1
|
|
|
<?php |
2
|
|
|
/** |
3
|
|
|
* Tokenizes PHP code. |
4
|
|
|
* |
5
|
|
|
* @author Greg Sherwood <[email protected]> |
6
|
|
|
* @copyright 2006-2015 Squiz Pty Ltd (ABN 77 084 670 600) |
7
|
|
|
* @license https://github.com/squizlabs/Symplify\PHP7_CodeSniffer/blob/master/licence.txt BSD Licence |
8
|
|
|
*/ |
9
|
|
|
|
10
|
|
|
namespace Symplify\PHP7_CodeSniffer\Tokenizers; |
11
|
|
|
|
12
|
|
|
use Symplify\PHP7_CodeSniffer\Util; |
13
|
|
|
|
14
|
|
|
class PHP extends Tokenizer |
15
|
|
|
{ |
16
|
|
|
|
17
|
|
|
|
18
|
|
|
/** |
19
|
|
|
* A list of tokens that are allowed to open a scope. |
20
|
|
|
* |
21
|
|
|
* This array also contains information about what kind of token the scope |
22
|
|
|
* opener uses to open and close the scope, if the token strictly requires |
23
|
|
|
* an opener, if the token can share a scope closer, and who it can be shared |
24
|
|
|
* with. An example of a token that shares a scope closer is a CASE scope. |
25
|
|
|
* |
26
|
|
|
* @var array |
27
|
|
|
*/ |
28
|
|
|
public $scopeOpeners = array( |
29
|
|
|
T_IF => array( |
30
|
|
|
'start' => array( |
31
|
|
|
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
32
|
|
|
T_COLON => T_COLON, |
33
|
|
|
), |
34
|
|
|
'end' => array( |
35
|
|
|
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
36
|
|
|
T_ENDIF => T_ENDIF, |
37
|
|
|
T_ELSE => T_ELSE, |
38
|
|
|
T_ELSEIF => T_ELSEIF, |
39
|
|
|
), |
40
|
|
|
'strict' => false, |
41
|
|
|
'shared' => false, |
42
|
|
|
'with' => array( |
43
|
|
|
T_ELSE => T_ELSE, |
44
|
|
|
T_ELSEIF => T_ELSEIF, |
45
|
|
|
), |
46
|
|
|
), |
47
|
|
|
T_TRY => array( |
48
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
49
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
50
|
|
|
'strict' => true, |
51
|
|
|
'shared' => false, |
52
|
|
|
'with' => array(), |
53
|
|
|
), |
54
|
|
|
T_CATCH => array( |
55
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
56
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
57
|
|
|
'strict' => true, |
58
|
|
|
'shared' => false, |
59
|
|
|
'with' => array(), |
60
|
|
|
), |
61
|
|
|
T_FINALLY => array( |
62
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
63
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
64
|
|
|
'strict' => true, |
65
|
|
|
'shared' => false, |
66
|
|
|
'with' => array(), |
67
|
|
|
), |
68
|
|
|
T_ELSE => array( |
69
|
|
|
'start' => array( |
70
|
|
|
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
71
|
|
|
T_COLON => T_COLON, |
72
|
|
|
), |
73
|
|
|
'end' => array( |
74
|
|
|
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
75
|
|
|
T_ENDIF => T_ENDIF, |
76
|
|
|
), |
77
|
|
|
'strict' => false, |
78
|
|
|
'shared' => false, |
79
|
|
|
'with' => array( |
80
|
|
|
T_IF => T_IF, |
81
|
|
|
T_ELSEIF => T_ELSEIF, |
82
|
|
|
), |
83
|
|
|
), |
84
|
|
|
T_ELSEIF => array( |
85
|
|
|
'start' => array( |
86
|
|
|
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
87
|
|
|
T_COLON => T_COLON, |
88
|
|
|
), |
89
|
|
|
'end' => array( |
90
|
|
|
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
91
|
|
|
T_ENDIF => T_ENDIF, |
92
|
|
|
T_ELSE => T_ELSE, |
93
|
|
|
T_ELSEIF => T_ELSEIF, |
94
|
|
|
), |
95
|
|
|
'strict' => false, |
96
|
|
|
'shared' => false, |
97
|
|
|
'with' => array( |
98
|
|
|
T_IF => T_IF, |
99
|
|
|
T_ELSE => T_ELSE, |
100
|
|
|
), |
101
|
|
|
), |
102
|
|
|
T_FOR => array( |
103
|
|
|
'start' => array( |
104
|
|
|
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
105
|
|
|
T_COLON => T_COLON, |
106
|
|
|
), |
107
|
|
|
'end' => array( |
108
|
|
|
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
109
|
|
|
T_ENDFOR => T_ENDFOR, |
110
|
|
|
), |
111
|
|
|
'strict' => false, |
112
|
|
|
'shared' => false, |
113
|
|
|
'with' => array(), |
114
|
|
|
), |
115
|
|
|
T_FOREACH => array( |
116
|
|
|
'start' => array( |
117
|
|
|
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
118
|
|
|
T_COLON => T_COLON, |
119
|
|
|
), |
120
|
|
|
'end' => array( |
121
|
|
|
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
122
|
|
|
T_ENDFOREACH => T_ENDFOREACH, |
123
|
|
|
), |
124
|
|
|
'strict' => false, |
125
|
|
|
'shared' => false, |
126
|
|
|
'with' => array(), |
127
|
|
|
), |
128
|
|
|
T_INTERFACE => array( |
129
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
130
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
131
|
|
|
'strict' => true, |
132
|
|
|
'shared' => false, |
133
|
|
|
'with' => array(), |
134
|
|
|
), |
135
|
|
|
T_FUNCTION => array( |
136
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
137
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
138
|
|
|
'strict' => true, |
139
|
|
|
'shared' => false, |
140
|
|
|
'with' => array(), |
141
|
|
|
), |
142
|
|
|
T_CLASS => array( |
143
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
144
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
145
|
|
|
'strict' => true, |
146
|
|
|
'shared' => false, |
147
|
|
|
'with' => array(), |
148
|
|
|
), |
149
|
|
|
T_TRAIT => array( |
150
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
151
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
152
|
|
|
'strict' => true, |
153
|
|
|
'shared' => false, |
154
|
|
|
'with' => array(), |
155
|
|
|
), |
156
|
|
|
T_USE => array( |
157
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
158
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
159
|
|
|
'strict' => false, |
160
|
|
|
'shared' => false, |
161
|
|
|
'with' => array(), |
162
|
|
|
), |
163
|
|
|
T_DECLARE => array( |
164
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
165
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
166
|
|
|
'strict' => false, |
167
|
|
|
'shared' => false, |
168
|
|
|
'with' => array(), |
169
|
|
|
), |
170
|
|
|
T_NAMESPACE => array( |
171
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
172
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
173
|
|
|
'strict' => false, |
174
|
|
|
'shared' => false, |
175
|
|
|
'with' => array(), |
176
|
|
|
), |
177
|
|
|
T_WHILE => array( |
178
|
|
|
'start' => array( |
179
|
|
|
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
180
|
|
|
T_COLON => T_COLON, |
181
|
|
|
), |
182
|
|
|
'end' => array( |
183
|
|
|
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
184
|
|
|
T_ENDWHILE => T_ENDWHILE, |
185
|
|
|
), |
186
|
|
|
'strict' => false, |
187
|
|
|
'shared' => false, |
188
|
|
|
'with' => array(), |
189
|
|
|
), |
190
|
|
|
T_DO => array( |
191
|
|
|
'start' => array(T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET), |
192
|
|
|
'end' => array(T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET), |
193
|
|
|
'strict' => true, |
194
|
|
|
'shared' => false, |
195
|
|
|
'with' => array(), |
196
|
|
|
), |
197
|
|
|
T_SWITCH => array( |
198
|
|
|
'start' => array( |
199
|
|
|
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
200
|
|
|
T_COLON => T_COLON, |
201
|
|
|
), |
202
|
|
|
'end' => array( |
203
|
|
|
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
204
|
|
|
T_ENDSWITCH => T_ENDSWITCH, |
205
|
|
|
), |
206
|
|
|
'strict' => true, |
207
|
|
|
'shared' => false, |
208
|
|
|
'with' => array(), |
209
|
|
|
), |
210
|
|
|
T_CASE => array( |
211
|
|
|
'start' => array( |
212
|
|
|
T_COLON => T_COLON, |
213
|
|
|
T_SEMICOLON => T_SEMICOLON, |
214
|
|
|
), |
215
|
|
|
'end' => array( |
216
|
|
|
T_BREAK => T_BREAK, |
217
|
|
|
T_RETURN => T_RETURN, |
218
|
|
|
T_CONTINUE => T_CONTINUE, |
219
|
|
|
T_THROW => T_THROW, |
220
|
|
|
T_EXIT => T_EXIT, |
221
|
|
|
), |
222
|
|
|
'strict' => true, |
223
|
|
|
'shared' => true, |
224
|
|
|
'with' => array( |
225
|
|
|
T_DEFAULT => T_DEFAULT, |
226
|
|
|
T_CASE => T_CASE, |
227
|
|
|
T_SWITCH => T_SWITCH, |
228
|
|
|
), |
229
|
|
|
), |
230
|
|
|
T_DEFAULT => array( |
231
|
|
|
'start' => array( |
232
|
|
|
T_COLON => T_COLON, |
233
|
|
|
T_SEMICOLON => T_SEMICOLON, |
234
|
|
|
), |
235
|
|
|
'end' => array( |
236
|
|
|
T_BREAK => T_BREAK, |
237
|
|
|
T_RETURN => T_RETURN, |
238
|
|
|
T_CONTINUE => T_CONTINUE, |
239
|
|
|
T_THROW => T_THROW, |
240
|
|
|
T_EXIT => T_EXIT, |
241
|
|
|
), |
242
|
|
|
'strict' => true, |
243
|
|
|
'shared' => true, |
