1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
declare(strict_types=1); |
4
|
|
|
|
5
|
|
|
namespace Antlr\Antlr4\Runtime\Atn; |
6
|
|
|
|
7
|
|
|
use Antlr\Antlr4\Runtime\Atn\Actions\LexerAction; |
8
|
|
|
use Antlr\Antlr4\Runtime\Atn\Actions\LexerIndexedCustomAction; |
9
|
|
|
use Antlr\Antlr4\Runtime\CharStream; |
10
|
|
|
use Antlr\Antlr4\Runtime\Comparison\Equality; |
11
|
|
|
use Antlr\Antlr4\Runtime\Comparison\Equatable; |
12
|
|
|
use Antlr\Antlr4\Runtime\Comparison\Hasher; |
13
|
|
|
use Antlr\Antlr4\Runtime\Lexer; |
14
|
|
|
|
15
|
|
|
/** |
16
|
|
|
* Represents an executor for a sequence of lexer actions which traversed during |
17
|
|
|
* the matching operation of a lexer rule (token). |
18
|
|
|
* |
19
|
|
|
* The executor tracks position information for position-dependent lexer actions |
20
|
|
|
* efficiently, ensuring that actions appearing only at the end of the rule do |
21
|
|
|
* not cause bloating of the {@see DFA} created for the lexer. |
22
|
|
|
* |
23
|
|
|
* @author Sam Harwell |
24
|
|
|
*/ |
25
|
|
|
final class LexerActionExecutor implements Equatable |
26
|
|
|
{ |
27
|
|
|
/** @var array<LexerAction> */ |
28
|
|
|
private $lexerActions; |
29
|
|
|
|
30
|
|
|
/** |
31
|
|
|
* Caches the result of {@see LexerActionExecutor::hashCode()} since |
32
|
|
|
* the hash code is an element of the performance-critical |
33
|
|
|
* {@see LexerATNConfig::hashCode()} operation. |
34
|
|
|
* |
35
|
|
|
* @var int|null |
36
|
|
|
*/ |
37
|
|
|
private $cachedHashCode; |
38
|
|
|
|
39
|
|
|
/** |
40
|
|
|
* @param array<LexerAction> $lexerActions |
41
|
|
|
*/ |
42
|
2 |
|
public function __construct(array $lexerActions) |
43
|
|
|
{ |
44
|
2 |
|
$this->lexerActions = $lexerActions; |
45
|
2 |
|
} |
46
|
|
|
|
47
|
|
|
/** |
48
|
|
|
* Creates a {@see LexerActionExecutor} which executes the actions for |
49
|
|
|
* the input `lexerActionExecutor` followed by a specified `lexerAction`. |
50
|
|
|
* |
51
|
|
|
* @param LexerActionExecutor|null $lexerActionExecutor The executor for actions |
52
|
|
|
* already traversed by |
53
|
|
|
* the lexer while matching |
54
|
|
|
* a token within a particular |
55
|
|
|
* {@see LexerATNConfig}. |
56
|
|
|
* If this is `null`, |
57
|
|
|
* the method behaves as |
58
|
|
|
* though it were an |
59
|
|
|
* empty executor. |
60
|
|
|
* @param LexerAction $lexerAction The lexer action to |
61
|
|
|
* execute after the |
62
|
|
|
* actions specified in |
63
|
|
|
* `lexerActionExecutor`. |
64
|
|
|
* |
65
|
|
|
* @return self A {@see LexerActionExecutor} for executing the combine actions |
66
|
|
|
* of `lexerActionExecutor` and `lexerAction`. |
67
|
|
|
*/ |
68
|
2 |
|
public static function append( |
69
|
|
|
?LexerActionExecutor $lexerActionExecutor, |
70
|
|
|
LexerAction $lexerAction |
71
|
|
|
) : self { |
72
|
2 |
|
if ($lexerActionExecutor === null) { |
73
|
2 |
|
return new LexerActionExecutor([$lexerAction]); |
74
|
|
|
} |
75
|
|
|
|
76
|
|
|
$lexerActions = \array_merge($lexerActionExecutor->lexerActions, [$lexerAction]); |
