1
|
|
|
<?php |
2
|
|
|
/** |
3
|
|
|
* This file is part of the ZBateson\MailMimeParser project. |
4
|
|
|
* |
5
|
|
|
* @license http://opensource.org/licenses/bsd-license.php BSD |
6
|
|
|
*/ |
7
|
|
|
namespace ZBateson\MailMimeParser\Header\Consumer; |
8
|
|
|
|
9
|
|
|
use ZBateson\MailMimeParser\Header\Consumer\ConsumerService; |
10
|
|
|
use ZBateson\MailMimeParser\Header\Part\HeaderPartFactory; |
11
|
|
|
use SplFixedArray; |
12
|
|
|
use Iterator; |
13
|
|
|
use NoRewindIterator; |
14
|
|
|
|
15
|
|
|
/** |
16
|
|
|
* Abstract base class for all header token consumers. |
17
|
|
|
* |
18
|
|
|
* Defines the base parser that loops over tokens, consuming them and creating |
19
|
|
|
* header parts. |
20
|
|
|
* |
21
|
|
|
* @author Zaahid Bateson |
22
|
|
|
*/ |
23
|
|
|
abstract class AbstractConsumer |
24
|
|
|
{ |
25
|
|
|
/** |
26
|
|
|
* @var \ZBateson\MailMimeParser\Header\Consumer\ConsumerService used to |
27
|
|
|
* get consumer instances for sub-consumers |
28
|
|
|
*/ |
29
|
|
|
protected $consumerService; |
30
|
|
|
|
31
|
|
|
/** |
32
|
|
|
* @var \ZBateson\MailMimeParser\Header\Part\HeaderPartFactory used to construct |
33
|
|
|
* HeaderPart objects |
34
|
|
|
*/ |
35
|
|
|
protected $partFactory; |
36
|
|
|
|
37
|
|
|
/** |
38
|
|
|
* Initializes the instance. |
39
|
|
|
* |
40
|
|
|
* @param ConsumerService $consumerService |
41
|
|
|
* @param HeaderPartFactory $partFactory |
42
|
|
|
*/ |
43
|
1 |
|
protected function __construct(ConsumerService $consumerService, HeaderPartFactory $partFactory) |
44
|
|
|
{ |
45
|
1 |
|
$this->consumerService = $consumerService; |
46
|
1 |
|
$this->partFactory = $partFactory; |
47
|
1 |
|
} |
48
|
|
|
|
49
|
|
|
/** |
50
|
|
|
* Returns the singleton instance for the class. |
51
|
|
|
* |
52
|
|
|
* @param ConsumerService $consumerService |
53
|
|
|
* @param HeaderPartFactory $partFactory |
54
|
|
|
*/ |
55
|
31 |
|
public static function getInstance(ConsumerService $consumerService, HeaderPartFactory $partFactory) |
56
|
|
|
{ |
57
|
31 |
|
static $instances = []; |
58
|
31 |
|
$class = get_called_class(); |
59
|
31 |
|
if (!isset($instances[$class])) { |
60
|
1 |
|
$instances[$class] = new static($consumerService, $partFactory); |
61
|
1 |
|
} |
62
|
31 |
|
return $instances[$class]; |
63
|
|
|
} |
64
|
|
|
|
65
|
|
|
/** |
66
|
|
|
* Invokes parsing of a header's value into header parts. |
67
|
|
|
* |
68
|
|
|
* @param string $value the raw header value |
69
|
|
|
* @return \ZBateson\MailMimeParser\Header\Part\HeaderPart[] the array of parsed |
70
|
|
|
* parts |
71
|
|
|
*/ |
72
|
26 |
|
public function __invoke($value) |
73
|
|
|
{ |
74
|
26 |
|
if ($value !== '') { |
75
|
25 |
|
return $this->parseRawValue($value); |
76
|
|
|
} |
77
|
1 |
|
return []; |
78
|
|
|
} |
79
|
|
|
|
80
|
|
|
/** |
81
|
|
|
* Called during construction to set up the list of sub-consumers that will |
82
|
|
|
* take control from this consumer should a token match a sub-consumer's |
83
|
|
|
* start token. |
84
|
|
|
* |
85
|
|
|
* @return AbstractConsumer[] the array of consumers |
86
|
|
|
*/ |
87
|
|
|
abstract protected function getSubConsumers(); |
88
|
|
|
|
89
|
|
|
/** |
90
|
|
|
* Returns this consumer and all unique sub consumers. |
91
|
|
|
* |
92
|
|
|
* Loops into the sub-consumers (and their sub-consumers, etc...) finding |
93
|
|
|
* all unique consumers, and returns them in an array. |
94
|
|
|
* |
95
|
|
|
* @return \ZBateson\MailMimeParser\Header\AbstractConsumer[] |
96
|
|
|
*/ |
97
|
