17, got=%d", got)
}
for i := 0; i < b.N; i++ {
Index(benchmarkString, "v")
}
}
func BenchmarkLastIndex(b *testing.B) {
if got := Index(benchmarkString, "v"); got != 17 {
b.Fatalf("wrong index: expected 17, got=%d", got)
}
for i := 0; i < b.N; i++ {
LastIndex(benchmarkString, "v")
}
}
func BenchmarkIndexByte(b *testing.B) {
if got := IndexByte(benchmarkString, 'v'); got != 17 {
b.Fatalf("wrong index: expected 17, got=%d", got)
}
for i := 0; i < b.N; i++ {
IndexByte(benchmarkString, 'v')
}
}
type SplitTest struct {
s string
sep string
n int
a []string
}
var splittests = []SplitTest{
{"", "", -1, []string{}},
{abcd, "", 2, []string{"a", "bcd"}},
{abcd, "", 4, []string{"a", "b", "c", "d"}},
{abcd, "", -1, []string{"a", "b", "c", "d"}},
{faces, "", -1, []string{"☺", "☻", "☹"}},
{faces, "", 3, []string{"☺", "☻", "☹"}},
{faces, "", 17, []string{"☺", "☻", "☹"}},
{"☺�☹", "", -1, []string{"☺", "�", "☹"}},
{abcd, "a", 0, nil},
{abcd, "a", -1, []string{"", "bcd"}},
{abcd, "z", -1, []string{"abcd"}},
{commas, ",", -1, []string{"1", "2", "3", "4"}},
{dots, "...", -1, []string{"1", ".2", ".3", ".4"}},
{faces, "☹", -1, []string{"☺☻", ""}},
{faces, "~", -1, []string{faces}},
{"1 2 3 4", " ", 3, []string{"1", "2", "3 4"}},
{"1 2", " ", 3, []string{"1", "2"}},
{"", "T", math.MaxInt / 4, []string{""}},
}
func TestSplit(t *testing.T) {
for _, tt := range splittests {
a := SplitN(tt.s, tt.sep, tt.n)
if !eq(a, tt.a) {
t.Errorf("Split(%q, %q, %d) = %v; want %v", tt.s, tt.sep, tt.n, a, tt.a)
continue
}
if tt.n == 0 {
continue
}
s := Join(a, tt.sep)
if s != tt.s {
t.Errorf("Join(Split(%q, %q, %d), %q) = %q", tt.s, tt.sep, tt.n, tt.sep, s)
}
if tt.n < 0 {
b := Split(tt.s, tt.sep)
if !reflect.DeepEqual(a, b) {
t.Errorf("Split disagrees with SplitN(%q, %q, %d) = %v; want %v", tt.s, tt.sep, tt.n, b, a)
}
}
}
}
var splitaftertests = []SplitTest{
{abcd, "a", -1, []string{"a", "bcd"}},
{abcd, "z", -1, []string{"abcd"}},
{abcd, "", -1, []string{"a", "b", "c", "d"}},
{commas, ",", -1, []string{"1,", "2,", "3,", "4"}},
{dots, "...", -1, []string{"1...", ".2...", ".3...", ".4"}},
{faces, "☹", -1, []string{"☺☻☹", ""}},
{faces, "~", -1, []string{faces}},
{faces, "", -1, []string{"☺", "☻", "☹"}},
{"1 2 3 4", " ", 3, []string{"1 ", "2 ", "3 4"}},
{"1 2 3", " ", 3, []string{"1 ", "2 ", "3"}},
{"1 2", " ", 3, []string{"1 ", "2"}},
{"123", "", 2, []string{"1", "23"}},
{"123", "", 17, []string{"1", "2", "3"}},
}
func TestSplitAfter(t *testing.T) {
for _, tt := range splitaftertests {
a := SplitAfterN(tt.s, tt.sep, tt.n)
if !eq(a, tt.a) {
t.Errorf(`Split(%q, %q, %d) = %v; want %v`, tt.s, tt.sep, tt.n, a, tt.a)
continue
}
s := Join(a, "")
if s != tt.s {
t.Errorf(`Join(Split(%q, %q, %d), %q) = %q`, tt.s, tt.sep, tt.n, tt.sep, s)
}
if tt.n < 0 {
b := SplitAfter(tt.s, tt.sep)
if !reflect.DeepEqual(a, b) {
t.Errorf("SplitAfter disagrees with SplitAfterN(%q, %q, %d) = %v; want %v", tt.s, tt.sep, tt.n, b, a)
}
}
}
}
type FieldsTest struct {
s string
a []string
}
var fieldstests = []FieldsTest{