244
|
|
|
'with' => array( |
245
|
|
|
T_CASE => T_CASE, |
246
|
|
|
T_SWITCH => T_SWITCH, |
247
|
|
|
), |
248
|
|
|
), |
249
|
|
|
T_START_HEREDOC => array( |
250
|
|
|
'start' => array(T_START_HEREDOC => T_START_HEREDOC), |
251
|
|
|
'end' => array(T_END_HEREDOC => T_END_HEREDOC), |
252
|
|
|
'strict' => true, |
253
|
|
|
'shared' => false, |
254
|
|
|
'with' => array(), |
255
|
|
|
), |
256
|
|
|
); |
257
|
|
|
|
258
|
|
|
/** |
259
|
|
|
* A list of tokens that end the scope. |
260
|
|
|
* |
261
|
|
|
* This array is just a unique collection of the end tokens |
262
|
|
|
* from the _scopeOpeners array. The data is duplicated here to |
263
|
|
|
* save time during parsing of the file. |
264
|
|
|
* |
265
|
|
|
* @var array |
266
|
|
|
*/ |
267
|
|
|
public $endScopeTokens = array( |
268
|
|
|
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
269
|
|
|
T_ENDIF => T_ENDIF, |
270
|
|
|
T_ENDFOR => T_ENDFOR, |
271
|
|
|
T_ENDFOREACH => T_ENDFOREACH, |
272
|
|
|
T_ENDWHILE => T_ENDWHILE, |
273
|
|
|
T_ENDSWITCH => T_ENDSWITCH, |
274
|
|
|
T_BREAK => T_BREAK, |
275
|
|
|
T_END_HEREDOC => T_END_HEREDOC, |
276
|
|
|
); |
277
|
|
|
|
278
|
|
|
/** |
279
|
|
|
* Known lengths of tokens. |
280
|
|
|
* |
281
|
|
|
* @var array<int, int> |
282
|
|
|
*/ |
283
|
|
|
public $knownLengths = array( |
284
|
|
|
T_ABSTRACT => 8, |
285
|
|
|
T_AND_EQUAL => 2, |
286
|
|
|
T_ARRAY => 5, |
287
|
|
|
T_AS => 2, |
288
|
|
|
T_BOOLEAN_AND => 2, |
289
|
|
|
T_BOOLEAN_OR => 2, |
290
|
|
|
T_BREAK => 5, |
291
|
|
|
T_CALLABLE => 8, |
292
|
|
|
T_CASE => 4, |
293
|
|
|
T_CATCH => 5, |
294
|
|
|
T_CLASS => 5, |
295
|
|
|
T_CLASS_C => 9, |
296
|
|
|
T_CLONE => 5, |
297
|
|
|
T_CONCAT_EQUAL => 2, |
298
|
|
|
T_CONST => 5, |
299
|
|
|
T_CONTINUE => 8, |
300
|
|
|
T_CURLY_OPEN => 2, |
301
|
|
|
T_DEC => 2, |
302
|
|
|
T_DECLARE => 7, |
303
|
|
|
T_DEFAULT => 7, |
304
|
|
|
T_DIR => 7, |
305
|
|
|
T_DIV_EQUAL => 2, |
306
|
|
|
T_DO => 2, |
307
|
|
|
T_DOLLAR_OPEN_CURLY_BRACES => 2, |
308
|
|
|
T_DOUBLE_ARROW => 2, |
309
|
|
|
T_DOUBLE_COLON => 2, |
310
|
|
|
T_ECHO => 4, |
311
|
|
|
T_ELSE => 4, |
312
|
|
|
T_ELSEIF => 6, |
313
|
|
|
T_EMPTY => 5, |
314
|
|
|
T_ENDDECLARE => 10, |
315
|
|
|
T_ENDFOR => 6, |
316
|
|
|
T_ENDFOREACH => 10, |
317
|
|
|
T_ENDIF => 5, |
318
|
|
|
T_ENDSWITCH => 9, |
319
|
|
|
T_ENDWHILE => 8, |
320
|
|
|
T_EVAL => 4, |
321
|
|
|
T_EXTENDS => 7, |
322
|
|
|
T_FILE => 8, |
323
|
|
|
T_FINAL => 5, |
324
|
|
|
T_FINALLY => 7, |
325
|
|
|
T_FOR => 3, |
326
|
|
|
T_FOREACH => 7, |
327
|
|
|
T_FUNCTION => 8, |
328
|
|
|
T_FUNC_C => 12, |
329
|
|
|
T_GLOBAL => 6, |
330
|
|
|
T_GOTO => 4, |
331
|
|
|
T_HALT_COMPILER => 15, |
332
|
|
|
T_IF => 2, |
333
|
|
|
T_IMPLEMENTS => 10, |
334
|
|
|
T_INC => 2, |
335
|
|
|
T_INCLUDE => 7, |
336
|
|
|
T_INCLUDE_ONCE => 12, |
337
|
|
|
T_INSTANCEOF => 10, |
338
|
|
|
T_INSTEADOF => 9, |
339
|
|
|
T_INTERFACE => 9, |
340
|
|
|
T_ISSET => 5, |
341
|
|
|
T_IS_EQUAL => 2, |
342
|
|
|
T_IS_GREATER_OR_EQUAL => 2, |
343
|
|
|
T_IS_IDENTICAL => 3, |
344
|
|
|
T_IS_NOT_EQUAL => 2, |
345
|
|
|
T_IS_NOT_IDENTICAL => 3, |
346
|
|
|
T_IS_SMALLER_OR_EQUAL => 2, |
347
|
|
|
T_LINE => 8, |
348
|
|
|
T_LIST => 4, |
349
|
|
|
T_LOGICAL_AND => 3, |
350
|
|
|
T_LOGICAL_OR => 2, |
351
|
|
|
T_LOGICAL_XOR => 3, |
352
|
|
|
T_METHOD_C => 10, |
353
|
|
|
T_MINUS_EQUAL => 2, |
354
|
|
|
T_POW_EQUAL => 3, |
355
|
|
|
T_MOD_EQUAL => 2, |
356
|
|
|
T_MUL_EQUAL => 2, |
357
|
|
|
T_NAMESPACE => 9, |
358
|
|
|
T_NS_C => 13, |
359
|
|
|
T_NS_SEPARATOR => 1, |
360
|
|
|
T_NEW => 3, |
361
|
|
|
T_OBJECT_OPERATOR => 2, |
362
|
|
|
T_OPEN_TAG_WITH_ECHO => 3, |
363
|
|
|
T_OR_EQUAL => 2, |
364
|
|
|
T_PLUS_EQUAL => 2, |
365
|
|
|
T_PRINT => 5, |
366
|
|
|
T_PRIVATE => 7, |
367
|
|
|
T_PUBLIC => 6, |
368
|
|
|
T_PROTECTED => 9, |
369
|
|
|
T_REQUIRE => 7, |
370
|
|
|
T_REQUIRE_ONCE => 12, |
371
|
|
|
T_RETURN => 6, |
372
|
|
|
T_STATIC => 6, |
373
|
|
|
T_SWITCH => 6, |
374
|
|
|
T_THROW => 5, |
375
|
|
|
T_TRAIT => 5, |
376
|
|
|
T_TRAIT_C => 9, |
377
|
|
|
T_TRY => 3, |
378
|
|
|
T_UNSET => 5, |
379
|
|
|
T_USE => 3, |
380
|
|
|
T_VAR => 3, |
381
|
|
|
T_WHILE => 5, |
382
|
|
|
T_XOR_EQUAL => 2, |
383
|
|
|
T_YIELD => 5, |
384
|
|
|
T_OPEN_CURLY_BRACKET => 1, |
385
|
|
|
T_CLOSE_CURLY_BRACKET => 1, |
386
|
|
|
T_OPEN_SQUARE_BRACKET => 1, |
387
|
|
|
T_CLOSE_SQUARE_BRACKET => 1, |
388
|
|
|
T_OPEN_PARENTHESIS => 1, |
389
|
|
|
T_CLOSE_PARENTHESIS => 1, |
390
|
|
|
T_COLON => 1, |
391
|
|
|
T_STRING_CONCAT => 1, |
392
|
|
|
T_INLINE_THEN => 1, |
393
|
|
|
T_INLINE_ELSE => 1, |
394
|
|
|
T_NULL => 4, |
395
|
|
|
T_FALSE => 5, |
396
|
|
|
T_TRUE => 4, |
397
|
|
|
T_SEMICOLON => 1, |
398
|
|
|
T_EQUAL => 1, |
399
|
|
|
T_MULTIPLY => 1, |
400
|
|
|
T_DIVIDE => 1, |
401
|
|
|
T_PLUS => 1, |
402
|
|
|
T_MINUS => 1, |
403
|
|
|
T_MODULUS => 1, |
404
|
|
|
T_POW => 2, |
405
|
|
|
T_SPACESHIP => 3, |
406
|
|
|
T_COALESCE => 2, |
407
|
|
|
T_BITWISE_AND => 1, |
408
|
|
|
T_BITWISE_OR => 1, |
409
|
|
|
T_BITWISE_XOR => 1, |
410
|
|
|
T_SL => 2, |
411
|
|
|
T_SR => 2, |
412
|
|
|
T_SL_EQUAL => 3, |
413
|
|
|
T_SR_EQUAL => 3, |
414
|
|
|
T_ARRAY_HINT => 5, |
415
|
|
|
T_GREATER_THAN => 1, |
416
|
|
|
T_LESS_THAN => 1, |
417
|
|
|
T_BOOLEAN_NOT => 1, |
418
|
|
|
T_SELF => 4, |
419
|
|
|
T_PARENT => 6, |
420
|
|
|
T_COMMA => 1, |
421
|
|
|
T_THIS => 4, |
422
|
|
|
T_CLOSURE => 8, |
423
|
|
|
T_BACKTICK => 1, |
424
|
|
|
T_OPEN_SHORT_ARRAY => 1, |
425
|
|
|
T_CLOSE_SHORT_ARRAY => 1, |
426
|
|
|
); |
427
|
|
|
|
428
|
|
|
|
429
|
|
|
/** |
430
|
|
|
* A cache of different token types, resolved into arrays. |
431
|
|
|
* |
432
|
|
|
* @var array |
433
|
|
|
* @see standardiseToken() |
434
|
|
|
*/ |
435
|
|
|
private static $_resolveTokenCache = array(); |
436
|
|
|
|
437
|
|
|
|
438
|
|
|
/** |
439
|
|
|
* Creates an array of tokens when given some PHP code. |
440
|
|
|
* |
441
|
|
|
* Starts by using token_get_all() but does a lot of extra processing |
442
|
|
|
* to insert information about the context of the token. |
443
|
|
|
* |
444
|
|
|
* @param string $string The string to tokenize. |
445
|
|
|
* |
446
|
|
|
* @return array |
447
|
|
|
*/ |
448
|
|
|
protected function tokenize($string) |
449
|
|
|
{ |
450
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
451
|
|
|
echo "\t*** START PHP TOKENIZING ***".PHP_EOL; |
452
|
|
|
$isWin = false; |
453
|
|
|
if (strtoupper(substr(PHP_OS, 0, 3)) === 'WIN') { |
454
|
|
|
$isWin = true; |
455
|
|
|
} |
456
|
|
|
} |
457
|
|
|
|
458
|
|
|
$tokens = @token_get_all($string); |
459
|
|
|
$finalTokens = array(); |
460
|
|
|
|
461
|
|
|
$newStackPtr = 0; |
462
|
|
|
$numTokens = count($tokens); |
463
|
|
|
$lastNotEmptyToken = 0; |
464
|
|
|
|
465
|
|
|
$insideInlineIf = array(); |
466
|
|
|
$insideUseGroup = false; |
467
|
|
|
|
468
|
|
|
$commentTokenizer = new Comment(); |
469
|
|
|
|
470
|
|
|
for ($stackPtr = 0; $stackPtr < $numTokens; $stackPtr++) { |
471
|
|
|
$token = (array) $tokens[$stackPtr]; |
472
|
|
|
$tokenIsArray = isset($token[1]); |
473
|
|
|
|
474
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
475
|
|
|
if ($tokenIsArray === true) { |
476
|
|
|
$type = token_name($token[0]); |
477
|
|
|
$content = Util\Common::prepareForOutput($token[1]); |
478
|
|
|
} else { |
479
|
|
|
$newToken = self::resolveSimpleToken($token[0]); |
480
|
|
|
$type = $newToken['type']; |
481
|
|
|
$content = Util\Common::prepareForOutput($token[0]); |
482
|
|
|
} |
483
|
|
|
|
484
|
|
|
echo "\tProcess token "; |
485
|
|
|
if ($tokenIsArray === true) { |
486
|
|
|
echo "[$stackPtr]"; |
487
|
|
|
} else { |
488
|
|
|
echo " $stackPtr "; |
489
|
|
|
} |
490
|
|
|
|
491
|
|
|
echo ": $type => $content"; |
492
|
|
|
}//end if |
493
|
|
|
|
494
|
|
|
if ($newStackPtr > 0 && $finalTokens[($newStackPtr - 1)]['code'] !== T_WHITESPACE) { |
495
|
|
|
$lastNotEmptyToken = ($newStackPtr - 1); |
496
|
|
|
} |
497
|
|
|
|
498
|
|
|
/* |
499
|
|
|
If we are using \r\n newline characters, the \r and \n are sometimes |
500
|
|
|
split over two tokens. This normally occurs after comments. We need |
501
|
|
|
to merge these two characters together so that our line endings are |
