1
|
|
|
<?php |
2
|
|
|
/** |
3
|
|
|
* Lexer adapted from Simple Test: http://sourceforge.net/projects/simpletest/ |
4
|
|
|
* For an intro to the Lexer see: |
5
|
|
|
* https://web.archive.org/web/20120125041816/http://www.phppatterns.com/docs/develop/simple_test_lexer_notes |
6
|
|
|
* |
7
|
|
|
* @author Marcus Baker http://www.lastcraft.com |
8
|
|
|
*/ |
9
|
|
|
|
10
|
|
|
namespace dokuwiki\Lexer; |
11
|
|
|
|
12
|
|
|
// FIXME move elsewhere |
13
|
|
|
define("DOKU_LEXER_ENTER", 1); |
14
|
|
|
define("DOKU_LEXER_MATCHED", 2); |
15
|
|
|
define("DOKU_LEXER_UNMATCHED", 3); |
16
|
|
|
define("DOKU_LEXER_EXIT", 4); |
17
|
|
|
define("DOKU_LEXER_SPECIAL", 5); |
18
|
|
|
|
19
|
|
|
/** |
20
|
|
|
* Accepts text and breaks it into tokens. |
21
|
|
|
* |
22
|
|
|
* Some optimisation to make the sure the content is only scanned by the PHP regex |
23
|
|
|
* parser once. Lexer modes must not start with leading underscores. |
24
|
|
|
*/ |
25
|
|
|
class Lexer |
26
|
|
|
{ |
27
|
|
|
/** @var ParallelRegex[] */ |
28
|
|
|
protected $regexes; |
29
|
|
|
/** @var \Doku_Handler */ |
30
|
|
|
protected $handler; |
31
|
|
|
/** @var StateStack */ |
32
|
|
|
protected $mode; |
33
|
|
|
/** @var array mode "rewrites" */ |
34
|
|
|
protected $mode_handlers; |
35
|
|
|
/** @var bool case sensitive? */ |
36
|
|
|
protected $case; |
37
|
|
|
|
38
|
|
|
/** |
39
|
|
|
* Sets up the lexer in case insensitive matching by default. |
40
|
|
|
* |
41
|
|
|
* @param \Doku_Handler $handler Handling strategy by reference. |
42
|
|
|
* @param string $start Starting handler. |
43
|
|
|
* @param boolean $case True for case sensitive. |
44
|
|
|
*/ |
45
|
|
|
public function __construct($handler, $start = "accept", $case = false) |
46
|
|
|
{ |
47
|
|
|
$this->case = $case; |
48
|
|
|
$this->regexes = array(); |
49
|
|
|
$this->handler = $handler; |
50
|
|
|
$this->mode = new StateStack($start); |
51
|
|
|
$this->mode_handlers = array(); |
52
|
|
|
} |
53
|
|
|
|
54
|
|
|
/** |
55
|
|
|
* Adds a token search pattern for a particular parsing mode. |
56
|
|
|
* |
57
|
|
|
* The pattern does not change the current mode. |
58
|
|
|
* |
59
|
|
|
* @param string $pattern Perl style regex, but ( and ) |
60
|
|
|
* lose the usual meaning. |
61
|
|
|
* @param string $mode Should only apply this |
62
|
|
|
* pattern when dealing with |
63
|
|
|
* this type of input. |
64
|
|
|
*/ |
65
|
|
|
public function addPattern($pattern, $mode = "accept") |
66
|
|
|
{ |
67
|
|
|
if (! isset($this->regexes[$mode])) { |
68
|
|
|
$this->regexes[$mode] = new ParallelRegex($this->case); |
69
|
|
|
} |
70
|
|
|
$this->regexes[$mode]->addPattern($pattern); |
71
|
|
|
} |
72
|
|
|
|
73
|
|
|
/** |
74
|
|
|
* Adds a pattern that will enter a new parsing mode. |
75
|
|
|
* |
76
|
|
|
* Useful for entering parenthesis, strings, tags, etc. |
77
|
|
|
* |
78
|
|
|
* @param string $pattern Perl style regex, but ( and ) lose the usual meaning. |
79
|
|
|
* @param string $mode Should only apply this pattern when dealing with this type of input. |
80
|
|
|
* @param string $new_mode Change parsing to this new nested mode. |
81
|
|
|
*/ |
82
|
|
|
public function addEntryPattern($pattern, $mode, $new_mode) |
83
|
|
|
{ |
84
|
|
|
if (! isset($this->regexes[$mode])) { |
85
|
|
|
$this->regexes[$mode] = new ParallelRegex($this->case); |