77
|
|
|
|
78
|
|
|
return new LexerActionExecutor($lexerActions); |
79
|
|
|
} |
80
|
|
|
|
81
|
|
|
/** |
82
|
|
|
* Creates a {@see LexerActionExecutor} which encodes the current offset |
83
|
|
|
* for position-dependent lexer actions. |
84
|
|
|
* |
85
|
|
|
* Normally, when the executor encounters lexer actions where |
86
|
|
|
* {@see LexerAction::isPositionDependent()} returns `true`, it calls |
87
|
|
|
* {@see IntStream::seek()} on the input {@see CharStream} to set the input |
88
|
|
|
* position to the <em>end</em> of the current token. This behavior provides |
89
|
|
|
* for efficient DFA representation of lexer actions which appear at the end |
90
|
|
|
* of a lexer rule, even when the lexer rule matches a variable number of |
91
|
|
|
* characters. |
92
|
|
|
* |
93
|
|
|
* Prior to traversing a match transition in the ATN, the current offset |
94
|
|
|
* from the token start index is assigned to all position-dependent lexer |
95
|
|
|
* actions which have not already been assigned a fixed offset. By storing |
96
|
|
|
* the offsets relative to the token start index, the DFA representation of |
97
|
|
|
* lexer actions which appear in the middle of tokens remains efficient due |
98
|
|
|
* to sharing among tokens of the same length, regardless of their absolute |
99
|
|
|
* position in the input stream. |
100
|
|
|
* |
101
|
|
|
* If the current executor already has offsets assigned to all |
102
|
|
|
* position-dependent lexer actions, the method returns `this`. |
103
|
|
|
* |
104
|
|
|
* @param int $offset The current offset to assign to all position-dependent |
105
|
|
|
* lexer actions which do not already have offsets assigned. |
106
|
|
|
* |
107
|
|
|
* @return self A {@see LexerActionExecutor} which stores input stream offsets |
108
|
|
|
* for all position-dependent lexer actions. |
109
|
|
|
*/ |
110
|
|
|
public function fixOffsetBeforeMatch(int $offset) : self |
111
|
|
|
{ |
112
|
|
|
$updatedLexerActions = null; |
113
|
|
|
|
114
|
|
|
for ($i = 0, $count = \count($this->lexerActions); $i < $count; $i++) { |
115
|
|
|
if ($this->lexerActions[$i]->isPositionDependent() |
116
|
|
|
&& !$this->lexerActions[$i] instanceof LexerIndexedCustomAction) { |
117
|
|
|
if ($updatedLexerActions === null) { |
118
|
|
|
$updatedLexerActions = \array_merge($this->lexerActions, []); |
119
|
|
|
} |
120
|
|
|
|
121
|
|
|
$updatedLexerActions[$i] = new LexerIndexedCustomAction($offset, $this->lexerActions[$i]); |
122
|
|
|
} |
123
|
|
|
} |
124
|
|
|
|
125
|
|
|
if ($updatedLexerActions === null) { |
|
|
|
|
126
|
|
|
return $this; |
127
|
|
|
} |
128
|
|
|
|
129
|
|
|
return new LexerActionExecutor($updatedLexerActions); |
130
|
|
|
} |
131
|
|
|
|
132
|
|
|
/** |
133
|
|
|
* Gets the lexer actions to be executed by this executor. |
134
|
|
|
* |
135
|
|
|
* @return array<LexerAction> The lexer actions to be executed by this executor. |
136
|
|
|
*/ |
137
|
|
|
public function getLexerActions() : array |
138
|
|
|
{ |
