|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace SilverStripe\View\Parsers; |
|
4
|
|
|
|
|
5
|
|
|
use SilverStripe\Core\Object; |
|
6
|
|
|
|
|
7
|
|
|
/** |
|
8
|
|
|
* Support class for converting unicode strings into a suitable 7-bit ASCII equivalent. |
|
9
|
|
|
* |
|
10
|
|
|
* Usage: |
|
11
|
|
|
* |
|
12
|
|
|
* <code> |
|
13
|
|
|
* $tr = new Transliterator(); |
|
14
|
|
|
* $ascii = $tr->toASCII($unicode); |
|
15
|
|
|
* </code> |
|
16
|
|
|
*/ |
|
17
|
|
|
class Transliterator extends Object |
|
18
|
|
|
{ |
|
19
|
|
|
/** |
|
20
|
|
|
* @config |
|
21
|
|
|
* @var boolean Allow the use of iconv() to perform transliteration. Set to false to disable. |
|
22
|
|
|
* Even if this variable is true, iconv() won't be used if it's not installed. |
|
23
|
|
|
*/ |
|
24
|
|
|
private static $use_iconv = false; |
|
25
|
|
|
|
|
26
|
|
|
/** |
|
27
|
|
|
* Convert the given utf8 string to a safe ASCII source |
|
28
|
|
|
* |
|
29
|
|
|
* @param string $source |
|
30
|
|
|
* @return string |
|
31
|
|
|
*/ |
|
32
|
|
|
public function toASCII($source) |
|
33
|
|
|
{ |
|
34
|
|
|
if (function_exists('iconv') && $this->config()->use_iconv) { |
|
35
|
|
|
return $this->useIconv($source); |
|
36
|
|
|
} else { |
|
37
|
|
|
return $this->useStrTr($source); |
|
38
|
|
|
} |
|
39
|
|
|
} |
|
40
|
|
|
|
|
41
|
|
|
/** |
|
42
|
|
|
* Transliteration using strtr() and a lookup table |
|
43
|
|
|
* |
|
44
|
|
|
* @param string $source |
|
45
|
|
|
* @return string |
|
46
|
|
|
*/ |
|
47
|
|
|
protected function useStrTr($source) |
|
48
|
|
|
{ |
|
49
|
|
|
$table = array( |
|
50
|
|
|
'Š'=>'S', 'š'=>'s', 'Đ'=>'Dj', 'đ'=>'dj', 'Ž'=>'Z', 'ž'=>'z', 'Č'=>'C', 'č'=>'c', 'Ć'=>'C', 'ć'=>'c', |
|
51
|
|
|
'À'=>'A', 'Á'=>'A', 'Â'=>'A', 'Ã'=>'A', 'Ä'=>'Ae', 'Å'=>'A', 'Æ'=>'A', 'Ç'=>'C', 'È'=>'E', 'É'=>'E', |
|
52
|
|
|
'Ê'=>'E', 'Ë'=>'E', 'Ì'=>'I', 'Í'=>'I', 'Î'=>'I', 'Ï'=>'I', 'Ñ'=>'N', 'Ò'=>'O', 'Ó'=>'O', 'Ô'=>'O', |
|
53
|
|
|
'Õ'=>'O', 'Ö'=>'Oe', 'Ø'=>'O', 'Ù'=>'U', 'Ú'=>'U', 'Û'=>'U', 'Ü'=>'Ue', 'Ý'=>'Y', 'Þ'=>'B', 'ß'=>'ss', |
|
54
|
|
|
'à'=>'a', 'á'=>'a', 'â'=>'a', 'ã'=>'a', 'ä'=>'ae', 'å'=>'a', 'æ'=>'ae', 'ç'=>'c', 'è'=>'e', 'é'=>'e', |
|
55
|
|
|
'ê'=>'e', 'ë'=>'e', 'ì'=>'i', 'í'=>'i', 'î'=>'i', 'ï'=>'i', 'ð'=>'o', 'ñ'=>'n', 'ò'=>'o', 'ó'=>'o', |
|
56
|
|
|
'ô'=>'o', 'õ'=>'o', 'ö'=>'oe', 'ø'=>'o', 'ù'=>'u', 'ú'=>'u', 'û'=>'u', 'ü'=>'ue', 'ý'=>'y', 'ý'=>'y', |
|
57
|
|
|
'þ'=>'b', 'ÿ'=>'y', 'Ŕ'=>'R', 'ŕ'=>'r', |
|
58
|
|
|
'Ā'=>'A', 'ā'=>'a', 'Ē'=>'E', 'ē'=>'e', 'Ī'=>'I', 'ī'=>'i', 'Ō'=>'O', 'ō'=>'o', 'Ū'=>'U', 'ū'=>'u', |
|
59
|
|
|
'œ'=>'oe', 'ß'=>'ss', 'ij'=>'ij', 'ą'=>'a','ę'=>'e', 'ė'=>'e', 'į'=>'i','ų'=>'u','ū'=>'u', 'Ą'=>'A', |
|
60
|
|
|
'Ę'=>'E', 'Ė'=>'E', 'Į'=>'I','Ų'=>'U','Ū'=>'U', |
|
61
|
|
|
"ľ"=>"l", "Ľ"=>"L", "ť"=>"t", "Ť"=>"T", "ů"=>"u", "Ů"=>"U", |
|
62
|
|
|
'ł'=>'l', 'Ł'=>'L', 'ń'=>'n', 'Ń'=>'N', 'ś'=>'s', 'Ś'=>'S', 'ź'=>'z', 'Ź'=>'Z', 'ż'=>'z', 'Ż'=>'Z', |
|
63
|
|
|
'а'=>"a",'б'=>"b",'в'=>"v",'г'=>"g",'д'=>"d",'е'=>"e",'ё'=>"yo",'ж'=>"zh",'з'=>"z",'и'=>"i", |
|
64
|
|
|
'й'=>"y",'к'=>"k",'л'=>"l",'м'=>"m",'н'=>"n",'о'=>"o",'п'=>"p",'р'=>"r",'с'=>"s",'т'=>"t", |
|
65
|
|
|
'у'=>"u",'ф'=>"f",'х'=>"kh",'ц'=>"ts",'ч'=>"ch",'ш'=>"sh",'щ'=>"shch",'ы'=>"y",'э'=>"e",'ю'=>"yu", |
|
66
|
|
|
'я'=>"ya", |
|
67
|
|
|
'А'=>"A",'Б'=>"B",'В'=>"V",'Г'=>"G",'Д'=>"D",'Е'=>"E",'Ё'=>"YO",'Ж'=>"ZH",'З'=>"Z",'И'=>"I", |
|
68
|
|
|
'Й'=>"Y",'К'=>"K",'Л'=>"L",'М'=>"M",'Н'=>"N",'О'=>"O",'П'=>"P",'Р'=>"R",'С'=>"S",'Т'=>"T", |
|
69
|
|
|
'У'=>"U",'Ф'=>"F",'Х'=>"KH",'Ц'=>"TS",'Ч'=>"CH",'Ш'=>"SH",'Щ'=>"SHCH",'Ы'=>"Y",'Э'=>"E",'Ю'=>"YU", |
|
70
|
|
|
'Я'=>"YA", |
|
71
|
|
|
'α'=>'a', 'Α'=>'A', 'ά'=>'a', 'Ά'=>'A', 'β'=>'v', 'Β'=>'V', 'γ'=>'g', 'Γ'=>'G', 'δ'=>'d', 'Δ'=>'D', |
|
72
|
|
|
'ε'=>'e', 'ϵ'=>'e', 'Ε'=>'E', 'έ'=>'e', 'Έ'=>'E', 'ζ'=>'z', 'Ζ'=>'Z', 'η'=>'i', 'Η'=>'I', |
|
73
|
|
|
'θ'=>'th', 'ϑ' => 'th', 'Θ'=>'TH', 'ι'=>'i', 'Ι'=>'I', 'ί'=>'i', 'Ί'=>'I', 'κ'=>'k', 'ϰ'=>'k', 'Κ'=>'K', |
|
74
|
|
|
'λ'=>'l', 'Λ'=>'L', 'μ'=>'m', 'Μ'=>'M', 'ν'=>'n', 'Ν'=>'N', 'ή'=>'n', 'Ή'=>'N', 'ἠ'=>'n', 'Ἠ'=>'N', |
|
75
|
|
|
'ο'=>'o', 'Ο'=>'O', 'ό'=>'o', 'Ό'=>'O', 'π'=>'p', 'Π'=>'P', 'ρ'=>'r', 'ϱ'=>'r', 'Ρ'=>'R', 'ῤ'=>'rh', |
|
76
|
|
|
'σ'=>'s', 'ς'=>'s', 'Σ'=>'S', 'τ'=>'t', 'Τ'=>'T', 'υ'=>'y', 'Υ'=>'Y', 'ύ'=>'y', 'Ύ'=>'Y', 'ὐ'=>'y', |
|
77
|
|
|
'φ'=>'f', 'ϕ'=>'f', 'Φ'=>'F', 'χ'=>'ch', 'Χ'=>'CH', 'ψ'=>'ps', 'Ψ'=>'PS', 'ξ'=>'x', 'Ξ'=>'X', |
|
78
|
|
|
'ω'=>'w', 'Ω'=>'W', 'ώ'=>'o', 'Ώ'=>'O', 'ὠ'=>'o', 'Ὠ'=>'O', |
|
79
|
|
|
); |
|
80
|
|
|
|
|
81
|
|
|
return strtr($source, $table); |
|
82
|
|
|
} |
|
83
|
|
|
|
|
84
|
|
|
/** |
|
85
|
|
|
* Transliteration using iconv() |
|
86
|
|
|
* |
|
87
|
|
|
* @param string $source |
|
88
|
|
|
* @return string |
|
89
|
|
|
*/ |
|
90
|
|
|
protected function useIconv($source) |
|
91
|
|
|
{ |
|
92
|
|
|
return iconv("utf-8", "us-ascii//IGNORE//TRANSLIT", $source); |
|
93
|
|
|
} |
|
94
|
|
|
} |
|
95
|
|
|
|