502
|
|
|
consistent for all lines. |
503
|
|
|
*/ |
504
|
|
|
|
505
|
|
|
if ($tokenIsArray === true && substr($token[1], -1) === "\r") { |
506
|
|
|
if (isset($tokens[($stackPtr + 1)]) === true |
507
|
|
|
&& is_array($tokens[($stackPtr + 1)]) === true |
508
|
|
|
&& $tokens[($stackPtr + 1)][1][0] === "\n" |
509
|
|
|
) { |
510
|
|
|
$token[1] .= "\n"; |
511
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
512
|
|
|
if ($isWin === true) { |
|
|
|
|
513
|
|
|
echo '\n'; |
514
|
|
|
} else { |
515
|
|
|
echo "\033[30;1m\\n\033[0m"; |
516
|
|
|
} |
517
|
|
|
} |
518
|
|
|
|
519
|
|
|
if ($tokens[($stackPtr + 1)][1] === "\n") { |
520
|
|
|
// This token's content has been merged into the previous, |
521
|
|
|
// so we can skip it. |
522
|
|
|
$tokens[($stackPtr + 1)] = ''; |
523
|
|
|
} else { |
524
|
|
|
$tokens[($stackPtr + 1)][1] = substr($tokens[($stackPtr + 1)][1], 1); |
525
|
|
|
} |
526
|
|
|
} |
527
|
|
|
}//end if |
528
|
|
|
|
529
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
530
|
|
|
echo PHP_EOL; |
531
|
|
|
} |
532
|
|
|
|
533
|
|
|
/* |
534
|
|
|
Parse doc blocks into something that can be easily iterated over. |
535
|
|
|
*/ |
536
|
|
|
|
537
|
|
|
if ($tokenIsArray === true && $token[0] === T_DOC_COMMENT) { |
538
|
|
|
$commentTokens = $commentTokenizer->tokenizeString($token[1], $this->eolChar, $newStackPtr); |
539
|
|
|
foreach ($commentTokens as $commentToken) { |
540
|
|
|
$finalTokens[$newStackPtr] = $commentToken; |
541
|
|
|
$newStackPtr++; |
542
|
|
|
} |
543
|
|
|
|
544
|
|
|
continue; |
545
|
|
|
} |
546
|
|
|
|
547
|
|
|
/* |
548
|
|
|
If this is a double quoted string, PHP will tokenize the whole |
549
|
|
|
thing which causes problems with the scope map when braces are |
550
|
|
|
within the string. So we need to merge the tokens together to |
551
|
|
|
provide a single string. |
552
|
|
|
*/ |
553
|
|
|
|
554
|
|
|
if ($tokenIsArray === false && ($token[0] === '"' || $token[0] === 'b"')) { |
555
|
|
|
// Binary casts need a special token. |
556
|
|
View Code Duplication |
if ($token[0] === 'b"') { |
|
|
|
|
557
|
|
|
$finalTokens[$newStackPtr] = array( |
558
|
|
|
'code' => T_BINARY_CAST, |
559
|
|
|
'type' => 'T_BINARY_CAST', |
560
|
|
|
'content' => 'b', |
561
|
|
|
); |
562
|
|
|
$newStackPtr++; |
563
|
|
|
} |
564
|
|
|
|
565
|
|
|
$tokenContent = '"'; |
566
|
|
|
$nestedVars = array(); |
567
|
|
|
for ($i = ($stackPtr + 1); $i < $numTokens; $i++) { |
568
|
|
|
$subToken = (array) $tokens[$i]; |
569
|
|
|
$subTokenIsArray = isset($subToken[1]); |
570
|
|
|
|
571
|
|
|
if ($subTokenIsArray === true) { |
572
|
|
|
$tokenContent .= $subToken[1]; |
573
|
|
|
if ($subToken[1] === '{' |
574
|
|
|
&& $subToken[0] !== T_ENCAPSED_AND_WHITESPACE |
575
|
|
|
) { |
576
|
|
|
$nestedVars[] = $i; |
577
|
|
|
} |
578
|
|
|
} else { |
579
|
|
|
$tokenContent .= $subToken[0]; |
580
|
|
|
if ($subToken[0] === '}') { |
581
|
|
|
array_pop($nestedVars); |
582
|
|
|
} |
583
|
|
|
} |
584
|
|
|
|
585
|
|
|
if ($subTokenIsArray === false |
586
|
|
|
&& $subToken[0] === '"' |
587
|
|
|
&& empty($nestedVars) === true |
588
|
|
|
) { |
589
|
|
|
// We found the other end of the double quoted string. |
590
|
|
|
break; |
591
|
|
|
} |
592
|
|
|
}//end for |
593
|
|
|
|
594
|
|
|
$stackPtr = $i; |
595
|
|
|
|
596
|
|
|
// Convert each line within the double quoted string to a |
597
|
|
|
// new token, so it conforms with other multiple line tokens. |
598
|
|
|
$tokenLines = explode($this->eolChar, $tokenContent); |
599
|
|
|
$numLines = count($tokenLines); |
600
|
|
|
$newToken = array(); |
601
|
|
|
|
602
|
|
|
for ($j = 0; $j < $numLines; $j++) { |
603
|
|
|
$newToken['content'] = $tokenLines[$j]; |
604
|
|
View Code Duplication |
if ($j === ($numLines - 1)) { |
|
|
|
|
605
|
|
|
if ($tokenLines[$j] === '') { |
606
|
|
|
break; |
607
|
|
|
} |
608
|
|
|
} else { |
609
|
|
|
$newToken['content'] .= $this->eolChar; |
610
|
|
|
} |
611
|
|
|
|
612
|
|
|
$newToken['code'] = T_DOUBLE_QUOTED_STRING; |
613
|
|
|
$newToken['type'] = 'T_DOUBLE_QUOTED_STRING'; |
614
|
|
|
$finalTokens[$newStackPtr] = $newToken; |
615
|
|
|
$newStackPtr++; |
616
|
|
|
} |
617
|
|
|
|
618
|
|
|
// Continue, as we're done with this token. |
619
|
|
|
continue; |
620
|
|
|
}//end if |
621
|
|
|
|
622
|
|
|
/* |
623
|
|
|
If this is a heredoc, PHP will tokenize the whole |
624
|
|
|
thing which causes problems when heredocs don't |
625
|
|
|
contain real PHP code, which is almost never. |
626
|
|
|
We want to leave the start and end heredoc tokens |
627
|
|
|
alone though. |
628
|
|
|
*/ |
629
|
|
|
|
630
|
|
|
if ($tokenIsArray === true && $token[0] === T_START_HEREDOC) { |
631
|
|
|
// Add the start heredoc token to the final array. |
632
|
|
|
$finalTokens[$newStackPtr] = self::standardiseToken($token); |
633
|
|
|
|
634
|
|
|
// Check if this is actually a nowdoc and use a different token |
635
|
|
|
// to help the sniffs. |
636
|
|
|
$nowdoc = false; |
637
|
|
|
if ($token[1][3] === "'") { |
638
|
|
|
$finalTokens[$newStackPtr]['code'] = T_START_NOWDOC; |
639
|
|
|
$finalTokens[$newStackPtr]['type'] = 'T_START_NOWDOC'; |
640
|
|
|
$nowdoc = true; |
641
|
|
|
} |
642
|
|
|
|
643
|
|
|
$tokenContent = ''; |
644
|
|
|
for ($i = ($stackPtr + 1); $i < $numTokens; $i++) { |
645
|
|
|
$subTokenIsArray = is_array($tokens[$i]); |
646
|
|
|
if ($subTokenIsArray === true |
647
|
|
|
&& $tokens[$i][0] === T_END_HEREDOC |
648
|
|
|
) { |
649
|
|
|
// We found the other end of the heredoc. |
650
|
|
|
break; |
651
|
|
|
} |
652
|
|
|
|
653
|
|
|
if ($subTokenIsArray === true) { |
654
|
|
|
$tokenContent .= $tokens[$i][1]; |
655
|
|
|
} else { |
656
|
|
|
$tokenContent .= $tokens[$i]; |
657
|
|
|
} |
658
|
|
|
} |
659
|
|
|
|
660
|
|
|
if ($i === $numTokens) { |
661
|
|
|
// We got to the end of the file and never |
662
|
|
|
// found the closing token, so this probably wasn't |
663
|
|
|
// a heredoc. |
664
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
665
|
|
|
$type = $finalTokens[$newStackPtr]['type']; |
666
|
|
|
echo "\t\t* failed to find the end of the here/nowdoc".PHP_EOL; |
667
|
|
|
echo "\t\t* token $stackPtr changed from $type to T_STRING".PHP_EOL; |
668
|
|
|
} |
669
|
|
|
|
670
|
|
|
$finalTokens[$newStackPtr]['code'] = T_STRING; |
671
|
|
|
$finalTokens[$newStackPtr]['type'] = 'T_STRING'; |
672
|
|
|
$newStackPtr++; |
673
|
|
|
continue; |
674
|
|
|
} |
675
|
|
|
|
676
|
|
|
$stackPtr = $i; |
677
|
|
|
$newStackPtr++; |
678
|
|
|
|
679
|
|
|
// Convert each line within the heredoc to a |
680
|
|
|
// new token, so it conforms with other multiple line tokens. |
681
|
|
|
$tokenLines = explode($this->eolChar, $tokenContent); |
682
|
|
|
$numLines = count($tokenLines); |
683
|
|
|
$newToken = array(); |
684
|
|
|
|
685
|
|
|
for ($j = 0; $j < $numLines; $j++) { |
686
|
|
|
$newToken['content'] = $tokenLines[$j]; |
687
|
|
View Code Duplication |
if ($j === ($numLines - 1)) { |
|
|
|
|
688
|
|
|
if ($tokenLines[$j] === '') { |
689
|
|
|
break; |
690
|
|
|
} |
691
|
|
|
} else { |
692
|
|
|
$newToken['content'] .= $this->eolChar; |
693
|
|
|
} |
694
|
|
|
|
695
|
|
|
if ($nowdoc === true) { |
696
|
|
|
$newToken['code'] = T_NOWDOC; |
697
|
|
|
$newToken['type'] = 'T_NOWDOC'; |
698
|
|
|
} else { |
699
|
|
|
$newToken['code'] = T_HEREDOC; |
700
|
|
|
$newToken['type'] = 'T_HEREDOC'; |
701
|
|
|
} |
702
|
|
|
|
703
|
|
|
$finalTokens[$newStackPtr] = $newToken; |
704
|
|
|
$newStackPtr++; |
705
|
|
|
}//end for |
706
|
|
|
|
707
|
|
|
// Add the end heredoc token to the final array. |
708
|
|
|
$finalTokens[$newStackPtr] = self::standardiseToken($tokens[$stackPtr]); |
709
|
|
|
|
710
|
|
|
if ($nowdoc === true) { |
711
|
|
|
$finalTokens[$newStackPtr]['code'] = T_END_NOWDOC; |
712
|
|
|
$finalTokens[$newStackPtr]['type'] = 'T_END_NOWDOC'; |
713
|
|
|
$nowdoc = true; |
|
|
|
|
714
|
|
|
} |
715
|
|
|
|
716
|
|
|
$newStackPtr++; |
717
|
|
|
|
718
|
|
|
// Continue, as we're done with this token. |
719
|
|
|
continue; |
720
|
|
|
}//end if |
721
|
|
|
|
722
|
|
|
/* |
723
|
|
|
Before PHP 5.6, the ... operator was tokenized as three |
724
|
|
|
T_STRING_CONCAT tokens in a row. So look for and combine |
