github.com/karrick/go@v0.0.0-20170817181416-d5b0ec858b37/test/utf.go (about)

     1  // run
     2  
     3  // Copyright 2009 The Go Authors. All rights reserved.
     4  // Use of this source code is governed by a BSD-style
     5  // license that can be found in the LICENSE file.
     6  
     7  // Test UTF-8 in strings and character constants.
     8  
     9  package main
    10  
    11  import "unicode/utf8"
    12  
    13  func main() {
    14  	var chars [6]rune
    15  	chars[0] = 'a'
    16  	chars[1] = 'b'
    17  	chars[2] = 'c'
    18  	chars[3] = '\u65e5'
    19  	chars[4] = '\u672c'
    20  	chars[5] = '\u8a9e'
    21  	s := ""
    22  	for i := 0; i < 6; i++ {
    23  		s += string(chars[i])
    24  	}
    25  	var l = len(s)
    26  	for w, i, j := 0, 0, 0; i < l; i += w {
    27  		var r rune
    28  		r, w = utf8.DecodeRuneInString(s[i:len(s)])
    29  		if w == 0 {
    30  			panic("zero width in string")
    31  		}
    32  		if r != chars[j] {
    33  			panic("wrong value from string")
    34  		}
    35  		j++
    36  	}
    37  	// encoded as bytes:  'a' 'b' 'c' e6 97 a5 e6 9c ac e8 aa 9e
    38  	const L = 12
    39  	if L != l {
    40  		panic("wrong length constructing array")
    41  	}
    42  	a := make([]byte, L)
    43  	a[0] = 'a'
    44  	a[1] = 'b'
    45  	a[2] = 'c'
    46  	a[3] = 0xe6
    47  	a[4] = 0x97
    48  	a[5] = 0xa5
    49  	a[6] = 0xe6
    50  	a[7] = 0x9c
    51  	a[8] = 0xac
    52  	a[9] = 0xe8
    53  	a[10] = 0xaa
    54  	a[11] = 0x9e
    55  	for w, i, j := 0, 0, 0; i < L; i += w {
    56  		var r rune
    57  		r, w = utf8.DecodeRune(a[i:L])
    58  		if w == 0 {
    59  			panic("zero width in bytes")
    60  		}
    61  		if r != chars[j] {
    62  			panic("wrong value from bytes")
    63  		}
    64  		j++
    65  	}
    66  }