{"", []string{}},
{" ", []string{}},
{" \t ", []string{}},
{"\u2000", []string{}},
{" abc ", []string{"abc"}},
{"1 2 3 4", []string{"1", "2", "3", "4"}},
{"1 2 3 4", []string{"1", "2", "3", "4"}},
{"1\t\t2\t\t3\t4", []string{"1", "2", "3", "4"}},
{"1\u20002\u20013\u20024", []string{"1", "2", "3", "4"}},
{"\u2000\u2001\u2002", []string{}},
{"\n™\t™\n", []string{"™", "™"}},
{"\n\u20001™2\u2000 \u2001 ™", []string{"1™2", "™"}},
{"\n1\uFFFD \uFFFD2\u20003\uFFFD4", []string{"1\uFFFD", "\uFFFD2", "3\uFFFD4"}},
{"1\xFF\u2000\xFF2\xFF \xFF", []string{"1\xFF", "\xFF2\xFF", "\xFF"}},
{faces, []string{faces}},
}
func TestFields(t *testing.T) {
for _, tt := range fieldstests {
a := Fields(tt.s)
if !eq(a, tt.a) {
t.Errorf("Fields(%q) = %v; want %v", tt.s, a, tt.a)
continue
}
}
}
var FieldsFuncTests = []FieldsTest{
{"", []string{}},
{"XX", []string{}},
{"XXhiXXX", []string{"hi"}},
{"aXXbXXXcX", []string{"a", "b", "c"}},
}
func TestFieldsFunc(t *testing.T) {
for _, tt := range fieldstests {
a := FieldsFunc(tt.s, unicode.IsSpace)
if !eq(a, tt.a) {
t.Errorf("FieldsFunc(%q, unicode.IsSpace) = %v; want %v", tt.s, a, tt.a)
continue
}
}
pred := func(c rune) bool { return c == 'X' }
for _, tt := range FieldsFuncTests {
a := FieldsFunc(tt.s, pred)
if !eq(a, tt.a) {
t.Errorf("FieldsFunc(%q) = %v, want %v", tt.s, a, tt.a)
}
}
}
// Test case for any function which accepts and returns a single string.
type StringTest struct {
in, out string
}
// Execute f on each test case. funcName should be the name of f; it's used
// in failure reports.
func runStringTests(t *testing.T, f func(string) string, funcName string, testCases []StringTest) {
for _, tc := range testCases {
actual := f(tc.in)
if actual != tc.out {
t.Errorf("%s(%q) = %q; want %q", funcName, tc.in, actual, tc.out)
}
}
}
var upperTests = []StringTest{
{"", ""},
{"ONLYUPPER", "ONLYUPPER"},
{"abc", "ABC"},
{"AbC123", "ABC123"},
{"azAZ09_", "AZAZ09_"},
{"longStrinGwitHmixofsmaLLandcAps", "LONGSTRINGWITHMIXOFSMALLANDCAPS"},
{"long\u0250string\u0250with\u0250nonascii\u2C6Fchars", "LONG\u2C6FSTRING\u2C6FWITH\u2C6FNONASCII\u2C6FCHARS"},
{"\u0250\u0250\u0250\u0250\u0250", "\u2C6F\u2C6F\u2C6F\u2C6F\u2C6F"}, // grows one byte per char
{"a\u0080\U0010FFFF", "A\u0080\U0010FFFF"}, // test utf8.RuneSelf and utf8.MaxRune
}
var lowerTests = []StringTest{
{"", ""},
{"abc", "abc"},
{"AbC123", "abc123"},
{"azAZ09_", "azaz09_"},
{"longStrinGwitHmixofsmaLLandcAps", "longstringwithmixofsmallandcaps"},
{"LONG\u2C6FSTRING\u2C6FWITH\u2C6FNONASCII\u2C6FCHARS", "long\u0250string\u0250with\u0250nonascii\u0250chars"},
{"\u2C6D\u2C6D\u2C6D\u2C6D\u2C6D", "\u0251\u0251\u0251\u0251\u0251"}, // shrinks one byte per char
{"A\u0080\U0010FFFF", "a\u0080\U0010FFFF"}, // test utf8.RuneSelf and utf8.MaxRune
}
const space = "\t\v\r\f\n\u0085\u00a0\u2000\u3000"
var trimSpaceTests = []StringTest{
{"", ""},
{"abc", "abc"},
{space + "abc" + space, "abc"},
{" ", ""},
{" \t\r\n \t\t\r\r\n\n ", ""},
{" \t\r\n x\t\t\r\r\n\n ", "x"},
{" \u2000\t\r\n x\t\t\r\r\ny\n \u3000", "x\t\t\r\r\ny"},
