// 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"
"io"
"math/rand"
"reflect"
. "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},
{"xx01234567"[:9], "01234567", -1},
{"", "0123456789", -1},
{"3214567844", "0123456789", -1},
{"0123456789", "0123456789", 0},
{"x0123456789", "0123456789", 1},
{"xyz0123456789"[:12], "0123456789", -1},
{"x01234567x89", "0123456789", -1},
{"", "0123456789012345", -1},
{"3214567889012345", "0123456789012345", -1},
{"0123456789012345", "0123456789012345", 0},
{"x0123456789012345", "0123456789012345", 1},
{"", "01234567890123456789", -1},
{"32145678890123456789", "01234567890123456789", -1},
{"01234567890123456789", "01234567890123456789", 0},
{"x01234567890123456789", "01234567890123456789", 1},
{"xyz01234567890123456789"[:22], "01234567890123456789", -1},
{"", "0123456789012345678901234567890", -1},
{"321456788901234567890123456789012345678911", "0123456789012345678901234567890", -1},
{"0123456789012345678901234567890", "0123456789012345678901234567890", 0},
{"x0123456789012345678901234567890", "0123456789012345678901234567890", 1},
{"xyz0123456789012345678901234567890"[:33], "0123456789012345678901234567890", -1},
{"", "01234567890123456789012345678901", -1},
{"32145678890123456789012345678901234567890211", "01234567890123456789012345678901", -1},
{"01234567890123456789012345678901", "01234567890123456789012345678901", 0},
{"x01234567890123456789012345678901", "01234567890123456789012345678901", 1},
{"xyz01234567890123456789012345678901"[:34], "01234567890123456789012345678901", -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},
{"aaa", "a", 0},
{"abc", "xyz", -1},
{"abc", "xcz", 2},
{"a☺b☻c☹d", "uvw☻xyz", 2 + len("☺")},
{"aRegExp*", ".(|)*+?^$[]", 7},
{dots + dots + dots, " ", -1},
}
var lastIndexAnyTests = []IndexTest{
{"", "", -1},
{"", "a", -1},
{"", "abc", -1},
{"a", "", -1},
{"a", "a", 0},
{"aaa", "a", 2},
{"abc", "xyz", -1},
{"abc", "ab", 1},
{"a☺b☻c☹d", "uvw☻xyz", 2 + len("☺")},
{"a.RegExp*", ".(|)*+?^$[]", 8},
{dots + dots + dots, " ", -1},
}
// 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 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)
}
}
}
var indexRuneTests = []struct {
s string
rune rune
out int
}{
{"a A x", 'A', 2},
{"some_text=some_value", '=', 9},
{"☺a", 'a', 3},
{"a☻☺b", '☺', 4},
}
func TestIndexRune(t *testing.T) {
for _, test := range indexRuneTests {
if actual := IndexRune(test.s, test.rune); actual != test.out {
t.Errorf("IndexRune(%q,%d)= %v; want %v", test.s, test.rune, actual, test.out)
}
}
}
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, '☺')
}
}
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')
}
}
var explodetests = []struct {
s string
n int
a []string
}{
{"", -1, []string{}},
{abcd, 4, []string{"a", "b", "c", "d"}},
{faces, 3, []string{"☺", "☻", "☹"}},
{abcd, 2, []string{"a", "bcd"}},
}
func TestExplode(t *testing.T) {
for _, tt := range explodetests {
a := SplitN(tt.s, "", tt.n)
if !eq(a, tt.a) {
t.Errorf("explode(%q, %d) = %v; want %v", tt.s, tt.n, a, tt.a)
continue
}
s := Join(a, "")
if s != tt.s {
t.Errorf(`Join(explode(%q, %d), "") = %q`, tt.s, tt.n, s)
}
}
}
type SplitTest struct {
s string
sep string
n int
a []string
}
var splittests = []SplitTest{
{abcd, "a", 0, nil},
{abcd, "a", -1, []string{"", "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, []string{"1", "2"}},
{"123", "", 2, []string{"1", "23"}},
{"123", "", 17, []string{"1", "2", "3"}},
}
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{}},
{" 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{"™", "™"}},
{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{
{"", ""},
{"abc", "ABC"},
{"AbC123", "ABC123"},
{"azAZ09_", "AZAZ09_"},
{"\u0250\u0250\u0250\u0250\u0250", "\u2C6F\u2C6F\u2C6F\u2C6F\u2C6F"}, // grows one byte per char
}
var lowerTests = []StringTest{
{"", ""},
{"abc", "abc"},
{"AbC123", "abc123"},
{"azAZ09_", "azaz09_"},
{"\u2C6D\u2C6D\u2C6D\u2C6D\u2C6D", "\u0251\u0251\u0251\u0251\u0251"}, // shrinks one byte per char
}
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")
}
}
func TestToUpper(t *testing.T) { runStringTests(t, ToUpper, "ToUpper", upperTests) }
func TestToLower(t *testing.T) { runStringTests(t, ToLower, "ToLower", lowerTests) }
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"},
{"TrimRight", "abba", "a", "abb"},
{"Trim", " ", "", "",
"