725
|
|
|
these tokens in earlier versions. |
726
|
|
|
*/ |
727
|
|
|
|
728
|
|
|
if ($tokenIsArray === false |
729
|
|
|
&& $token[0] === '.' |
730
|
|
|
&& isset($tokens[($stackPtr + 1)]) === true |
731
|
|
|
&& isset($tokens[($stackPtr + 2)]) === true |
732
|
|
|
&& $tokens[($stackPtr + 1)] === '.' |
733
|
|
|
&& $tokens[($stackPtr + 2)] === '.' |
734
|
|
|
) { |
735
|
|
|
$newToken = array(); |
736
|
|
|
$newToken['code'] = T_ELLIPSIS; |
737
|
|
|
$newToken['type'] = 'T_ELLIPSIS'; |
738
|
|
|
$newToken['content'] = '...'; |
739
|
|
|
$finalTokens[$newStackPtr] = $newToken; |
740
|
|
|
|
741
|
|
|
$newStackPtr++; |
742
|
|
|
$stackPtr += 2; |
743
|
|
|
continue; |
744
|
|
|
} |
745
|
|
|
|
746
|
|
|
/* |
747
|
|
|
Before PHP 5.6, the ** operator was tokenized as two |
748
|
|
|
T_MULTIPLY tokens in a row. So look for and combine |
749
|
|
|
these tokens in earlier versions. |
750
|
|
|
*/ |
751
|
|
|
|
752
|
|
View Code Duplication |
if ($tokenIsArray === false |
|
|
|
|
753
|
|
|
&& $token[0] === '*' |
754
|
|
|
&& isset($tokens[($stackPtr + 1)]) === true |
755
|
|
|
&& $tokens[($stackPtr + 1)] === '*' |
756
|
|
|
) { |
757
|
|
|
$newToken = array(); |
758
|
|
|
$newToken['code'] = T_POW; |
759
|
|
|
$newToken['type'] = 'T_POW'; |
760
|
|
|
$newToken['content'] = '**'; |
761
|
|
|
$finalTokens[$newStackPtr] = $newToken; |
762
|
|
|
|
763
|
|
|
$newStackPtr++; |
764
|
|
|
$stackPtr++; |
765
|
|
|
continue; |
766
|
|
|
} |
767
|
|
|
|
768
|
|
|
/* |
769
|
|
|
Before PHP 5.6, the **= operator was tokenized as |
770
|
|
|
T_MULTIPLY followed by T_MUL_EQUAL. So look for and combine |
771
|
|
|
these tokens in earlier versions. |
772
|
|
|
*/ |
773
|
|
|
|
774
|
|
|
if ($tokenIsArray === false |
775
|
|
|
&& $token[0] === '*' |
776
|
|
|
&& isset($tokens[($stackPtr + 1)]) === true |
777
|
|
|
&& is_array($tokens[($stackPtr + 1)]) === true |
778
|
|
|
&& $tokens[($stackPtr + 1)][1] === '*=' |
779
|
|
|
) { |
780
|
|
|
$newToken = array(); |
781
|
|
|
$newToken['code'] = T_POW_EQUAL; |
782
|
|
|
$newToken['type'] = 'T_POW_EQUAL'; |
783
|
|
|
$newToken['content'] = '**='; |
784
|
|
|
$finalTokens[$newStackPtr] = $newToken; |
785
|
|
|
|
786
|
|
|
$newStackPtr++; |
787
|
|
|
$stackPtr++; |
788
|
|
|
continue; |
789
|
|
|
} |
790
|
|
|
|
791
|
|
|
/* |
792
|
|
|
Before PHP 7, the ?? operator was tokenized as |
793
|
|
|
T_INLINE_THEN followed by T_INLINE_THEN. |
794
|
|
|
So look for and combine these tokens in earlier versions. |
795
|
|
|
*/ |
796
|
|
|
|
797
|
|
View Code Duplication |
if ($tokenIsArray === false |
|
|
|
|
798
|
|
|
&& $token[0] === '?' |
799
|
|
|
&& isset($tokens[($stackPtr + 1)]) === true |
800
|
|
|
&& $tokens[($stackPtr + 1)][0] === '?' |
801
|
|
|
) { |
802
|
|
|
$newToken = array(); |
803
|
|
|
$newToken['code'] = T_COALESCE; |
804
|
|
|
$newToken['type'] = 'T_COALESCE'; |
805
|
|
|
$newToken['content'] = '??'; |
806
|
|
|
$finalTokens[$newStackPtr] = $newToken; |
807
|
|
|
|
808
|
|
|
$newStackPtr++; |
809
|
|
|
$stackPtr++; |
810
|
|
|
continue; |
811
|
|
|
} |
812
|
|
|
|
813
|
|
|
/* |
814
|
|
|
Before PHP 7, the <=> operator was tokenized as |
815
|
|
|
T_IS_SMALLER_OR_EQUAL followed by T_GREATER_THAN. |
816
|
|
|
So look for and combine these tokens in earlier versions. |
817
|
|
|
*/ |
818
|
|
|
|
819
|
|
View Code Duplication |
if ($tokenIsArray === true |
|
|
|
|
820
|
|
|
&& $token[0] === T_IS_SMALLER_OR_EQUAL |
821
|
|
|
&& isset($tokens[($stackPtr + 1)]) === true |
822
|
|
|
&& $tokens[($stackPtr + 1)][0] === '>' |
823
|
|
|
) { |
824
|
|
|
$newToken = array(); |
825
|
|
|
$newToken['code'] = T_SPACESHIP; |
826
|
|
|
$newToken['type'] = 'T_SPACESHIP'; |
827
|
|
|
$newToken['content'] = '<=>'; |
828
|
|
|
$finalTokens[$newStackPtr] = $newToken; |
829
|
|
|
|
830
|
|
|
$newStackPtr++; |
831
|
|
|
$stackPtr++; |
832
|
|
|
continue; |
833
|
|
|
} |
834
|
|
|
|
835
|
|
|
/* |
836
|
|
|
Emulate traits in PHP versions less than 5.4. |
837
|
|
|
*/ |
838
|
|
|
|
839
|
|
|
if ($tokenIsArray === true |
840
|
|
|
&& $token[0] === T_STRING |
841
|
|
|
&& strtolower($token[1]) === 'trait' |
842
|
|
|
&& $tokens[($stackPtr - 1)][0] !== T_OBJECT_OPERATOR |
843
|
|
|
) { |
844
|
|
|
$finalTokens[$newStackPtr] = array( |
845
|
|
|
'content' => $token[1], |
846
|
|
|
'code' => T_TRAIT, |
847
|
|
|
'type' => 'T_TRAIT', |
848
|
|
|
); |
849
|
|
|
|
850
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
851
|
|
|
echo "\t\t* token $stackPtr changed from T_STRING to T_TRAIT".PHP_EOL; |
852
|
|
|
} |
853
|
|
|
|
854
|
|
|
$newStackPtr++; |
855
|
|
|
continue; |
856
|
|
|
} |
857
|
|
|
|
858
|
|
|
/* |
859
|
|
|
PHP doesn't assign a token to goto labels, so we have to. |
860
|
|
|
These are just string tokens with a single colon after them. Double |
861
|
|
|
colons are already tokenized and so don't interfere with this check. |
862
|
|
|
But we do have to account for CASE statements, that look just like |
863
|
|
|
goto labels. |
864
|
|
|
*/ |
865
|
|
|
|
866
|
|
|
if ($tokenIsArray === true |
867
|
|
|
&& $token[0] === T_STRING |
868
|
|
|
&& isset($tokens[($stackPtr + 1)]) === true |
869
|
|
|
&& $tokens[($stackPtr + 1)] === ':' |
870
|
|
|
&& $tokens[($stackPtr - 1)][0] !== T_PAAMAYIM_NEKUDOTAYIM |
871
|
|
|
) { |
872
|
|
|
$stopTokens = array( |
873
|
|
|
T_CASE => true, |
874
|
|
|
T_SEMICOLON => true, |
875
|
|
|
T_OPEN_CURLY_BRACKET => true, |
876
|
|
|
T_INLINE_THEN => true, |
877
|
|
|
); |
878
|
|
|
|
879
|
|
|
for ($x = ($newStackPtr - 1); $x > 0; $x--) { |
880
|
|
|
if (isset($stopTokens[$finalTokens[$x]['code']]) === true) { |
881
|
|
|
break; |
882
|
|
|
} |
883
|
|
|
} |
884
|
|
|
|
885
|
|
|
if ($finalTokens[$x]['code'] !== T_CASE |
886
|
|
|
&& $finalTokens[$x]['code'] !== T_INLINE_THEN |
887
|
|
|
) { |
888
|
|
|
$finalTokens[$newStackPtr] = array( |
889
|
|
|
'content' => $token[1].':', |
890
|
|
|
'code' => T_GOTO_LABEL, |
891
|
|
|
'type' => 'T_GOTO_LABEL', |
892
|
|
|
); |
893
|
|
|
|
894
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
895
|
|
|
echo "\t\t* token $stackPtr changed from T_STRING to T_GOTO_LABEL".PHP_EOL; |
896
|
|
|
echo "\t\t* skipping T_COLON token ".($stackPtr + 1).PHP_EOL; |
897
|
|
|
} |
898
|
|
|
|
899
|
|
|
$newStackPtr++; |
900
|
|
|
$stackPtr++; |
901
|
|
|
continue; |
902
|
|
|
} |
903
|
|
|
}//end if |
904
|
|
|
|
905
|
|
|
/* |
906
|
|
|
HHVM 3.5 tokenizes "else[\s]+if" as a T_ELSEIF token while PHP |
907
|
|
|
proper only tokenizes "elseif" as a T_ELSEIF token. So split |
908
|
|
|
up the HHVM token to make it looks like proper PHP. |
909
|
|
|
*/ |
910
|
|
|
|
911
|
|
|
if ($tokenIsArray === true |
912
|
|
|
&& $token[0] === T_ELSEIF |
913
|
|
|
&& strtolower($token[1]) !== 'elseif' |
914
|
|
|
) { |
915
|
|
|
$finalTokens[$newStackPtr] = array( |
916
|
|
|
'content' => substr($token[1], 0, 4), |
917
|
|
|
'code' => T_ELSE, |
918
|
|
|
'type' => 'T_ELSE', |
919
|
|
|
); |
920
|
|
|
|
921
|
|
|
$newStackPtr++; |
922
|
|
|
$finalTokens[$newStackPtr] = array( |
923
|
|
|
'content' => substr($token[1], 4, -2), |
924
|
|
|
'code' => T_WHITESPACE, |
925
|
|
|
'type' => 'T_WHITESPACE', |
926
|
|
|
); |
927
|
|
|
|
928
|
|
|
$newStackPtr++; |
929
|
|
|
$finalTokens[$newStackPtr] = array( |
930
|
|
|
'content' => substr($token[1], -2), |
931
|
|
|
'code' => T_IF, |
932
|
|
|
'type' => 'T_IF', |
933
|
|
|
); |
934
|
|
|
|
935
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
936
|
|
|
echo "\t\t* token $stackPtr changed from T_ELSEIF to T_ELSE/T_WHITESPACE/T_IF".PHP_EOL; |
937
|
|
|
} |
938
|
|
|
|
939
|
|
|
$newStackPtr++; |
940
|
|
|
continue; |
941
|
|
|
}//end if |
942
|
|
|
|
943
|
|
|
/* |
944
|
|
|
HHVM 3.5 and 3.6 tokenizes a hashbang line such as #!/usr/bin/php |
945
|
|
|
as T_HASHANG while PHP proper uses T_INLINE_HTML. |
946
|
|
|
*/ |
947
|
|
|
|
948
|
|
|
if ($tokenIsArray === true && token_name($token[0]) === 'T_HASHBANG') { |
949
|
|
|
$finalTokens[$newStackPtr] = array( |
950
|
|
|
'content' => $token[1], |
951
|
|
|
'code' => T_INLINE_HTML, |
952
|
|
|
'type' => 'T_INLINE_HTML', |
953
|
|
|
); |
954
|
|
|
|
955
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