{"1 \t\r\n2", "1 \t\r\n2"},
{" x\x80", "x\x80"},
{" x\xc0", "x\xc0"},
{"x \xc0\xc0 ", "x \xc0\xc0"},
{"x \xc0", "x \xc0"},
{"x \xc0 ", "x \xc0"},
{"x \xc0\xc0 ", "x \xc0\xc0"},
{"x ☺\xc0\xc0 ", "x ☺\xc0\xc0"},
{"x ☺ ", "x ☺"},
}
func tenRunes(ch rune) string {
r := make([]rune, 10)
for i := range r {
r[i] = ch
}
return string(r)
}
// User-defined self-inverse mapping function
func rot13(r rune) rune {
step := rune(13)
if r >= 'a' && r <= 'z' {
return ((r - 'a' + step) % 26) + 'a'
}
if r >= 'A' && r <= 'Z' {
return ((r - 'A' + step) % 26) + 'A'
}
return r
}
func TestMap(t *testing.T) {
// Run a couple of awful growth/shrinkage tests
a := tenRunes('a')
// 1. Grow. This triggers two reallocations in Map.
maxRune := func(rune) rune { return unicode.MaxRune }
m := Map(maxRune, a)
expect := tenRunes(unicode.MaxRune)
if m != expect {
t.Errorf("growing: expected %q got %q", expect, m)
}
// 2. Shrink
minRune := func(rune) rune { return 'a' }
m = Map(minRune, tenRunes(unicode.MaxRune))
expect = a
if m != expect {
t.Errorf("shrinking: expected %q got %q", expect, m)
}
// 3. Rot13
m = Map(rot13, "a to zed")
expect = "n gb mrq"
if m != expect {
t.Errorf("rot13: expected %q got %q", expect, m)
}
// 4. Rot13^2
m = Map(rot13, Map(rot13, "a to zed"))
expect = "a to zed"
if m != expect {
t.Errorf("rot13: expected %q got %q", expect, m)
}
// 5. Drop
dropNotLatin := func(r rune) rune {
if unicode.Is(unicode.Latin, r) {
return r
}
return -1
}
m = Map(dropNotLatin, "Hello, 세계")
expect = "Hello"
if m != expect {
t.Errorf("drop: expected %q got %q", expect, m)
}
// 6. Identity
identity := func(r rune) rune {
return r
}
orig := "Input string that we expect not to be copied."
m = Map(identity, orig)
if (*reflect.StringHeader)(unsafe.Pointer(&orig)).Data !=
(*reflect.StringHeader)(unsafe.Pointer(&m)).Data {
t.Error("unexpected copy during identity map")
}
// 7. Handle invalid UTF-8 sequence
replaceNotLatin := func(r rune) rune {
if unicode.Is(unicode.Latin, r) {
return r
}
return utf8.RuneError
}
m = Map(replaceNotLatin, "Hello\255World")
expect = "Hello\uFFFDWorld"
if m != expect {
t.Errorf("replace invalid sequence: expected %q got %q", expect, m)
}
// 8. Check utf8.RuneSelf and utf8.MaxRune encoding
encode := func(r rune) rune {
switch r {
case utf8.RuneSelf:
return unicode.MaxRune
case unicode.MaxRune:
return utf8.RuneSelf
}
return r
}
s := string(rune(utf8.RuneSelf)) + string(utf8.MaxRune)
r := string(utf8.MaxRune) + string(rune(utf8.RuneSelf)) // reverse of s
m = Map(encode, s)
if m != r {
t.Errorf("encoding not handled correctly: expected %q got %q", r, m)
}
m = Map(encode, r)
if m != s {
t.Errorf("encoding not handled correctly: expected %q got %q", s, m)
}
// 9. Check mapping occurs in the front, middle and back
trimSpaces := func(r rune) rune {
if unicode.IsSpace(r) {
return -1
}
return r
}
m = Map(trimSpaces, " abc 123 ")
expect = "abc123"
if m != expect {
t.Errorf("trimSpaces: expected %q got %q", expect, m)
}
}