139
|
|
|
return $this->lexerActions; |
140
|
|
|
} |
141
|
|
|
|
142
|
|
|
/** |
143
|
|
|
* Execute the actions encapsulated by this executor within the context of a |
144
|
|
|
* particular {@see Lexer}. |
145
|
|
|
* |
146
|
|
|
* This method calls {@see IntStream::seek()} to set the position of the |
147
|
|
|
* `input` {@see CharStream} prior to calling {@see LexerAction::execute()} |
148
|
|
|
* on a position-dependent action. Before the method returns, the input |
149
|
|
|
* position will be restored to the same position it was in when the method |
150
|
|
|
* was invoked. |
151
|
|
|
* |
152
|
|
|
* @param Lexer $lexer The lexer instance. |
153
|
|
|
* @param CharStream $input The input stream which is the source for |
154
|
|
|
* the current token. When this method is called, |
155
|
|
|
* the current {@see IntStream::getIndex()} for |
156
|
|
|
* `input` should be the start of the following |
157
|
|
|
* token, i.e. 1 character past the end of the |
158
|
|
|
* current token. |
159
|
|
|
* @param int $startIndex The token start index. This value may be |
160
|
|
|
* passed to {@see IntStream::seek()} to set |
161
|
|
|
* the `input` position to the beginning |
162
|
|
|
* of the token. |
163
|
|
|
*/ |
164
|
5 |
|
public function execute(Lexer $lexer, CharStream $input, int $startIndex) : void |
165
|
|
|
{ |
166
|
5 |
|
$requiresSeek = false; |
167
|
5 |
|
$stopIndex = $input->getIndex(); |
168
|
|
|
|
169
|
|
|
try { |
170
|
5 |
|
foreach ($this->lexerActions as $lexerAction) { |
171
|
5 |
|
if ($lexerAction instanceof LexerIndexedCustomAction) { |
172
|
|
|
$offset = $lexerAction->getOffset(); |
173
|
|
|
$input->seek($startIndex + $offset); |
174
|
|
|
$lexerAction = $lexerAction->getAction(); |
175
|
|
|
$requiresSeek = $startIndex + $offset !== $stopIndex; |
176
|
5 |
|
} elseif ($lexerAction->isPositionDependent()) { |
177
|
|
|
$input->seek($stopIndex); |
178
|
|
|
$requiresSeek = false; |
179
|
|
|
} |
180
|
|
|
|
181
|
5 |
|
$lexerAction->execute($lexer); |
182
|
|
|
} |
183
|
5 |
|
} finally { |
184
|
5 |
|
if ($requiresSeek) { |
185
|
5 |
|
$input->seek($stopIndex); |
186
|
|
|
} |
187
|
|
|
} |
188
|
5 |
|
} |
189
|
|
|
|
190
|
|
|
public function hashCode() : int |
191
|
|
|
{ |
192
|
|
|
if ($this->cachedHashCode === null) { |
193
|
|
|
$this->cachedHashCode = Hasher::hash($this->lexerActions); |
194
|
|
|
} |
195
|
|
|
|
196
|
|
|
return $this->cachedHashCode; |
197
|
|
|
} |
198
|
|
|
|
199
|
|
|
public function equals(object $other) : bool |
200
|
|
|
{ |
201
|
|
|
if ($this === $other) { |
202
|
|
|
return true; |
203
|
|
|
} |
204
|
|
|
|
205
|
|
|
return $other instanceof self |
206
|
|
|
&& $this->hashCode() === $other->hashCode() |
207
|
|
|
&& Equality::equals($this->lexerActions, $other->lexerActions); |
208
|
|
|
} |
209
|
|
|
|
210
|
|
|
public function __toString() : string |
211
|
|
|
{ |
212
|
|
|
return \sprintf('LexerActionExecutor[%s]', \implode(', ', $this->lexerActions)); |
213
|
|
|
} |
214
|
|
|
} |
215
|
|
|
|