956
|
|
|
echo "\t\t* token $stackPtr changed from T_HASHBANG to T_INLINE_HTML".PHP_EOL; |
957
|
|
|
} |
958
|
|
|
|
959
|
|
|
$newStackPtr++; |
960
|
|
|
continue; |
961
|
|
|
}//end if |
962
|
|
|
|
963
|
|
|
/* |
964
|
|
|
If this token has newlines in its content, split each line up |
965
|
|
|
and create a new token for each line. We do this so it's easier |
966
|
|
|
to ascertain where errors occur on a line. |
967
|
|
|
Note that $token[1] is the token's content. |
968
|
|
|
*/ |
969
|
|
|
|
970
|
|
|
if ($tokenIsArray === true && strpos($token[1], $this->eolChar) !== false) { |
971
|
|
|
$tokenLines = explode($this->eolChar, $token[1]); |
972
|
|
|
$numLines = count($tokenLines); |
973
|
|
|
$newToken = array( |
974
|
|
|
'type' => token_name($token[0]), |
975
|
|
|
'code' => $token[0], |
976
|
|
|
'content' => '', |
977
|
|
|
); |
978
|
|
|
|
979
|
|
|
for ($i = 0; $i < $numLines; $i++) { |
980
|
|
|
$newToken['content'] = $tokenLines[$i]; |
981
|
|
View Code Duplication |
if ($i === ($numLines - 1)) { |
|
|
|
|
982
|
|
|
if ($tokenLines[$i] === '') { |
983
|
|
|
break; |
984
|
|
|
} |
985
|
|
|
} else { |
986
|
|
|
$newToken['content'] .= $this->eolChar; |
987
|
|
|
} |
988
|
|
|
|
989
|
|
|
$finalTokens[$newStackPtr] = $newToken; |
990
|
|
|
$newStackPtr++; |
991
|
|
|
} |
992
|
|
|
} else { |
993
|
|
|
if ($tokenIsArray === true && $token[0] === T_STRING) { |
994
|
|
|
// Some T_STRING tokens should remain that way |
995
|
|
|
// due to their context. |
996
|
|
|
$context = array( |
997
|
|
|
T_OBJECT_OPERATOR => true, |
998
|
|
|
T_FUNCTION => true, |
999
|
|
|
T_CLASS => true, |
1000
|
|
|
T_EXTENDS => true, |
1001
|
|
|
T_IMPLEMENTS => true, |
1002
|
|
|
T_NEW => true, |
1003
|
|
|
T_CONST => true, |
1004
|
|
|
T_NS_SEPARATOR => true, |
1005
|
|
|
T_USE => true, |
1006
|
|
|
T_NAMESPACE => true, |
1007
|
|
|
T_PAAMAYIM_NEKUDOTAYIM => true, |
1008
|
|
|
); |
1009
|
|
View Code Duplication |
if (isset($context[$finalTokens[$lastNotEmptyToken]['code']]) === true) { |
|
|
|
|
1010
|
|
|
$finalTokens[$newStackPtr] = array( |
1011
|
|
|
'content' => $token[1], |
1012
|
|
|
'code' => T_STRING, |
1013
|
|
|
'type' => 'T_STRING', |
1014
|
|
|
); |
1015
|
|
|
$newStackPtr++; |
1016
|
|
|
continue; |
1017
|
|
|
} |
1018
|
|
|
}//end if |
1019
|
|
|
|
1020
|
|
|
$newToken = null; |
1021
|
|
View Code Duplication |
if ($tokenIsArray === false) { |
|
|
|
|
1022
|
|
|
if (isset(self::$_resolveTokenCache[$token[0]]) === true) { |
1023
|
|
|
$newToken = self::$_resolveTokenCache[$token[0]]; |
1024
|
|
|
} |
1025
|
|
|
} else { |
1026
|
|
|
$cacheKey = null; |
1027
|
|
|
if ($token[0] === T_STRING) { |
1028
|
|
|
$cacheKey = strtolower($token[1]); |
1029
|
|
|
} else if ($token[0] !== T_CURLY_OPEN) { |
1030
|
|
|
$cacheKey = $token[0]; |
1031
|
|
|
} |
1032
|
|
|
|
1033
|
|
|
if ($cacheKey !== null && isset(self::$_resolveTokenCache[$cacheKey]) === true) { |
1034
|
|
|
$newToken = self::$_resolveTokenCache[$cacheKey]; |
1035
|
|
|
$newToken['content'] = $token[1]; |
1036
|
|
|
} |
1037
|
|
|
} |
1038
|
|
|
|
1039
|
|
|
if ($newToken === null) { |
1040
|
|
|
$newToken = self::standardiseToken($token); |
1041
|
|
|
} |
1042
|
|
|
|
1043
|
|
|
// Convert colons that are actually the ELSE component of an |
1044
|
|
|
// inline IF statement. |
1045
|
|
|
if ($newToken['code'] === T_INLINE_THEN) { |
1046
|
|
|
$insideInlineIf[] = $stackPtr; |
1047
|
|
|
} else if (empty($insideInlineIf) === false && $newToken['code'] === T_COLON) { |
1048
|
|
|
array_pop($insideInlineIf); |
1049
|
|
|
$newToken['code'] = T_INLINE_ELSE; |
1050
|
|
|
$newToken['type'] = 'T_INLINE_ELSE'; |
1051
|
|
|
} |
1052
|
|
|
|
1053
|
|
|
// This is a special condition for T_ARRAY tokens used for |
1054
|
|
|
// type hinting function arguments as being arrays. We want to keep |
1055
|
|
|
// the parenthesis map clean, so let's tag these tokens as |
1056
|
|
|
// T_ARRAY_HINT. |
1057
|
|
|
if ($newToken['code'] === T_ARRAY) { |
1058
|
|
|
// Recalculate number of tokens. |
1059
|
|
|
for ($i = $stackPtr; $i < $numTokens; $i++) { |
1060
|
|
|
if ($tokens[$i] === '(') { |
1061
|
|
|
break; |
1062
|
|
|
} else if ($tokens[$i][0] === T_VARIABLE) { |
1063
|
|
|
$newToken['code'] = T_ARRAY_HINT; |
1064
|
|
|
$newToken['type'] = 'T_ARRAY_HINT'; |
1065
|
|
|
break; |
1066
|
|
|
} |
1067
|
|
|
} |
1068
|
|
|
} |
1069
|
|
|
|
1070
|
|
|
// This is a special case when checking PHP 5.5+ code in PHP < 5.5 |
1071
|
|
|
// where "finally" should be T_FINALLY instead of T_STRING. |
1072
|
|
|
if ($newToken['code'] === T_STRING |
1073
|
|
|
&& strtolower($newToken['content']) === 'finally' |
1074
|
|
|
) { |
1075
|
|
|
$newToken['code'] = T_FINALLY; |
1076
|
|
|
$newToken['type'] = 'T_FINALLY'; |
1077
|
|
|
} |
1078
|
|
|
|
1079
|
|
|
// This is a special case for the PHP 5.5 classname::class syntax |
1080
|
|
|
// where "class" should be T_STRING instead of T_CLASS. |
1081
|
|
|
if ($newToken['code'] === T_CLASS |
1082
|
|
|
&& $finalTokens[($newStackPtr - 1)]['code'] === T_DOUBLE_COLON |
1083
|
|
|
) { |
1084
|
|
|
$newToken['code'] = T_STRING; |
1085
|
|
|
$newToken['type'] = 'T_STRING'; |
1086
|
|
|
} |
1087
|
|
|
|
1088
|
|
|
// This is a special case for PHP 5.6 use function and use const |
1089
|
|
|
// where "function" and "const" should be T_STRING instead of T_FUNCTION |
1090
|
|
|
// and T_CONST. |
1091
|
|
View Code Duplication |
if (($newToken['code'] === T_FUNCTION |
|
|
|
|
1092
|
|
|
|| $newToken['code'] === T_CONST) |
1093
|
|
|
&& $finalTokens[$lastNotEmptyToken]['code'] === T_USE |
1094
|
|
|
) { |
1095
|
|
|
$newToken['code'] = T_STRING; |
1096
|
|
|
$newToken['type'] = 'T_STRING'; |
1097
|
|
|
} |
1098
|
|
|
|
1099
|
|
|
// This is a special case for use groups in PHP 7+ where leaving |
1100
|
|
|
// the curly braces as their normal tokens would confuse |
1101
|
|
|
// the scope map and sniffs. |
1102
|
|
View Code Duplication |
if ($newToken['code'] === T_OPEN_CURLY_BRACKET |
|
|
|
|
1103
|
|
|
&& $finalTokens[$lastNotEmptyToken]['code'] === T_NS_SEPARATOR |
1104
|
|
|
) { |
1105
|
|
|
$newToken['code'] = T_OPEN_USE_GROUP; |
1106
|
|
|
$newToken['type'] = 'T_OPEN_USE_GROUP'; |
1107
|
|
|
$insideUseGroup = true; |
1108
|
|
|
} |
1109
|
|
|
|
1110
|
|
|
if ($insideUseGroup === true && $newToken['code'] === T_CLOSE_CURLY_BRACKET) { |
1111
|
|
|
$newToken['code'] = T_CLOSE_USE_GROUP; |
1112
|
|
|
$newToken['type'] = 'T_CLOSE_USE_GROUP'; |
1113
|
|
|
$insideUseGroup = false; |
1114
|
|
|
} |
1115
|
|
|
|
1116
|
|
|
$finalTokens[$newStackPtr] = $newToken; |
1117
|
|
|
$newStackPtr++; |
1118
|
|
|
}//end if |
1119
|
|
|
}//end for |
1120
|
|
|
|
1121
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1122
|
|
|
echo "\t*** END PHP TOKENIZING ***".PHP_EOL; |
1123
|
|
|
} |
1124
|
|
|
|
1125
|
|
|
return $finalTokens; |
1126
|
|
|
|
1127
|
|
|
}//end tokenize() |
1128
|
|
|
|
1129
|
|
|
|
1130
|
|
|
/** |
1131
|
|
|
* Performs additional processing after main tokenizing. |
1132
|
|
|
* |
1133
|
|
|
* This additional processing checks for CASE statements that are using curly |
1134
|
|
|
* braces for scope openers and closers. It also turns some T_FUNCTION tokens |
1135
|
|
|
* into T_CLOSURE when they are not standard function definitions. It also |
1136
|
|
|
* detects short array syntax and converts those square brackets into new tokens. |
1137
|
|
|
* It also corrects some usage of the static and class keywords. It also |
1138
|
|
|
* assigns tokens to function return types. |
1139
|
|
|
* |
1140
|
|
|
* @return void |
1141
|
|
|
*/ |
1142
|
|
|
protected function processAdditional() |
1143
|
|
|
{ |
1144
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1145
|
|
|
echo "\t*** START ADDITIONAL PHP PROCESSING ***".PHP_EOL; |
1146
|
|
|
} |
1147
|
|
|
|
1148
|
|
|
$numTokens = count($this->tokens); |
1149
|
|
|
for ($i = ($numTokens - 1); $i >= 0; $i--) { |
1150
|
|
|
// Check for any unset scope conditions due to alternate IF/ENDIF syntax. |
1151
|
|
View Code Duplication |
if (isset($this->tokens[$i]['scope_opener']) === true |
|
|
|
|
1152
|
|
|
&& isset($this->tokens[$i]['scope_condition']) === false |
1153