func TestToUpper(t *testing.T) { runStringTests(t, ToUpper, "ToUpper", upperTests) }
func TestToLower(t *testing.T) { runStringTests(t, ToLower, "ToLower", lowerTests) }
var toValidUTF8Tests = []struct {
in string
repl string
out string
}{
{"", "\uFFFD", ""},
{"abc", "\uFFFD", "abc"},
{"\uFDDD", "\uFFFD", "\uFDDD"},
{"a\xffb", "\uFFFD", "a\uFFFDb"},
{"a\xffb\uFFFD", "X", "aXb\uFFFD"},
{"a☺\xffb☺\xC0\xAFc☺\xff", "", "a☺b☺c☺"},
{"a☺\xffb☺\xC0\xAFc☺\xff", "日本語", "a☺日本語b☺日本語c☺日本語"},
{"\xC0\xAF", "\uFFFD", "\uFFFD"},
{"\xE0\x80\xAF", "\uFFFD", "\uFFFD"},
{"\xed\xa0\x80", "abc", "abc"},
{"\xed\xbf\xbf", "\uFFFD", "\uFFFD"},
{"\xF0\x80\x80\xaf", "☺", "☺"},
{"\xF8\x80\x80\x80\xAF", "\uFFFD", "\uFFFD"},
{"\xFC\x80\x80\x80\x80\xAF", "\uFFFD", "\uFFFD"},
}
func TestToValidUTF8(t *testing.T) {
for _, tc := range toValidUTF8Tests {
got := ToValidUTF8(tc.in, tc.repl)
if got != tc.out {
t.Errorf("ToValidUTF8(%q, %q) = %q; want %q", tc.in, tc.repl, got, tc.out)
}
}
}
func BenchmarkToUpper(b *testing.B) {
for _, tc := range upperTests {
b.Run(tc.in, func(b *testing.B) {
for i := 0; i < b.N; i++ {
actual := ToUpper(tc.in)
if actual != tc.out {
b.Errorf("ToUpper(%q) = %q; want %q", tc.in, actual, tc.out)
}
}
})
}
}
func BenchmarkToLower(b *testing.B) {
for _, tc := range lowerTests {
b.Run(tc.in, func(b *testing.B) {
for i := 0; i < b.N; i++ {
actual := ToLower(tc.in)
if actual != tc.out {
b.Errorf("ToLower(%q) = %q; want %q", tc.in, actual, tc.out)
}
}
})
}
}
func BenchmarkMapNoChanges(b *testing.B) {
identity := func(r rune) rune {
return r
}
for i := 0; i < b.N; i++ {
Map(identity, "Some string that won't be modified.")
}
}
func TestSpecialCase(t *testing.T) {
lower := "abcçdefgğhıijklmnoöprsştuüvyz"
upper := "ABCÇDEFGĞHIİJKLMNOÖPRSŞTUÜVYZ"
u := ToUpperSpecial(unicode.TurkishCase, upper)
if u != upper {
t.Errorf("Upper(upper) is %s not %s", u, upper)
}
u = ToUpperSpecial(unicode.TurkishCase, lower)
if u != upper {
t.Errorf("Upper(lower) is %s not %s", u, upper)
}
l := ToLowerSpecial(unicode.TurkishCase, lower)
if l != lower {
t.Errorf("Lower(lower) is %s not %s", l, lower)
}
l = ToLowerSpecial(unicode.TurkishCase, upper)
if l != lower {
t.Errorf("Lower(upper) is %s not %s", l, lower)
}
}
func TestTrimSpace(t *testing.T) { runStringTests(t, TrimSpace, "TrimSpace", trimSpaceTests) }
var trimTests = []struct {
f string
in, arg, out string
}{
{"Trim", "abba", "a", "bb"},
{"Trim", "abba", "ab", ""},
{"TrimLeft", "abba", "ab", ""},
{"TrimRight", "abba", "ab", ""},
{"TrimLeft", "abba", "a", "bba"},
{"TrimLeft", "abba", "b", "abba"},
{"TrimRight", "abba", "a", "abb"},
{"TrimRight", "abba", "b", "abba"},
{"Trim", " ", "", "",
"
