|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* @package s9e\RegexpBuilder |
|
5
|
|
|
* @copyright Copyright (c) 2016 The s9e Authors |
|
6
|
|
|
* @license http://www.opensource.org/licenses/mit-license.php The MIT License |
|
7
|
|
|
*/ |
|
8
|
|
|
namespace s9e\RegexpBuilder\Input; |
|
9
|
|
|
|
|
10
|
|
|
use InvalidArgumentException; |
|
11
|
|
|
|
|
12
|
|
|
class Utf8 extends BaseImplementation |
|
13
|
|
|
{ |
|
14
|
|
|
/** |
|
15
|
|
|
* @var bool Whether to use surrogates to represent higher codepoints |
|
16
|
|
|
*/ |
|
17
|
|
|
protected $useSurrogates; |
|
18
|
|
|
|
|
19
|
|
|
/** |
|
20
|
|
|
* {@inheritdoc} |
|
21
|
|
|
*/ |
|
22
|
9 |
|
public function __construct(array $options = []) |
|
23
|
|
|
{ |
|
24
|
9 |
|
$this->useSurrogates = !empty($options['useSurrogates']); |
|
25
|
9 |
|
} |
|
26
|
|
|
|
|
27
|
|
|
/** |
|
28
|
|
|
* {@inheritdoc} |
|
29
|
|
|
*/ |
|
30
|
9 |
|
public function split($string) |
|
31
|
|
|
{ |
|
32
|
9 |
|
if (preg_match_all('(.)us', $string, $matches) === false) |
|
33
|
|
|
{ |
|
34
|
1 |
|
throw new InvalidArgumentException('Invalid UTF-8 string'); |
|
35
|
|
|
} |
|
36
|
|
|
|
|
37
|
8 |
|
return ($this->useSurrogates) ? $this->charsToCodepointsWithSurrogates($matches[0]) : $this->charsToCodepoints($matches[0]); |
|
38
|
|
|
} |
|
39
|
|
|
|
|
40
|
|
|
/** |
|
41
|
|
|
* Convert a list of UTF-8 characters into a list of Unicode codepoint |
|
42
|
|
|
* |
|
43
|
|
|
* @param string[] $chars |
|
44
|
|
|
* @return integer[] |
|
45
|
|
|
*/ |
|
46
|
6 |
|
protected function charsToCodepoints(array $chars) |
|
47
|
|
|
{ |
|
48
|
6 |
|
return array_map([$this, 'cp'], $chars); |
|
49
|
|
|
} |
|
50
|
|
|
|
|
51
|
|
|
/** |
|
52
|
|
|
* Convert a list of UTF-8 characters into a list of Unicode codepoint with surrogates |
|
53
|
|
|
* |
|
54
|
|
|
* @param string[] $chars |
|
55
|
|
|
* @return integer[] |
|
56
|
|
|
*/ |
|
57
|
2 |
|
protected function charsToCodepointsWithSurrogates(array $chars) |
|
58
|
|
|
{ |
|
59
|
2 |
|
$codepoints = []; |
|
60
|
2 |
|
foreach ($chars as $char) |
|
61
|
|
|
{ |
|
62
|
2 |
|
$cp = $this->cp($char); |
|
63
|
2 |
|
if ($cp < 0x10000) |
|
64
|
|
|
{ |
|
65
|
2 |
|
$codepoints[] = $cp; |
|
66
|
|
|
} |
|
67
|
|
|
else |
|
68
|
|
|
{ |
|
69
|
1 |
|
$codepoints[] = 0xD7C0 + ($cp >> 10); |
|
70
|
2 |
|
$codepoints[] = 0xDC00 + ($cp & 0x3FF); |
|
71
|
|
|
} |
|
72
|
|
|
} |
|
73
|
|
|
|
|
74
|
2 |
|
return $codepoints; |
|
75
|
|
|
} |
|
76
|
|
|
|
|
77
|
|
|
/** |
|
78
|
|
|
* Compute and return the Unicode codepoint for given UTF-8 char |
|
79
|
|
|
* |
|
80
|
|
|
* @param string $char UTF-8 char |
|
81
|
|
|
* @return integer |
|
82
|
|
|
*/ |
|
83
|
7 |
|
protected function cp($char) |
|
84
|
|
|
{ |
|
85
|
7 |
|
$cp = ord($char[0]); |
|
86
|
7 |
|
if ($cp >= 0xF0) |
|
87
|
|
|
{ |
|
88
|
3 |
|
$cp = ($cp << 18) + (ord($char[1]) << 12) + (ord($char[2]) << 6) + ord($char[3]) - 0x3C82080; |
|
89
|
|
|
} |
|
90
|
7 |
|
elseif ($cp >= 0xE0) |
|
91
|
|
|
{ |
|
92
|
1 |
|
$cp = ($cp << 12) + (ord($char[1]) << 6) + ord($char[2]) - 0xE2080; |
|
93
|
|
|
} |
|
94
|
6 |
|
elseif ($cp >= 0xC0) |
|
95
|
|
|
{ |
|
96
|
2 |
|
$cp = ($cp << 6) + ord($char[1]) - 0x3080; |
|
97
|
|
|
} |
|
98
|
|
|
|
|
99
|
7 |
|
return $cp; |
|
100
|
|
|
} |
|
101
|
|
|
} |
s9e /
RegexpBuilder