|
|
|
) { |
1154
|
|
|
$this->tokens[$i]['scope_condition'] = $this->tokens[$this->tokens[$i]['scope_opener']]['scope_condition']; |
1155
|
|
|
} |
1156
|
|
|
|
1157
|
|
|
if ($this->tokens[$i]['code'] === T_FUNCTION) { |
1158
|
|
|
// Context sensitive keywords support. |
1159
|
|
View Code Duplication |
for ($x = ($i + 1); $x < $numTokens; $x++) { |
|
|
|
|
1160
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
1161
|
|
|
// Non-whitespace content. |
1162
|
|
|
break; |
1163
|
|
|
} |
1164
|
|
|
} |
1165
|
|
|
|
1166
|
|
|
if ($x === $numTokens) { |
1167
|
|
|
// We got to the end without finding any more |
1168
|
|
|
// non-whitespace content. |
1169
|
|
|
continue; |
1170
|
|
|
} |
1171
|
|
|
|
1172
|
|
View Code Duplication |
if (in_array($this->tokens[$x]['code'], array(T_STRING, T_OPEN_PARENTHESIS, T_BITWISE_AND), true) === false) { |
|
|
|
|
1173
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1174
|
|
|
$line = $this->tokens[$x]['line']; |
1175
|
|
|
$type = $this->tokens[$x]['type']; |
1176
|
|
|
echo "\t* token $x on line $line changed from $type to T_STRING".PHP_EOL; |
1177
|
|
|
} |
1178
|
|
|
|
1179
|
|
|
$this->tokens[$x]['code'] = T_STRING; |
1180
|
|
|
$this->tokens[$x]['type'] = 'T_STRING'; |
1181
|
|
|
} |
1182
|
|
|
|
1183
|
|
|
/* |
1184
|
|
|
Detect functions that are actually closures and |
1185
|
|
|
assign them a different token. |
1186
|
|
|
*/ |
1187
|
|
|
|
1188
|
|
|
if (isset($this->tokens[$i]['scope_opener']) === true) { |
1189
|
|
|
for ($x = ($i + 1); $x < $numTokens; $x++) { |
1190
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false |
1191
|
|
|
&& $this->tokens[$x]['code'] !== T_BITWISE_AND |
1192
|
|
|
) { |
1193
|
|
|
break; |
1194
|
|
|
} |
1195
|
|
|
} |
1196
|
|
|
|
1197
|
|
|
if ($this->tokens[$x]['code'] === T_OPEN_PARENTHESIS) { |
1198
|
|
|
$this->tokens[$i]['code'] = T_CLOSURE; |
1199
|
|
|
$this->tokens[$i]['type'] = 'T_CLOSURE'; |
1200
|
|
View Code Duplication |
if (PHP_CodeSniffer_VERBOSITY > 1) { |
|
|
|
|
1201
|
|
|
$line = $this->tokens[$i]['line']; |
1202
|
|
|
echo "\t* token $i on line $line changed from T_FUNCTION to T_CLOSURE".PHP_EOL; |
1203
|
|
|
} |
1204
|
|
|
|
1205
|
|
View Code Duplication |
for ($x = ($this->tokens[$i]['scope_opener'] + 1); $x < $this->tokens[$i]['scope_closer']; $x++) { |
|
|
|
|
1206
|
|
|
if (isset($this->tokens[$x]['conditions'][$i]) === false) { |
1207
|
|
|
continue; |
1208
|
|
|
} |
1209
|
|
|
|
1210
|
|
|
$this->tokens[$x]['conditions'][$i] = T_CLOSURE; |
1211
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1212
|
|
|
$type = $this->tokens[$x]['type']; |
1213
|
|
|
echo "\t\t* cleaned $x ($type) *".PHP_EOL; |
1214
|
|
|
} |
1215
|
|
|
} |
1216
|
|
|
} |
1217
|
|
|
|
1218
|
|
|
$tokenAfterReturnTypeHint = $this->tokens[$i]['scope_opener']; |
1219
|
|
|
} else if (isset($this->tokens[$i]['parenthesis_closer']) === true) { |
1220
|
|
|
$tokenAfterReturnTypeHint = null; |
1221
|
|
|
for ($x = ($this->tokens[$i]['parenthesis_closer'] + 1); $x < $numTokens; $x++) { |
1222
|
|
|
if ($this->tokens[$x]['code'] === T_SEMICOLON) { |
1223
|
|
|
$tokenAfterReturnTypeHint = $x; |
1224
|
|
|
break; |
1225
|
|
|
} |
1226
|
|
|
} |
1227
|
|
|
|
1228
|
|
|
if ($tokenAfterReturnTypeHint === null) { |
1229
|
|
|
// Probably a syntax error. |
1230
|
|
|
continue; |
1231
|
|
|
} |
1232
|
|
|
} else { |
1233
|
|
|
// Probably a syntax error. |
1234
|
|
|
continue; |
1235
|
|
|
}//end if |
1236
|
|
|
|
1237
|
|
|
/* |
1238
|
|
|
Detect function return values and assign them |
1239
|
|
|
a special token, because PHP doesn't. |
1240
|
|
|
*/ |
1241
|
|
|
|
1242
|
|
|
for ($x = ($tokenAfterReturnTypeHint - 1); $x > $i; $x--) { |
1243
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
1244
|
|
|
if (in_array($this->tokens[$x]['code'], array(T_STRING, T_ARRAY, T_CALLABLE, T_SELF, T_PARENT), true) === true) { |
1245
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1246
|
|
|
$line = $this->tokens[$x]['line']; |
1247
|
|
|
$type = $this->tokens[$x]['type']; |
1248
|
|
|
echo "\t* token $x on line $line changed from $type to T_RETURN_TYPE".PHP_EOL; |
1249
|
|
|
} |
1250
|
|
|
|
1251
|
|
|
$this->tokens[$x]['code'] = T_RETURN_TYPE; |
1252
|
|
|
$this->tokens[$x]['type'] = 'T_RETURN_TYPE'; |
1253
|
|
|
} |
1254
|
|
|
|
1255
|
|
|
break; |
1256
|
|
|
} |
1257
|
|
|
} |
1258
|
|
|
|
1259
|
|
|
continue; |
1260
|
|
|
} else if ($this->tokens[$i]['code'] === T_CLASS && isset($this->tokens[$i]['scope_opener']) === true) { |
1261
|
|
|
/* |
1262
|
|
|
Detect anonymous classes and assign them a different token. |
1263
|
|
|
*/ |
1264
|
|
|
|
1265
|
|
View Code Duplication |
for ($x = ($i + 1); $x < $numTokens; $x++) { |
|
|
|
|
1266
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
1267
|
|
|
break; |
1268
|
|
|
} |
1269
|
|
|
} |
1270
|
|
|
|
1271
|
|
|
if ($this->tokens[$x]['code'] === T_OPEN_PARENTHESIS |
1272
|
|
|
|| $this->tokens[$x]['code'] === T_OPEN_CURLY_BRACKET |
1273
|
|
|
|| $this->tokens[$x]['code'] === T_EXTENDS |
1274
|
|
|
|| $this->tokens[$x]['code'] === T_IMPLEMENTS |
1275
|
|
|
) { |
1276
|
|
|
$this->tokens[$i]['code'] = T_ANON_CLASS; |
1277
|
|
|
$this->tokens[$i]['type'] = 'T_ANON_CLASS'; |
1278
|
|
View Code Duplication |
if (PHP_CodeSniffer_VERBOSITY > 1) { |
|
|
|
|
1279
|
|
|
$line = $this->tokens[$i]['line']; |
1280
|
|
|
echo "\t* token $i on line $line changed from T_CLASS to T_ANON_CLASS".PHP_EOL; |
1281
|
|
|
} |
1282
|
|
|
|
1283
|
|
View Code Duplication |
for ($x = ($this->tokens[$i]['scope_opener'] + 1); $x < $this->tokens[$i]['scope_closer']; $x++) { |
|
|
|
|
1284
|
|
|
if (isset($this->tokens[$x]['conditions'][$i]) === false) { |
1285
|
|
|
continue; |
1286
|
|
|
} |
1287
|
|
|
|
1288
|
|
|
$this->tokens[$x]['conditions'][$i] = T_ANON_CLASS; |
1289
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1290
|
|
|
$type = $this->tokens[$x]['type']; |
1291
|
|
|
echo "\t\t* cleaned $x ($type) *".PHP_EOL; |
1292
|
|
|
} |
1293
|
|
|
} |
1294
|
|
|
} |
1295
|
|
|
|
1296
|
|
|
continue; |
1297
|
|
|
} else if ($this->tokens[$i]['code'] === T_OPEN_SQUARE_BRACKET) { |
1298
|
|
|
// Unless there is a variable or a bracket before this token, |
1299
|
|
|
// it is the start of an array being defined using the short syntax. |
1300
|
|
View Code Duplication |
for ($x = ($i - 1); $x > 0; $x--) { |
|
|
|
|
1301
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
1302
|
|
|
break; |
1303
|
|
|
} |
1304
|
|
|
} |
1305
|
|
|
|
1306
|
|
|
$allowed = array( |
1307
|
|
|
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
1308
|
|
|
T_CLOSE_SQUARE_BRACKET => T_CLOSE_SQUARE_BRACKET, |
1309
|
|
|
T_CLOSE_PARENTHESIS => T_CLOSE_PARENTHESIS, |
1310
|
|
|
T_VARIABLE => T_VARIABLE, |
1311
|
|
|
T_STRING => T_STRING, |
1312
|
|
|
); |
1313
|
|
|
|
1314
|
|
|
if (isset($allowed[$this->tokens[$x]['code']]) === false |
1315
|
|
|
&& isset($this->tokens[$i]['bracket_closer']) === true |
1316
|
|
|
) { |
1317
|
|
|
$this->tokens[$i]['code'] = T_OPEN_SHORT_ARRAY; |
1318
|
|
|
$this->tokens[$i]['type'] = 'T_OPEN_SHORT_ARRAY'; |
1319
|
|
|
|
1320
|
|
|
$closer = $this->tokens[$i]['bracket_closer']; |
1321
|
|
|
$this->tokens[$closer]['code'] = T_CLOSE_SHORT_ARRAY; |
1322
|
|
|
$this->tokens[$closer]['type'] = 'T_CLOSE_SHORT_ARRAY'; |
1323
|
|
View Code Duplication |
if (PHP_CodeSniffer_VERBOSITY > 1) { |
|
|
|
|
1324
|
|
|
$line = $this->tokens[$i]['line']; |
1325
|
|
|
echo "\t* token $i on line $line changed from T_OPEN_SQUARE_BRACKET to T_OPEN_SHORT_ARRAY".PHP_EOL; |
1326
|
|
|
$line = $this->tokens[$closer]['line']; |
1327
|
|
|
echo "\t* token $closer on line $line changed from T_CLOSE_SQUARE_BRACKET to T_CLOSE_SHORT_ARRAY".PHP_EOL; |
1328
|
|
|
} |
1329
|
|
|
} |
1330
|
|
|
|
1331
|
|
|
continue; |
1332
|
|
|
} else if ($this->tokens[$i]['code'] === T_STATIC) { |
1333
|
|
View Code Duplication |
for ($x = ($i - 1); $x > 0; $x--) { |
|
|
|
|
1334
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
1335
|
|
|
break; |
1336
|
|
|
} |
1337
|
|
|
} |
1338
|
|
|
|
1339
|
|
|
if ($this->tokens[$x]['code'] === T_INSTANCEOF) { |
1340
|
|
|
$this->tokens[$i]['code'] = T_STRING; |
1341
|