helloworld") } func BenchmarkLastIndexHard1(b *testing.B) { benchmarkLastIndexHard(b, "<>") } func BenchmarkLastIndexHard2(b *testing.B) { benchmarkLastIndexHard(b, "") } func BenchmarkLastIndexHard3(b *testing.B) { benchmarkLastIndexHard(b, "hello world") } func BenchmarkCountHard1(b *testing.B) { benchmarkCountHard(b, "<>") } func BenchmarkCountHard2(b *testing.B) { benchmarkCountHard(b, "") } func BenchmarkCountHard3(b *testing.B) { benchmarkCountHard(b, "hello world") } var benchInputTorture = Repeat("ABC", 1<<10) + "123" + Repeat("ABC", 1<<10) var benchNeedleTorture = Repeat("ABC", 1<<10+1) func BenchmarkIndexTorture(b *testing.B) { for i := 0; i < b.N; i++ { Index(benchInputTorture, benchNeedleTorture) } } func BenchmarkCountTorture(b *testing.B) { for i := 0; i < b.N; i++ { Count(benchInputTorture, benchNeedleTorture) } } func BenchmarkCountTortureOverlapping(b *testing.B) { A := Repeat("ABC", 1<<20) B := Repeat("ABC", 1<<10) for i := 0; i < b.N; i++ { Count(A, B) } } func BenchmarkCountByte(b *testing.B) { indexSizes := []int{10, 32, 4 << 10, 4 << 20, 64 << 20} benchStr := Repeat(benchmarkString, (indexSizes[len(indexSizes)-1]+len(benchmarkString)-1)/len(benchmarkString)) benchFunc := func(b *testing.B, benchStr string) { b.SetBytes(int64(len(benchStr))) for i := 0; i < b.N; i++ { Count(benchStr, "=") } } for _, size := range indexSizes { b.Run(fmt.Sprintf("%d", size), func(b *testing.B) { benchFunc(b, benchStr[:size]) }) } } var makeFieldsInput = func() string { x := make([]byte, 1<<20) // Input is ~10% space, ~10% 2-byte UTF-8, rest ASCII non-space. for i := range x { switch rand.Intn(10) { case 0: x[i] = ' ' case 1: if i > 0 && x[i-1] == 'x' { copy(x[i-1:], "χ") break } fallthrough default: x[i] = 'x' } } return string(x) } var makeFieldsInputASCII = func() string { x := make([]byte, 1<<20) // Input is ~10% space, rest ASCII non-space. for i := range x { if rand.Intn(10) == 0 { x[i] = ' ' } else { x[i] = 'x' } } return string(x) } var stringdata = []struct{ name, data string }{ {"ASCII", makeFieldsInputASCII()}, {"Mixed", makeFieldsInput()}, } func BenchmarkFields(b *testing.B) { for _, sd := range stringdata { b.Run(sd.name, func(b *testing.B) { for j := 1 << 4; j <= 1<<20; j <<= 4 { b.Run(fmt.Sprintf("%d", j), func(b *testing.B) { b.ReportAllocs() b.SetBytes(int64(j)) data := sd.data[:j] for i := 0; i < b.N; i++ { Fields(data) } }) } }) } } func BenchmarkFieldsFunc(b *testing.B) { for _, sd := range stringdata { b.Run(sd.name, func(b *testing.B) { for j := 1 << 4; j <= 1<<20; j <<= 4 { b.Run(fmt.Sprintf("%d", j), func(b *testing.B) { b.ReportAllocs() b.SetBytes(int64(j)) data := sd.data[:j] for i := 0; i < b.N; i++ { FieldsFunc(data, unicode.IsSpace) } }) } }) } } func BenchmarkSplitEmptySeparator(b *testing.B) { for i := 0; i < b.N; i++ { Split(benchInputHard, "") } } func BenchmarkSplitSingleByteSeparator(b *testing.B) { for i := 0; i < b.N; i++ { Split(benchInputHard, "/") } } func BenchmarkSplitMultiByteSeparator(b *testing.B) { for i := 0; i < b.N; i++ { Split(benchInputHard, "hello") } } func BenchmarkSplitNSingleByteSeparator(b *testing.B) { for i := 0; i < b.N; i++ { SplitN(benchInputHard, "/", 10) } } func BenchmarkSplitNMultiByteSeparator(b *testing.B) { for i := 0; i < b.N; i++ { SplitN(benchInputHard, "hello", 10) } } func BenchmarkRepeat(b *testing.B) { s := "0123456789" for _, n := range []int{5, 10} { for _, c := range []int{1, 2, 6} { b.Run(fmt.Sprintf("%dx%d", n, c), func(b *testing.B) { for i := 0; i < b.N; i++ { Repeat(s[:n], c) } }) } } } func BenchmarkIndexAnyASCII(b *testing.B) { x := Repeat("#", 2048) // Never matches set cs := "0123456789abcdefghijklmnopqrstuvwxyz0123456789abcdefghijklmnopqrstuvwxyz" for k := 1; k <= 2048; k <<= 4 { for j := 1; j <= 64; j <<= 1 { b.Run(fmt.Sprintf("%d:%d", k, j), func(b *testing.B) { for i := 0; i < b.N; i++ { IndexAny(x[:k], cs[:j]) } }) } } } func BenchmarkIndexAnyUTF8(b *testing.B) { x := Repeat("#", 2048) // Never matches set cs := "你好世界, hello world. 你好世界, hello world. 你好世界, hello world." for k := 1; k <= 2048; k <<= 4 { for j := 1; j <= 64; j <<= 1 { b.Run(fmt.Sprintf("%d:%d", k, j), func(b *testing.B) { for i := 0; i < b.N; i++ { IndexAny(x[:k], cs[:j]) } }) } } } func BenchmarkLastIndexAnyASCII(b *testing.B) { x := Repeat("#", 2048) // Never matches set cs := "0123456789abcdefghijklmnopqrstuvwxyz0123456789abcdefghijklmnopqrstuvwxyz" for k := 1; k <= 2048; k <<= 4 { for j := 1; j <= 64; j <<= 1 { b.Run(fmt.Sprintf("%d:%d", k, j), func(b *testing.B) { for i := 0; i < b.N; i++ { LastIndexAny(x[:k], cs[:j]) } }) } } } func BenchmarkLastIndexAnyUTF8(b *testing.B) { x := Repeat("#", 2048) // Never matches set cs := "你好世界, hello world. 你好世界, hello world. 你好世界, hello world." for k := 1; k <= 2048; k <<= 4 { for j := 1; j <= 64; j <<= 1 { b.Run(fmt.Sprintf("%d:%d", k, j), func(b *testing.B) { for i := 0; i < b.N; i++ { LastIndexAny(x[:k], cs[:j]) } }) } } } func BenchmarkTrimASCII(b *testing.B) { cs := "0123456789abcdef" for k := 1; k <= 4096; k <<= 4 { for j := 1; j <= 16; j <<= 1 { b.Run(fmt.Sprintf("%d:%d", k, j), func(b *testing.B) { x := Repeat(cs[:j], k) // Always matches set for i := 0; i < b.N; i++ { Trim(x[:k], cs[:j]) } }) } } } func BenchmarkTrimByte(b *testing.B) { x := " the quick brown fox " for i := 0; i < b.N; i++ { Trim(x, " ") } } func BenchmarkIndexPeriodic(b *testing.B) { key := "aa" for _, skip := range [...]int{2, 4, 8, 16, 32, 64} { b.Run(fmt.Sprintf("IndexPeriodic%d", skip), func(b *testing.B) { s := Repeat("a"+Repeat(" ", skip-1), 1<<16/skip) for i := 0; i < b.N; i++ { Index(s, key) } }) } } func BenchmarkJoin(b *testing.B) { vals := []string{"red", "yellow", "pink", "green", "purple", "orange", "blue"} for l := 0; l <= len(vals); l++ { b.Run(strconv.Itoa(l), func(b *testing.B) { b.ReportAllocs() vals := vals[:l] for i := 0; i < b.N; i++ { Join(vals, " and ") } }) } } func BenchmarkTrimSpace(b *testing.B) { tests := []struct{ name, input string }{ {"NoTrim", "typical"}, {"ASCII", " foo bar "}, {"SomeNonASCII", " \u2000\t\r\n x\t\t\r\r\ny\n \u3000 "}, {"JustNonASCII", "\u2000\u2000\u2000☺☺☺☺\u3000\u3000\u3000"}, } for _, test := range tests { b.Run(test.name, func(b *testing.B) { for i := 0; i < b.N; i++ { TrimSpace(test.input) } }) } } var stringSink string func BenchmarkReplaceAll(b *testing.B) { b.ReportAllocs() for i := 0; i < b.N; i++ { stringSink = ReplaceAll("banana", "a", "<>") } }