|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* This file is part of Railt package. |
|
4
|
|
|
* |
|
5
|
|
|
* For the full copyright and license information, please view the LICENSE |
|
6
|
|
|
* file that was distributed with this source code. |
|
7
|
|
|
*/ |
|
8
|
|
|
declare(strict_types=1); |
|
9
|
|
|
|
|
10
|
|
|
namespace Railt\Compiler\Grammar\PP2; |
|
11
|
|
|
|
|
12
|
|
|
use Railt\Lexer\Driver\NativeStateless; |
|
13
|
|
|
|
|
14
|
|
|
/** |
|
15
|
|
|
* Class Lexer |
|
16
|
|
|
*/ |
|
17
|
|
|
class Lexer extends NativeStateless |
|
18
|
|
|
{ |
|
19
|
|
|
/**@#+ |
|
20
|
|
|
* List of tokens used inside grammar files. |
|
21
|
|
|
*/ |
|
22
|
|
|
public const T_WHITESPACE = 'T_WHITESPACE'; |
|
23
|
|
|
public const T_COMMENT = 'T_COMMENT'; |
|
24
|
|
|
public const T_BLOCK_COMMENT = 'T_BLOCK_COMMENT'; |
|
25
|
|
|
public const T_PRAGMA = 'T_PRAGMA'; |
|
26
|
|
|
public const T_TOKEN = 'T_TOKEN'; |
|
27
|
|
|
public const T_SKIP = 'T_SKIP'; |
|
28
|
|
|
public const T_INCLUDE = 'T_INCLUDE'; |
|
29
|
|
|
public const T_NODE_DEFINITION = 'T_NODE_DEFINITION'; |
|
30
|
|
|
public const T_OR = 'T_OR'; |
|
31
|
|
|
public const T_ZERO_OR_ONE = 'T_ZERO_OR_ONE'; |
|
32
|
|
|
public const T_ONE_OR_MORE = 'T_ONE_OR_MORE'; |
|
33
|
|
|
public const T_ZERO_OR_MORE = 'T_ZERO_OR_MORE'; |
|
34
|
|
|
public const T_N_TO_M = 'T_N_TO_M'; |
|
35
|
|
|
public const T_ZERO_TO_M = 'T_ZERO_TO_M'; |
|
36
|
|
|
public const T_N_OR_MORE = 'T_N_OR_MORE'; |
|
37
|
|
|
public const T_EXACTLY_N = 'T_EXACTLY_N'; |
|
38
|
|
|
public const T_SKIPPED = 'T_SKIPPED'; |
|
39
|
|
|
public const T_KEPT = 'T_KEPT'; |
|
40
|
|
|
public const T_NAMED = 'T_NAMED'; |
|
41
|
|
|
public const T_NODE = 'T_NODE'; |
|
42
|
|
|
public const T_GROUP_OPEN = 'T_GROUP_OPEN'; |
|
43
|
|
|
public const T_GROUP_CLOSE = 'T_GROUP_CLOSE'; |
|
44
|
|
|
/**#@-*/ |
|
45
|
|
|
|
|
46
|
|
|
/** |
|
47
|
|
|
* @var array|string[] Tokens list |
|
48
|
|
|
*/ |
|
49
|
|
|
private const TOKENS_LIST = [ |
|
50
|
|
|
self::T_WHITESPACE => '\s+', |
|
51
|
|
|
self::T_COMMENT => '//[^\\n]*', |
|
52
|
|
|
self::T_BLOCK_COMMENT => '/\\*.*?\\*/', |
|
53
|
|
|
self::T_PRAGMA => '%pragma\h+([\w\.]+)\h+(.+?)\s+', |
|
54
|
|
|
self::T_TOKEN => '%token\h+(\w+)\h+(.+?)(?:\s*->\h*\\$(\d+)\h*)?\n', |
|
55
|
|
|
self::T_SKIP => '%skip\h+(\w+)\h+(.+?)\s+', |
|
56
|
|
|
self::T_INCLUDE => '%include\h+(.+?)\s+', |
|
57
|
|
|
self::T_NODE_DEFINITION => '(#?\w+)(?:\s*->\h*(.+?)\h*)?\s*:', |
|
58
|
|
|
self::T_OR => '\\|', |
|
59
|
|
|
self::T_ZERO_OR_ONE => '\\?', |
|
60
|
|
|
self::T_ONE_OR_MORE => '\\+', |
|
61
|
|
|
self::T_ZERO_OR_MORE => '\\*', |
|
62
|
|
|
self::T_N_TO_M => '{\h*(\d+),\h*(\d+)\h*}', |
|
63
|
|
|
self::T_ZERO_TO_M => '{\h*,\h*(\d+)\h*}', |
|
64
|
|
|
self::T_N_OR_MORE => '{\h*(\d+)\h*,\h*}', |
|
65
|
|
|
self::T_EXACTLY_N => '{(\d+)}', |
|
66
|
|
|
self::T_SKIPPED => '::(\w+)::', |
|
67
|
|
|
self::T_KEPT => '<(\w+)>', |
|
68
|
|
|
self::T_NAMED => '(\w+)\\(\\)', |
|
69
|
|
|
self::T_NODE => '#(\w+)', |
|
70
|
|
|
self::T_GROUP_OPEN => '\\(', |
|
71
|
|
|
self::T_GROUP_CLOSE => '\\)', |
|
72
|
|
|
]; |
|
73
|
|
|
|
|
74
|
|
|
/** |
|
75
|
|
|
* A list of skipped tokens |
|
76
|
|
|
*/ |
|
77
|
|
|
private const TOKENS_SKIP = [ |
|
78
|
|
|
self::T_WHITESPACE, |
|
79
|
|
|
self::T_COMMENT, |
|
80
|
|
|
self::T_BLOCK_COMMENT, |
|
81
|
|
|
]; |
|
82
|
|
|
|
|
83
|
|
|
/** |
|
84
|
|
|
* Lexer constructor. |
|
85
|
|
|
*/ |
|
86
|
|
|
public function __construct() |
|
87
|
|
|
{ |
|
88
|
|
|
parent::__construct(); |
|
89
|
|
|
|
|
90
|
|
|
foreach (self::TOKENS_LIST as $name => $pcre) { |
|
91
|
|
|
$this->add($name, $pcre, \in_array($name, self::TOKENS_SKIP, true)); |
|
92
|
|
|
} |
|
93
|
|
|
} |
|
94
|
|
|
} |
|
95
|
|
|
|
railt /
compiler