|
|
$this->tokens[$i]['type'] = 'T_STRING'; |
1342
|
|
|
|
1343
|
|
View Code Duplication |
if (PHP_CodeSniffer_VERBOSITY > 1) { |
|
|
|
|
1344
|
|
|
$line = $this->tokens[$i]['line']; |
1345
|
|
|
echo "\t* token $i on line $line changed from T_STATIC to T_STRING".PHP_EOL; |
1346
|
|
|
} |
1347
|
|
|
} |
1348
|
|
|
|
1349
|
|
|
continue; |
1350
|
|
|
} else if ($this->tokens[$i]['code'] === T_ECHO && $this->tokens[$i]['content'] === '<?=') { |
1351
|
|
|
// HHVM tokenizes <?= as T_ECHO but it should be T_OPEN_TAG_WITH_ECHO. |
1352
|
|
|
$this->tokens[$i]['code'] = T_OPEN_TAG_WITH_ECHO; |
1353
|
|
|
$this->tokens[$i]['type'] = 'T_OPEN_TAG_WITH_ECHO'; |
1354
|
|
|
|
1355
|
|
View Code Duplication |
if (PHP_CodeSniffer_VERBOSITY > 1) { |
|
|
|
|
1356
|
|
|
$line = $this->tokens[$i]['line']; |
1357
|
|
|
echo "\t* token $i on line $line changed from T_ECHO to T_OPEN_TAG_WITH_ECHO".PHP_EOL; |
1358
|
|
|
} |
1359
|
|
|
} else if ($this->tokens[$i]['code'] === T_TRUE |
1360
|
|
|
|| $this->tokens[$i]['code'] === T_FALSE |
1361
|
|
|
|| $this->tokens[$i]['code'] === T_NULL |
1362
|
|
|
) { |
1363
|
|
View Code Duplication |
for ($x = ($i + 1); $i < $numTokens; $x++) { |
|
|
|
|
1364
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
1365
|
|
|
// Non-whitespace content. |
1366
|
|
|
break; |
1367
|
|
|
} |
1368
|
|
|
} |
1369
|
|
|
|
1370
|
|
|
$context = array( |
1371
|
|
|
T_OBJECT_OPERATOR => true, |
1372
|
|
|
T_NS_SEPARATOR => true, |
1373
|
|
|
T_PAAMAYIM_NEKUDOTAYIM => true, |
1374
|
|
|
); |
1375
|
|
View Code Duplication |
if (isset($context[$this->tokens[$x]['code']]) === true) { |
|
|
|
|
1376
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1377
|
|
|
$line = $this->tokens[$i]['line']; |
1378
|
|
|
$type = $this->tokens[$i]['type']; |
1379
|
|
|
echo "\t* token $i on line $line changed from $type to T_STRING".PHP_EOL; |
1380
|
|
|
} |
1381
|
|
|
|
1382
|
|
|
$this->tokens[$i]['code'] = T_STRING; |
1383
|
|
|
$this->tokens[$i]['type'] = 'T_STRING'; |
1384
|
|
|
} |
1385
|
|
|
} else if ($this->tokens[$i]['code'] === T_CONST) { |
1386
|
|
|
// Context sensitive keywords support. |
1387
|
|
View Code Duplication |
for ($x = ($i + 1); $i < $numTokens; $x++) { |
|
|
|
|
1388
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
1389
|
|
|
// Non-whitespace content. |
1390
|
|
|
break; |
1391
|
|
|
} |
1392
|
|
|
} |
1393
|
|
|
|
1394
|
|
View Code Duplication |
if ($this->tokens[$x]['code'] !== T_STRING) { |
|
|
|
|
1395
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1396
|
|
|
$line = $this->tokens[$x]['line']; |
1397
|
|
|
$type = $this->tokens[$x]['type']; |
1398
|
|
|
echo "\t* token $x on line $line changed from $type to T_STRING".PHP_EOL; |
1399
|
|
|
} |
1400
|
|
|
|
1401
|
|
|
$this->tokens[$x]['code'] = T_STRING; |
1402
|
|
|
$this->tokens[$x]['type'] = 'T_STRING'; |
1403
|
|
|
} |
1404
|
|
|
} else if ($this->tokens[$i]['code'] === T_PAAMAYIM_NEKUDOTAYIM) { |
1405
|
|
|
// Context sensitive keywords support. |
1406
|
|
View Code Duplication |
for ($x = ($i + 1); $i < $numTokens; $x++) { |
|
|
|
|
1407
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
1408
|
|
|
// Non-whitespace content. |
1409
|
|
|
break; |
1410
|
|
|
} |
1411
|
|
|
} |
1412
|
|
|
|
1413
|
|
View Code Duplication |
if (in_array($this->tokens[$x]['code'], array(T_STRING, T_VARIABLE, T_DOLLAR), true) === false) { |
|
|
|
|
1414
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1415
|
|
|
$line = $this->tokens[$x]['line']; |
1416
|
|
|
$type = $this->tokens[$x]['type']; |
1417
|
|
|
echo "\t* token $x on line $line changed from $type to T_STRING".PHP_EOL; |
1418
|
|
|
} |
1419
|
|
|
|
1420
|
|
|
$this->tokens[$x]['code'] = T_STRING; |
1421
|
|
|
$this->tokens[$x]['type'] = 'T_STRING'; |
1422
|
|
|
} |
1423
|
|
|
}//end if |
1424
|
|
|
|
1425
|
|
|
if (($this->tokens[$i]['code'] !== T_CASE |
1426
|
|
|
&& $this->tokens[$i]['code'] !== T_DEFAULT) |
1427
|
|
|
|| isset($this->tokens[$i]['scope_opener']) === false |
1428
|
|
|
) { |
1429
|
|
|
// Only interested in CASE and DEFAULT statements from here on in. |
1430
|
|
|
continue; |
1431
|
|
|
} |
1432
|
|
|
|
1433
|
|
|
$scopeOpener = $this->tokens[$i]['scope_opener']; |
1434
|
|
|
$scopeCloser = $this->tokens[$i]['scope_closer']; |
1435
|
|
|
|
1436
|
|
|
// If the first char after the opener is a curly brace |
1437
|
|
|
// and that brace has been ignored, it is actually |
1438
|
|
|
// opening this case statement and the opener and closer are |
1439
|
|
|
// probably set incorrectly. |
1440
|
|
View Code Duplication |
for ($x = ($scopeOpener + 1); $x < $numTokens; $x++) { |
|
|
|
|
1441
|
|
|
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
1442
|
|
|
// Non-whitespace content. |
1443
|
|
|
break; |
1444
|
|
|
} |
1445
|
|
|
} |
1446
|
|
|
|
1447
|
|
|
if ($this->tokens[$x]['code'] === T_CASE || $this->tokens[$x]['code'] === T_DEFAULT) { |
1448
|
|
|
// Special case for multiple CASE statements that share the same |
1449
|
|
|
// closer. Because we are going backwards through the file, this next |
1450
|
|
|
// CASE statement is already fixed, so just use its closer and don't |
1451
|
|
|
// worry about fixing anything. |
1452
|
|
|
$newCloser = $this->tokens[$x]['scope_closer']; |
1453
|
|
|
$this->tokens[$i]['scope_closer'] = $newCloser; |
1454
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1455
|
|
|
$oldType = $this->tokens[$scopeCloser]['type']; |
1456
|
|
|
$newType = $this->tokens[$newCloser]['type']; |
1457
|
|
|
$line = $this->tokens[$i]['line']; |
1458
|
|
|
echo "\t* token $i (T_CASE) on line $line closer changed from $scopeCloser ($oldType) to $newCloser ($newType)".PHP_EOL; |
1459
|
|
|
} |
1460
|
|
|
|
1461
|
|
|
continue; |
1462
|
|
|
} |
1463
|
|
|
|
1464
|
|
|
if ($this->tokens[$x]['code'] !== T_OPEN_CURLY_BRACKET |
1465
|
|
|
|| isset($this->tokens[$x]['scope_condition']) === true |
1466
|
|
|
) { |
1467
|
|
|
// Not a CASE/DEFAULT with a curly brace opener. |
1468
|
|
|
continue; |
1469
|
|
|
} |
1470
|
|
|
|
1471
|
|
|
// The closer for this CASE/DEFAULT should be the closing curly brace and |
1472
|
|
|
// not whatever it already is. The opener needs to be the opening curly |
1473
|
|
|
// brace so everything matches up. |
1474
|
|
|
$newCloser = $this->tokens[$x]['bracket_closer']; |
1475
|
|
View Code Duplication |
foreach (array($i, $x, $newCloser) as $index) { |
|
|
|
|
1476
|
|
|
$this->tokens[$index]['scope_condition'] = $i; |
1477
|
|
|
$this->tokens[$index]['scope_opener'] = $x; |
1478
|
|
|
$this->tokens[$index]['scope_closer'] = $newCloser; |
1479
|
|
|
} |
1480
|
|
|
|
1481
|
|
|
unset($this->tokens[$scopeOpener]['scope_condition']); |
1482
|
|
|
unset($this->tokens[$scopeOpener]['scope_opener']); |
1483
|
|
|
unset($this->tokens[$scopeOpener]['scope_closer']); |
1484
|
|
|
unset($this->tokens[$scopeCloser]['scope_condition']); |
1485
|
|
|
unset($this->tokens[$scopeCloser]['scope_opener']); |
1486
|
|
|
unset($this->tokens[$scopeCloser]['scope_closer']); |
1487
|
|
|
unset($this->tokens[$x]['bracket_opener']); |
1488
|
|
|
unset($this->tokens[$x]['bracket_closer']); |
1489
|
|
|
unset($this->tokens[$newCloser]['bracket_opener']); |
1490
|
|
|
unset($this->tokens[$newCloser]['bracket_closer']); |
1491
|
|
|
$this->tokens[$scopeCloser]['conditions'][] = $i; |
1492
|
|
|
|
1493
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1494
|
|
|
$line = $this->tokens[$i]['line']; |
1495
|
|
|
$tokenType = $this->tokens[$i]['type']; |
1496
|
|
|
|
1497
|
|
|
$oldType = $this->tokens[$scopeOpener]['type']; |
1498
|
|
|
$newType = $this->tokens[$x]['type']; |
1499
|
|
|
echo "\t* token $i ($tokenType) on line $line opener changed from $scopeOpener ($oldType) to $x ($newType)".PHP_EOL; |
1500
|
|
|
|
1501
|
|
|
$oldType = $this->tokens[$scopeCloser]['type']; |
1502
|
|
|
$newType = $this->tokens[$newCloser]['type']; |
1503
|
|
|
echo "\t* token $i ($tokenType) on line $line closer changed from $scopeCloser ($oldType) to $newCloser ($newType)".PHP_EOL; |
1504
|
|
|
} |
1505
|
|
|
|
1506
|
|
|
// Now fix up all the tokens that think they are |
1507
|
|
|
// inside the CASE/DEFAULT statement when they are really outside. |