86
|
|
|
} |
87
|
|
|
$this->regexes[$mode]->addPattern($pattern, $new_mode); |
88
|
|
|
} |
89
|
|
|
|
90
|
|
|
/** |
91
|
|
|
* Adds a pattern that will exit the current mode and re-enter the previous one. |
92
|
|
|
* |
93
|
|
|
* @param string $pattern Perl style regex, but ( and ) lose the usual meaning. |
94
|
|
|
* @param string $mode Mode to leave. |
95
|
|
|
*/ |
96
|
|
|
public function addExitPattern($pattern, $mode) |
97
|
|
|
{ |
98
|
|
|
if (! isset($this->regexes[$mode])) { |
99
|
|
|
$this->regexes[$mode] = new ParallelRegex($this->case); |
100
|
|
|
} |
101
|
|
|
$this->regexes[$mode]->addPattern($pattern, "__exit"); |
102
|
|
|
} |
103
|
|
|
|
104
|
|
|
/** |
105
|
|
|
* Adds a pattern that has a special mode. |
106
|
|
|
* |
107
|
|
|
* Acts as an entry and exit pattern in one go, effectively calling a special |
108
|
|
|
* parser handler for this token only. |
109
|
|
|
* |
110
|
|
|
* @param string $pattern Perl style regex, but ( and ) lose the usual meaning. |
111
|
|
|
* @param string $mode Should only apply this pattern when dealing with this type of input. |
112
|
|
|
* @param string $special Use this mode for this one token. |
113
|
|
|
*/ |
114
|
|
|
public function addSpecialPattern($pattern, $mode, $special) |
115
|
|
|
{ |
116
|
|
|
if (! isset($this->regexes[$mode])) { |
117
|
|
|
$this->regexes[$mode] = new ParallelRegex($this->case); |
118
|
|
|
} |
119
|
|
|
$this->regexes[$mode]->addPattern($pattern, "_$special"); |
120
|
|
|
} |
121
|
|
|
|
122
|
|
|
/** |
123
|
|
|
* Adds a mapping from a mode to another handler. |
124
|
|
|
* |
125
|
|
|
* @param string $mode Mode to be remapped. |
126
|
|
|
* @param string $handler New target handler. |
127
|
|
|
*/ |
128
|
|
|
public function mapHandler($mode, $handler) |
129
|
|
|
{ |
130
|
|
|
$this->mode_handlers[$mode] = $handler; |
131
|
|
|
} |
132
|
|
|
|
133
|
|
|
/** |
134
|
|
|
* Splits the page text into tokens. |
135
|
|
|
* |
136
|
|
|
* Will fail if the handlers report an error or if no content is consumed. If successful then each |
137
|
|
|
* unparsed and parsed token invokes a call to the held listener. |
138
|
|
|
* |
139
|
|
|
* @param string $raw Raw HTML text. |
140
|
|
|
* @return boolean True on success, else false. |
141
|
|
|
*/ |
142
|
|
|
public function parse($raw) |
143
|
|
|
{ |
144
|
|
|
if (! isset($this->handler)) { |
145
|
|
|
return false; |
146
|
|
|
} |
147
|
|
|
$initialLength = strlen($raw); |
148
|
|
|
$length = $initialLength; |
149
|
|
|
$pos = 0; |
150
|
|
|
while (is_array($parsed = $this->reduce($raw))) { |
151
|
|
|
list($unmatched, $matched, $mode) = $parsed; |
152
|
|
|
$currentLength = strlen($raw); |
153
|
|
|
$matchPos = $initialLength - $currentLength - strlen($matched); |
154
|
|
|
if (! $this->dispatchTokens($unmatched, $matched, $mode, $pos, $matchPos)) { |
155
|
|
|
return false; |
156
|
|
|
} |
157
|
|
|
if ($currentLength == $length) { |
158
|
|
|
return false; |
159
|
|
|
} |
160
|
|
|
$length = $currentLength; |
161
|
|
|
$pos = $initialLength - $currentLength; |
162
|
|
|
} |
163
|
|
|
if (!$parsed) { |
164
|
|
|
return false; |
165
|
|
|
} |
166
|
|
|
return $this->invokeHandler($raw, DOKU_LEXER_UNMATCHED, $pos); |
|
|
|
|
167
|
|
|
} |
168
|
|
|
|
169
|
|
|
/** |
170
|
|
|
* Sends the matched token and any leading unmatched |
171
|
|
|
* text to the parser changing the lexer to a new |
172
|
|
|
* mode if one is listed. |
173
|
|
|
* |
174
|
|
|
* @param string $unmatched Unmatched leading portion. |
175
|
|
|
* @param string $matched Actual token match. |
176
|
|
|
* @param bool|string $mode Mode after match. A boolean false mode causes no change. |
177
|
|
|
* @param int $initialPos |
178
|
|
|
* @param int $matchPos Current byte index location in raw doc thats being parsed |
179
|
|
|
* @return boolean False if there was any error from the parser. |
180
|
|
|
*/ |
181
|
|
|
protected function dispatchTokens($unmatched, $matched, $mode = false, $initialPos, $matchPos) |
182
|
|
|
{ |
183
|
|
|
if (! $this->invokeHandler($unmatched, DOKU_LEXER_UNMATCHED, $initialPos)) { |
|
|
|
|
184
|
|
|
return false; |
185
|
|
|
} |
186
|
|
|
if ($this->isModeEnd($mode)) { |
|
|
|
|
187
|
|
|
if (! $this->invokeHandler($matched, DOKU_LEXER_EXIT, $matchPos)) { |
|
|
|
|
188
|
|
|
return false; |
189
|
|
|
} |
190
|
|
|
return $this->mode->leave(); |
191
|
|
|
} |
192
|
|
|
if ($this->isSpecialMode($mode)) { |
|
|
|
|
193
|
|
|
$this->mode->enter($this->decodeSpecial($mode)); |
|
|
|
|
194
|
|
|
if (! $this->invokeHandler($matched, DOKU_LEXER_SPECIAL, $matchPos)) { |
|
|
|
|
195
|
|
|
return false; |
196
|
|
|
} |
197
|
|
|
return $this->mode->leave(); |
198
|
|
|
} |
199
|
|
|
if (is_string($mode)) { |
200
|
|
|
$this->mode->enter($mode); |
201
|
|
|
return $this->invokeHandler($matched, DOKU_LEXER_ENTER, $matchPos); |
|
|
|
|
202
|
|
|
} |
203
|
|
|
return $this->invokeHandler($matched, DOKU_LEXER_MATCHED, $matchPos); |
|
|
|
|
204
|
|
|
} |
205
|
|
|
|
206
|
|
|
/** |
207
|
|
|
* Tests to see if the new mode is actually to leave the current mode and pop an item from the matching |
208
|
|
|
* mode stack. |
209
|
|
|
* |
210
|
|
|
* @param string $mode Mode to test. |
211
|
|
|
* @return boolean True if this is the exit mode. |
212
|
|
|
*/ |
213
|
|
|
protected function isModeEnd($mode) |
214
|
|
|
{ |
215
|
|
|
return ($mode === "__exit"); |
216
|
|
|
} |
217
|
|
|
|
218
|
|
|
/** |
219
|
|
|
* Test to see if the mode is one where this mode is entered for this token only and automatically |
220
|
|
|
* leaves immediately afterwoods. |
221
|
|
|
* |
222
|
|
|
* @param string $mode Mode to test. |
223
|
|
|
* @return boolean True if this is the exit mode. |
224
|
|
|
*/ |
225
|
|
|
protected function isSpecialMode($mode) |
226
|
|
|
{ |
227
|
|
|
return (strncmp($mode, "_", 1) == 0); |
228
|
|
|
} |
229
|
|
|
|
230
|
|
|
/** |
231
|
|
|
* Strips the magic underscore marking single token modes. |
232
|
|
|
* |
233
|
|
|
* @param string $mode Mode to decode. |
234
|
|
|
* @return string Underlying mode name. |
235
|
|
|
*/ |
236
|
|
|
protected function decodeSpecial($mode) |
237
|
|
|
{ |
238
|
|
|
return substr($mode, 1); |
239
|
|
|
} |
240
|
|
|
|
241
|
|
|
/** |
242
|
|
|
* Calls the parser method named after the current mode. |
243
|
|
|
* |
244
|
|
|
* Empty content will be ignored. The lexer has a parser handler for each mode in the lexer. |
245
|
|
|
* |
246
|
|
|
* @param string $content Text parsed. |
247
|
|
|
* @param boolean $is_match Token is recognised rather |
248
|
|
|
* than unparsed data. |
249
|
|
|
* @param int $pos Current byte index location in raw doc |
250
|
|
|
* thats being parsed |
251
|
|
|
* @return bool |
252
|
|
|
*/ |
253
|
|
|
protected function invokeHandler($content, $is_match, $pos) |
254
|
|
|
{ |
255
|
|
|
if (($content === "") || ($content === false)) { |
256
|
|
|
return true; |
257
|
|
|
} |
258
|
|
|
$handler = $this->mode->getCurrent(); |
259
|
|
|
if (isset($this->mode_handlers[$handler])) { |
260
|
|
|
$handler = $this->mode_handlers[$handler]; |
261
|
|
|
} |
262
|
|
|
|
263
|
|
|
// modes starting with plugin_ are all handled by the same |
264
|
|
|
// handler but with an additional parameter |
265
|
|
|
if (substr($handler, 0, 7)=='plugin_') { |
266
|
|
|
list($handler,$plugin) = explode('_', $handler, 2); |
267
|
|
|
return $this->handler->$handler($content, $is_match, $pos, $plugin); |
268
|
|
|
} |
269
|
|
|
|
270
|
|
|
return $this->handler->$handler($content, $is_match, $pos); |
271
|
|
|
} |
272
|
|
|
|
273
|
|
|
/** |
274
|
|
|
* Tries to match a chunk of text and if successful removes the recognised chunk and any leading |
275
|
|
|
* unparsed data. Empty strings will not be matched. |
276
|
|
|
* |
277
|
|
|
* @param string $raw The subject to parse. This is the content that will be eaten. |
278
|
|
|
* @return array|bool Three item list of unparsed content followed by the |
279
|
|
|
* recognised token and finally the action the parser is to take. |
280
|
|
|
* True if no match, false if there is a parsing error. |
281
|
|
|
*/ |
282
|
|
|
protected function reduce(&$raw) |
283
|
|
|
{ |
284
|
|
|
if (! isset($this->regexes[$this->mode->getCurrent()])) { |
285
|
|
|
return false; |
286
|
|
|
} |
287
|
|
|
if ($raw === "") { |
288
|
|
|
return true; |
289
|
|
|
} |
290
|
|
|
if ($action = $this->regexes[$this->mode->getCurrent()]->split($raw, $split)) { |
291
|
|
|
list($unparsed, $match, $raw) = $split; |
292
|
|
|
return array($unparsed, $match, $action); |
293
|
|
|
} |
294
|
|
|
return true; |
295
|
|
|
} |
296
|
|
|
|
297
|
|
|
/** |
298
|
|
|
* Escapes regex characters other than (, ) and / |
299
|
|
|
* |
300
|
|
|
* @param string $str |
301
|
|
|
* @return string |
302
|
|
|
*/ |
303
|
|
|
public static function escape($str) |
304
|
|
|
{ |
305
|
|
|
$chars = array( |
306
|
|
|
'/\\\\/', |
307
|
|
|
'/\./', |
308
|
|
|
'/\+/', |
309
|
|
|
'/\*/', |
310
|
|
|
'/\?/', |
311
|
|
|
'/\[/', |
312
|
|
|
'/\^/', |
313
|
|
|
'/\]/', |
314
|
|
|
'/\$/', |
315
|
|
|
'/\{/', |
316
|
|
|
'/\}/', |
317
|
|
|
'/\=/', |
318
|
|
|
'/\!/', |
319
|
|
|
'/\</', |
320
|
|
|
'/\>/', |
321
|
|
|
'/\|/', |
322
|
|
|
'/\:/' |
323
|
|
|
); |
324
|
|
|
|
325
|
|
|
$escaped = array( |
326
|
|
|
'\\\\\\\\', |
327
|
|
|
'\.', |
328
|
|
|
'\+', |
329
|
|
|
'\*', |
330
|
|
|
'\?', |
331
|
|
|
'\[', |
332
|
|
|
'\^', |
333
|
|
|
'\]', |
334
|
|
|
'\$', |
335
|
|
|
'\{', |
336
|
|
|
'\}', |
337
|
|
|
'\=', |
338
|
|
|
'\!', |
339
|
|
|
'\<', |
340
|
|
|
'\>', |
341
|
|
|
'\|', |
342
|
|
|
'\:' |
343
|
|
|
); |
344
|
|
|
return preg_replace($chars, $escaped, $str); |
345
|
|
|
} |
346
|
|
|
} |
347
|
|
|
|
It seems like the type of the argument is not accepted by the function/method which you are calling.
In some cases, in particular if PHP’s automatic type-juggling kicks in this might be fine. In other cases, however this might be a bug.
We suggest to add an explicit type cast like in the following example: