1
|
|
|
<?php declare(strict_types = 1); |
2
|
|
|
|
3
|
|
|
namespace Apicart\FQL\Tokenizer; |
4
|
|
|
|
5
|
|
|
use Apicart\FQL\Token\Token\GroupBegin; |
6
|
|
|
use Apicart\FQL\Value\Token; |
7
|
|
|
use RuntimeException; |
8
|
|
|
|
9
|
|
|
abstract class AbstractTokenExtractor |
10
|
|
|
{ |
11
|
|
|
|
12
|
|
|
/** |
13
|
|
|
* Return the token at the given $position of the $string. |
14
|
|
|
* |
15
|
|
|
* @throws RuntimeException On PCRE regex error |
16
|
|
|
*/ |
17
|
507 |
|
final public function extract(string $string, int $position): Token |
18
|
|
|
{ |
19
|
507 |
|
$byteOffset = $this->getByteOffset($string, $position); |
20
|
507 |
|
foreach ($this->getExpressionTypeMap() as $expression => $type) { |
21
|
507 |
|
$success = preg_match($expression, $string, $matches, 0, $byteOffset); |
22
|
507 |
|
if ($success === false) { |
23
|
1 |
|
throw new RuntimeException('PCRE regex error code: ' . preg_last_error()); |
24
|
|
|
} |
25
|
506 |
|
if ($success === 0) { |
26
|
502 |
|
continue; |
27
|
|
|
} |
28
|
506 |
|
if (isset($matches['domain'])) { |
29
|
|
|
$matches['domain'] = trim($matches['domain'], "'"); |
30
|
19 |
|
} |
31
|
|
|
|
32
|
|
|
return $this->createToken($type, $position, $matches); |
33
|
|
|
} |
34
|
|
|
return new Token(Tokenizer::TOKEN_BAILOUT, mb_substr($string, $position, 1), $position); |
35
|
|
|
} |
36
|
|
|
|
37
|
|
|
|
38
|
|
|
/** |
39
|
|
|
* Return a map of regular expressions to token types. |
40
|
|
|
* |
41
|
|
|
* The returned map must be an array where key is a regular expression |
42
|
|
|
* and value is a corresponding token type. Regular expression must define |
43
|
|
|
* named capturing group 'lexeme' that identifies part of the input string |
44
|
|
|
* recognized as token. |
45
|
|
|
*/ |
46
|
|
|
abstract protected function getExpressionTypeMap(): array; |
47
|
|
|
|
48
|
|
|
|
49
|
|
|
/** |
50
|
|
|
* Create a term type token by the given parameters. |
51
|
|
|
* |
52
|
|
|
* @throw RuntimeException If token could not be created from the given $matches data |
53
|
|
|
*/ |
54
|
|
|
abstract protected function createTermToken(int $position, array $data): Token; |
55
|
|
|
|
56
|
74 |
|
|
57
|
|
|
/** |
58
|
74 |
|
* Create an instance of Group token by the given parameters. |
59
|
|
|
*/ |
60
|
|
|
protected function createGroupBeginToken(int $position, array $data): GroupBegin |
61
|
|
|
{ |
62
|
506 |
|
return new GroupBegin($data['lexeme'], $position, $data['delimiter'], $data['domain']); |
63
|
|
|
} |
64
|
506 |
|
|
65
|
88 |
|
|
66
|
|
|
private function createToken(int $type, int $position, array $data): Token |
67
|
506 |
|
{ |
68
|
420 |
|
if ($type === Tokenizer::TOKEN_GROUP_BEGIN) { |
69
|
|
|
return $this->createGroupBeginToken($position, $data); |
70
|
293 |
|
} |
71
|
|
|
if ($type === Tokenizer::TOKEN_TERM) { |
72
|
|
|
return $this->createTermToken($position, $data); |
73
|
|
|
} |
74
|
|
|
return new Token($type, $data['lexeme'], $position); |
75
|
|
|
} |
76
|
|
|
|
77
|
|
|
|
78
|
|
|
/** |
79
|
507 |
|
* Return the offset of the given $position in the input $string, in bytes. |
80
|
|
|
* |
81
|
507 |
|
* Offset in bytes is needed for preg_match $offset parameter. |
82
|
|
|
*/ |
83
|
|
|
private function getByteOffset(string $string, int $position): int |
84
|
|
|
{ |
85
|
|
|
return strlen(mb_substr($string, 0, $position)); |
86
|
|
|
} |
87
|
|
|
|
88
|
|
|
} |
89
|
|
|
|