1508
|
|
|
for ($x = $newCloser; $x < $scopeCloser; $x++) { |
1509
|
|
|
foreach ($this->tokens[$x]['conditions'] as $num => $oldCond) { |
1510
|
|
|
if ($oldCond === $this->tokens[$i]['code']) { |
1511
|
|
|
$oldConditions = $this->tokens[$x]['conditions']; |
1512
|
|
|
unset($this->tokens[$x]['conditions'][$num]); |
1513
|
|
|
|
1514
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1515
|
|
|
$type = $this->tokens[$x]['type']; |
1516
|
|
|
$oldConds = ''; |
1517
|
|
|
foreach ($oldConditions as $condition) { |
1518
|
|
|
$oldConds .= token_name($condition).','; |
1519
|
|
|
} |
1520
|
|
|
|
1521
|
|
|
$oldConds = rtrim($oldConds, ','); |
1522
|
|
|
|
1523
|
|
|
$newConds = ''; |
1524
|
|
|
foreach ($this->tokens[$x]['conditions'] as $condition) { |
1525
|
|
|
$newConds .= token_name($condition).','; |
1526
|
|
|
} |
1527
|
|
|
|
1528
|
|
|
$newConds = rtrim($newConds, ','); |
1529
|
|
|
|
1530
|
|
|
echo "\t\t* cleaned $x ($type) *".PHP_EOL; |
1531
|
|
|
echo "\t\t\t=> conditions changed from $oldConds to $newConds".PHP_EOL; |
1532
|
|
|
} |
1533
|
|
|
|
1534
|
|
|
break; |
1535
|
|
|
}//end if |
1536
|
|
|
}//end foreach |
1537
|
|
|
}//end for |
1538
|
|
|
}//end for |
1539
|
|
|
|
1540
|
|
|
if (PHP_CodeSniffer_VERBOSITY > 1) { |
1541
|
|
|
echo "\t*** END ADDITIONAL PHP PROCESSING ***".PHP_EOL; |
1542
|
|
|
} |
1543
|
|
|
|
1544
|
|
|
}//end processAdditional() |
1545
|
|
|
|
1546
|
|
|
|
1547
|
|
|
/** |
1548
|
|
|
* Takes a token produced from <code>token_get_all()</code> and produces a |
1549
|
|
|
* more uniform token. |
1550
|
|
|
* |
1551
|
|
|
* @param string|array $token The token to convert. |
1552
|
|
|
* |
1553
|
|
|
* @return array The new token. |
1554
|
|
|
*/ |
1555
|
|
|
public static function standardiseToken($token) |
1556
|
|
|
{ |
1557
|
|
View Code Duplication |
if (isset($token[1]) === false) { |
|
|
|
|
1558
|
|
|
if (isset(self::$_resolveTokenCache[$token[0]]) === true) { |
1559
|
|
|
return self::$_resolveTokenCache[$token[0]]; |
1560
|
|
|
} |
1561
|
|
|
} else { |
1562
|
|
|
$cacheKey = null; |
1563
|
|
|
if ($token[0] === T_STRING) { |
1564
|
|
|
$cacheKey = strtolower($token[1]); |
1565
|
|
|
} else if ($token[0] !== T_CURLY_OPEN) { |
1566
|
|
|
$cacheKey = $token[0]; |
1567
|
|
|
} |
1568
|
|
|
|
1569
|
|
|
if ($cacheKey !== null && isset(self::$_resolveTokenCache[$cacheKey]) === true) { |
1570
|
|
|
$newToken = self::$_resolveTokenCache[$cacheKey]; |
1571
|
|
|
$newToken['content'] = $token[1]; |
1572
|
|
|
return $newToken; |
1573
|
|
|
} |
1574
|
|
|
} |
1575
|
|
|
|
1576
|
|
|
if (isset($token[1]) === false) { |
1577
|
|
|
return self::resolveSimpleToken($token[0]); |
1578
|
|
|
} |
1579
|
|
|
|
1580
|
|
|
if ($token[0] === T_STRING) { |
1581
|
|
|
switch ($cacheKey) { |
|
|
|
|
1582
|
|
|
case 'false': |
1583
|
|
|
$newToken['type'] = 'T_FALSE'; |
|
|
|
|
1584
|
|
|
break; |
1585
|
|
|
case 'true': |
1586
|
|
|
$newToken['type'] = 'T_TRUE'; |
|
|
|
|
1587
|
|
|
break; |
1588
|
|
|
case 'null': |
1589
|
|
|
$newToken['type'] = 'T_NULL'; |
|
|
|
|
1590
|
|
|
break; |
1591
|
|
|
case 'self': |
1592
|
|
|
$newToken['type'] = 'T_SELF'; |
|
|
|
|
1593
|
|
|
break; |
1594
|
|
|
case 'parent': |
1595
|
|
|
$newToken['type'] = 'T_PARENT'; |
|
|
|
|
1596
|
|
|
break; |
1597
|
|
|
default: |
1598
|
|
|
$newToken['type'] = 'T_STRING'; |
|
|
|
|
1599
|
|
|
break; |
1600
|
|
|
} |
1601
|
|
|
|
1602
|
|
|
$newToken['code'] = constant($newToken['type']); |
1603
|
|
|
|
1604
|
|
|
self::$_resolveTokenCache[$cacheKey] = $newToken; |
1605
|
|
|
} else if ($token[0] === T_CURLY_OPEN) { |
1606
|
|
|
$newToken = array( |
1607
|
|
|
'code' => T_OPEN_CURLY_BRACKET, |
1608
|
|
|
'type' => 'T_OPEN_CURLY_BRACKET', |
1609
|
|
|
); |
1610
|
|
|
} else { |
1611
|
|
|
$newToken = array( |
1612
|
|
|
'code' => $token[0], |
1613
|
|
|
'type' => token_name($token[0]), |
1614
|
|
|
); |
1615
|
|
|
|
1616
|
|
|
self::$_resolveTokenCache[$token[0]] = $newToken; |
1617
|
|
|
}//end if |
1618
|
|
|
|
1619
|
|
|
$newToken['content'] = $token[1]; |
1620
|
|
|
return $newToken; |
1621
|
|
|
|
1622
|
|
|
}//end standardiseToken() |
1623
|
|
|
|
1624
|
|
|
|
1625
|
|
|
/** |
1626
|
|
|
* Converts simple tokens into a format that conforms to complex tokens |
1627
|
|
|
* produced by token_get_all(). |
1628
|
|
|
* |
1629
|
|
|
* Simple tokens are tokens that are not in array form when produced from |
1630
|
|
|
* token_get_all(). |
1631
|
|
|
* |
1632
|
|
|
* @param string $token The simple token to convert. |
1633
|
|
|
* |
1634
|
|
|
* @return array The new token in array format. |
1635
|
|
|
*/ |
1636
|
|
|
public static function resolveSimpleToken($token) |
1637
|
|
|
{ |
1638
|
|
|
$newToken = array(); |
1639
|
|
|
|
1640
|
|
|
switch ($token) { |
1641
|
|
|
case '{': |
1642
|
|
|
$newToken['type'] = 'T_OPEN_CURLY_BRACKET'; |
1643
|
|
|
break; |
1644
|
|
|
case '}': |
1645
|
|
|
$newToken['type'] = 'T_CLOSE_CURLY_BRACKET'; |
1646
|
|
|
break; |
1647
|
|
|
case '[': |
1648
|
|
|
$newToken['type'] = 'T_OPEN_SQUARE_BRACKET'; |
1649
|
|
|
break; |
1650
|
|
|
case ']': |
1651
|
|
|
$newToken['type'] = 'T_CLOSE_SQUARE_BRACKET'; |
1652
|
|
|
break; |
1653
|
|
|
case '(': |
1654
|
|
|
$newToken['type'] = 'T_OPEN_PARENTHESIS'; |
1655
|
|
|
break; |
1656
|
|
|
case ')': |
1657
|
|
|
$newToken['type'] = 'T_CLOSE_PARENTHESIS'; |
1658
|
|
|
break; |
1659
|
|
|
case ':': |
1660
|
|
|
$newToken['type'] = 'T_COLON'; |
1661
|
|
|
break; |
1662
|
|
|
case '.': |
1663
|
|
|
$newToken['type'] = 'T_STRING_CONCAT'; |
1664
|
|
|
break; |
1665
|
|
|
case '?': |
1666
|
|
|
$newToken['type'] = 'T_INLINE_THEN'; |
1667
|
|
|
break; |
1668
|
|
|
case ';': |
1669
|
|
|
$newToken['type'] = 'T_SEMICOLON'; |
1670
|
|
|
break; |
1671
|
|
|
case '=': |
1672
|
|
|
$newToken['type'] = 'T_EQUAL'; |
1673
|
|
|
break; |
1674
|
|
|
case '*': |
1675
|
|
|
$newToken['type'] = 'T_MULTIPLY'; |
1676
|
|
|
break; |
1677
|
|
|
case '/': |
1678
|
|
|
$newToken['type'] = 'T_DIVIDE'; |
1679
|
|
|
break; |
1680
|
|
|
case '+': |
1681
|
|
|
$newToken['type'] = 'T_PLUS'; |
1682
|
|
|
break; |
1683
|
|
|
case '-': |
1684
|
|
|
$newToken['type'] = 'T_MINUS'; |
1685
|
|
|
break; |
1686
|
|
|
case '%': |
1687
|
|
|
$newToken['type'] = 'T_MODULUS'; |
1688
|
|
|
break; |
1689
|
|
|
case '^': |
1690
|
|
|
$newToken['type'] = 'T_BITWISE_XOR'; |
1691
|
|
|
break; |
1692
|
|
|
case '&': |
1693
|
|
|
$newToken['type'] = 'T_BITWISE_AND'; |
1694
|
|
|
break; |
1695
|
|
|
case '|': |
1696
|
|
|
$newToken['type'] = 'T_BITWISE_OR'; |
1697
|
|
|
break; |
1698
|
|
|
case '<': |
1699
|
|
|
$newToken['type'] = 'T_LESS_THAN'; |
1700
|
|
|
break; |
1701
|
|
|
case '>': |
1702
|
|
|
$newToken['type'] = 'T_GREATER_THAN'; |
1703
|
|
|
break; |
1704
|
|
|
case '!': |
1705
|
|
|
$newToken['type'] = 'T_BOOLEAN_NOT'; |
1706
|
|
|
break; |
1707
|
|
|
case ',': |
1708
|
|
|
$newToken['type'] = 'T_COMMA'; |
1709
|
|
|
break; |
1710
|
|
|
case '@': |
1711
|
|
|
$newToken['type'] = 'T_ASPERAND'; |
1712
|
|
|
break; |
1713
|
|
|
case '$': |
1714
|
|
|
$newToken['type'] = 'T_DOLLAR'; |
1715
|
|
|
break; |
1716
|
|
|
case '`': |
1717
|
|
|
$newToken['type'] = 'T_BACKTICK'; |
1718
|
|
|
break; |
1719
|
|
|
default: |
1720
|
|
|
$newToken['type'] = 'T_NONE'; |
1721
|
|
|
break; |
1722
|
|
|
}//end switch |
1723
|
|
|
|
1724
|
|
|
$newToken['code'] = constant($newToken['type']); |
1725
|
|
|
$newToken['content'] = $token; |
1726
|
|
|
|
1727
|
|
|
self::$_resolveTokenCache[$token] = $newToken; |
1728
|
|
|
return $newToken; |
1729
|
|
|
|
1730
|
|
|
}//end resolveSimpleToken() |
1731
|
|
|
|
1732
|
|
|
|
1733
|
|
|
}//end class |
1734
|
|
|
|
If you define a variable conditionally, it can happen that it is not defined for all execution paths.
Let’s take a look at an example:
In the above example, the variable $x is defined if you pass “foo” or “bar” as argument for $a. However, since the switch statement has no default case statement, if you pass any other value, the variable $x would be undefined.
Available Fixes
Check for existence of the variable explicitly:
Define a default value for the variable:
Add a value for the missing path: