|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* For licensing information, please see the LICENSE file accompanied with this file. |
|
4
|
|
|
* |
|
5
|
|
|
* @author Gerard van Helden <[email protected]> |
|
6
|
|
|
* @copyright 2012 Gerard van Helden <http://melp.nl> |
|
7
|
|
|
*/ |
|
8
|
|
|
|
|
9
|
|
|
namespace Zicht\Tool\Script\Tokenizer; |
|
10
|
|
|
|
|
11
|
|
|
use Zicht\Tool\Script\Token; |
|
12
|
|
|
use Zicht\Tool\Script\TokenizerInterface; |
|
13
|
|
|
|
|
14
|
|
|
/** |
|
15
|
|
|
* Tokenizer for expressions |
|
16
|
|
|
*/ |
|
17
|
|
|
class Expression implements TokenizerInterface |
|
18
|
|
|
{ |
|
19
|
|
|
/** |
|
20
|
|
|
* @{inheritDoc} |
|
21
|
|
|
*/ |
|
22
|
22 |
|
public function getTokens($string, &$needle = 0) |
|
23
|
|
|
{ |
|
24
|
22 |
|
$depth = 1; |
|
25
|
22 |
|
$ret = array(); |
|
26
|
22 |
|
$len = strlen($string); |
|
27
|
22 |
|
while ($needle < $len) { |
|
28
|
22 |
|
$substr = substr($string, $needle); |
|
29
|
22 |
|
$before = $needle; |
|
30
|
|
|
|
|
31
|
22 |
|
if (preg_match('/^(==|=~|<=?|>=?|!=?|\?|:|\|\||&&|xor\b|or\b|and\b|\.|\[|\]|\(|\)|\+|-|\/|\*)/', $substr, $m)) { |
|
32
|
21 |
|
if ($m[0] == ')') { |
|
33
|
18 |
|
$depth--; |
|
34
|
18 |
|
if ($depth == 0) { |
|
35
|
15 |
|
$ret[] = new Token(Token::EXPR_END, ')'); |
|
36
|
15 |
|
$needle++; |
|
37
|
15 |
|
break; |
|
38
|
|
|
} |
|
39
|
13 |
|
} elseif ($m[0] === '(') { |
|
40
|
6 |
|
$depth++; |
|
41
|
6 |
|
} |
|
42
|
13 |
|
$ret[] = new Token(Token::OPERATOR, $m[0]); |
|
43
|
13 |
|
$needle += strlen($m[0]); |
|
44
|
22 |
View Code Duplication |
} elseif (preg_match('/^(true|false|in|as|null)\b/', $substr, $m)) { |
|
|
|
|
|
|
45
|
|
|
$ret[] = new Token(Token::KEYWORD, $m[0]); |
|
46
|
|
|
$needle += strlen($m[0]); |
|
47
|
22 |
|
} elseif (preg_match('/^[a-z_][\w]*/i', $substr, $m)) { |
|
48
|
16 |
|
$ret[] = new Token(Token::IDENTIFIER, $m[0]); |
|
49
|
16 |
|
$needle += strlen($m[0]); |
|
50
|
22 |
View Code Duplication |
} elseif (preg_match('/^\s+/', $substr, $m)) { |
|
|
|
|
|
|
51
|
6 |
|
$ret[] = new Token(Token::WHITESPACE, $m[0]); |
|
52
|
6 |
|
$needle += strlen($m[0]); |
|
53
|
11 |
|
} elseif (preg_match('/^([0-9]*.)?[0-9]+/', $substr, $m)) { |
|
54
|
4 |
|
$ret[] = new Token(Token::NUMBER, $m[0]); |
|
55
|
4 |
|
$needle += strlen($m[0]); |
|
56
|
9 |
|
} elseif (preg_match('/^[\?,]/', $substr, $m)) { |
|
57
|
1 |
|
$ret[] = new Token($m[0]); |
|
58
|
1 |
|
$needle++; |
|
59
|
6 |
|
} elseif ($string{$needle} == '"') { |
|
60
|
5 |
|
$strData = ''; |
|
61
|
|
|
|
|
62
|
5 |
|
$escape = false; |
|
63
|
5 |
|
for ($j = $needle + 1; $j < $len; $j++) { |
|
64
|
5 |
|
$ch = $string{$j}; |
|
65
|
|
|
|
|
66
|
5 |
|
if ($ch == '\\') { |
|
67
|
2 |
|
$escape = true; |
|
68
|
5 |
|
} elseif ($ch == '"') { |
|
69
|
5 |
|
if ($escape) { |
|
70
|
|
|
$escape = false; |
|
71
|
|
|
} else { |
|
72
|
5 |
|
$j++; |
|
73
|
5 |
|
break; |
|
74
|
|
|
} |
|
75
|
|
|
} else { |
|
76
|
5 |
|
if ($escape) { |
|
77
|
|
|
switch ($ch) { |
|
78
|
2 |
|
case 'n': |
|
79
|
|
|
$strData .= "\n"; |
|
80
|
|
|
break; |
|
81
|
2 |
|
case '\\': |
|
82
|
|
|
$strData .= "\\"; |
|
83
|
|
|
break; |
|
84
|
2 |
|
default: |
|
85
|
2 |
|
$strData .= '\\' . $ch; |
|
86
|
2 |
|
break; |
|
87
|
2 |
|
} |
|
88
|
2 |
|
$escape = false; |
|
89
|
2 |
|
} else { |
|
90
|
5 |
|
$strData .= $ch; |
|
91
|
|
|
} |
|
92
|
|
|
} |
|
93
|
5 |
|
} |
|
94
|
5 |
|
$ret[] = new Token(Token::STRING, $strData); |
|
95
|
5 |
|
$needle = $j; |
|
96
|
5 |
|
} |
|
97
|
22 |
|
if ($before === $needle) { |
|
98
|
|
|
// safety net. |
|
99
|
|
|
throw new \UnexpectedValueException( |
|
100
|
|
|
"Unexpected input near token {$string{$needle}}, unsupported character ($string)" |
|
101
|
|
|
); |
|
102
|
|
|
} |
|
103
|
22 |
|
} |
|
104
|
22 |
|
return $ret; |
|
105
|
|
|
} |
|
106
|
|
|
} |
|
107
|
|
|
|
zicht /
z
Loading history...
Duplicated code is one of the most pungent code smells. If you need to duplicate the same code in three or more different places, we strongly encourage you to look into extracting the code into a single class or operation.
You can also find more detailed suggestions in the “Code” section of your repository.