diff options
Diffstat (limited to 'src/strings')
| -rw-r--r-- | src/strings/strings.go | 44 | ||||
| -rw-r--r-- | src/strings/strings_test.go | 54 |
2 files changed, 86 insertions, 12 deletions
diff --git a/src/strings/strings.go b/src/strings/strings.go index 7fb05b7d0e..2789f5fb25 100644 --- a/src/strings/strings.go +++ b/src/strings/strings.go @@ -143,6 +143,14 @@ func IndexAny(s, chars string) int { // Avoid scanning all of s. return -1 } + if len(chars) == 1 { + // Avoid scanning all of s. + r := rune(chars[0]) + if r >= utf8.RuneSelf { + r = utf8.RuneError + } + return IndexRune(s, r) + } if len(s) > 8 { if as, isASCII := makeASCIISet(chars); isASCII { for i := 0; i < len(s); i++ { @@ -154,10 +162,8 @@ func IndexAny(s, chars string) int { } } for i, c := range s { - for _, m := range chars { - if c == m { - return i - } + if IndexRune(chars, c) >= 0 { + return i } } return -1 @@ -171,6 +177,16 @@ func LastIndexAny(s, chars string) int { // Avoid scanning all of s. return -1 } + if len(s) == 1 { + rc := rune(s[0]) + if rc >= utf8.RuneSelf { + rc = utf8.RuneError + } + if IndexRune(chars, rc) >= 0 { + return 0 + } + return -1 + } if len(s) > 8 { if as, isASCII := makeASCIISet(chars); isASCII { for i := len(s) - 1; i >= 0; i-- { @@ -181,13 +197,25 @@ func LastIndexAny(s, chars string) int { return -1 } } + if len(chars) == 1 { + rc := rune(chars[0]) + if rc >= utf8.RuneSelf { + rc = utf8.RuneError + } + for i := len(s); i > 0; { + r, size := utf8.DecodeLastRuneInString(s[:i]) + i -= size + if rc == r { + return i + } + } + return -1 + } for i := len(s); i > 0; { r, size := utf8.DecodeLastRuneInString(s[:i]) i -= size - for _, c := range chars { - if r == c { - return i - } + if IndexRune(chars, r) >= 0 { + return i } } return -1 diff --git a/src/strings/strings_test.go b/src/strings/strings_test.go index 984fecfa8d..c01c4dabc5 100644 --- a/src/strings/strings_test.go +++ b/src/strings/strings_test.go @@ -153,6 +153,7 @@ var indexAnyTests = []IndexTest{ {"", "abc", -1}, {"a", "", -1}, {"a", "a", 0}, + {"\x80", "\xffb", 0}, {"aaa", "a", 0}, {"abc", "xyz", -1}, {"abc", "xcz", 2}, @@ -163,6 +164,7 @@ var indexAnyTests = []IndexTest{ {dots + dots + dots, " ", -1}, {"012abcba210", "\xffb", 4}, {"012\x80bcb\x80210", "\xffb", 3}, + {"0123456\xcf\x80abc", "\xcfb\x80", 10}, } var lastIndexAnyTests = []IndexTest{ @@ -171,6 +173,7 @@ var lastIndexAnyTests = []IndexTest{ {"", "abc", -1}, {"a", "", -1}, {"a", "a", 0}, + {"\x80", "\xffb", 0}, {"aaa", "a", 2}, {"abc", "xyz", -1}, {"abc", "ab", 1}, @@ -181,6 +184,7 @@ var lastIndexAnyTests = []IndexTest{ {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 @@ -1787,10 +1791,24 @@ func BenchmarkRepeat(b *testing.B) { } func BenchmarkIndexAnyASCII(b *testing.B) { - x := Repeat("#", 4096) // Never matches set - cs := "0123456789abcdef" - for k := 1; k <= 4096; k <<= 4 { - for j := 1; j <= 16; j <<= 1 { + 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]) @@ -1800,6 +1818,34 @@ func BenchmarkIndexAnyASCII(b *testing.B) { } } +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 { |