25 |
|
private function getAllConsumers() |
98
|
|
|
{ |
99
|
25 |
|
$found = [$this]; |
100
|
|
|
do { |
101
|
25 |
|
$current = current($found); |
102
|
25 |
|
$subConsumers = $current->getSubConsumers(); |
103
|
25 |
|
foreach ($subConsumers as $consumer) { |
104
|
22 |
|
if (!in_array($consumer, $found)) { |
105
|
22 |
|
$found[] = $consumer; |
106
|
22 |
|
} |
107
|
25 |
|
} |
108
|
25 |
|
} while (next($found) !== false); |
109
|
25 |
|
return $found; |
110
|
|
|
} |
111
|
|
|
|
112
|
|
|
/** |
113
|
|
|
* Called by __invoke to parse the raw header value into header parts. |
114
|
|
|
* |
115
|
|
|
* Calls splitTokens to split the value into token part strings, then calls |
116
|
|
|
* parseParts to parse the returned array. |
117
|
|
|
* |
118
|
|
|
* @param string $value |
119
|
|
|
* @return \ZBateson\MailMimeParser\Header\Part\HeaderPart[] the array of parsed |
120
|
|
|
* parts |
121
|
|
|
*/ |
122
|
25 |
|
private function parseRawValue($value) |
123
|
|
|
{ |
124
|
25 |
|
$tokens = $this->splitRawValue($value); |
125
|
25 |
|
return $this->parseTokensIntoParts(new NoRewindIterator(SplFixedArray::fromArray($tokens))); |
126
|
|
|
} |
127
|
|
|
|
128
|
|
|
/** |
129
|
|
|
* Returns an array of regular expression separators specific to this |
130
|
|
|
* consumer. The returned patterns are used to split the header value into |
131
|
|
|
* tokens for the consumer to parse into parts. |
132
|
|
|
* |
133
|
|
|
* Each array element makes part of a generated regular expression that is |
134
|
|
|
* used in a call to preg_split(). RegEx patterns can be used, and care |
135
|
|
|
* should be taken to escape special characters. |
136
|
|
|
* |
137
|
|
|
* @return string[] the array of patterns |
138
|
|
|
*/ |
139
|
|
|
abstract protected function getTokenSeparators(); |
140
|
|
|
|
141
|
|
|
/** |
142
|
|
|
* Returns a list of regular expression markers for this consumer and all |
143
|
|
|
* sub-consumers by calling 'getTokenSeparators'.. |
144
|
|
|
* |
145
|
|
|
* @return string[] an array of regular expression markers |
146
|
|
|
*/ |
147
|
25 |
|
private function getAllTokenSeparators() |
148
|
|
|
{ |
149
|
25 |
|
$markers = $this->getTokenSeparators(); |
150
|
25 |
|
$subConsumers = $this->getAllConsumers(); |
151
|
25 |
|
foreach ($subConsumers as $consumer) { |
152
|
25 |
|
$markers = array_merge($consumer->getTokenSeparators(), $markers); |
153
|
25 |
|
} |
154
|
25 |
|
return array_unique($markers); |
155
|
|
|
} |
156
|
|
|
|
157
|
|
|
/** |
158
|
|
|
* Returns a regex pattern used to split the input header string. The |
159
|
|
|
* default implementation calls getAllTokenSeparators and implodes the |
160
|
|
|
* returned array with the regex OR '|' character as its glue. |
161
|
|
|
* |
162
|
|
|
* @return string the regex pattern |
163
|
|
|
*/ |
164
|
25 |
|
protected function getTokenSplitPattern() |
165
|
|
|
{ |
166
|
25 |
|
$sChars = implode('|', $this->getAllTokenSeparators()); |
167
|
25 |
|
return '~(\\\\.|' . $sChars . ')~'; |
168
|
|
|
} |
169
|
|
|
|
170
|
|
|
/** |
171
|
|
|
* Returns an array of split tokens from the input string. |
172
|
|
|
* |
173
|
|
|
* The method calls preg_split using getTokenSplitPattern. The split |
174
|
|
|
* array will not contain any empty parts and will contain the markers. |
175
|
|
|
* |
176
|
|
|
* @param string $rawValue the raw string |
177
|
|
|
* @return array the array of tokens |
178
|
|
|
*/ |
179
|
25 |
|
protected function splitRawValue($rawValue) |
180
|
|
|
{ |
181
|
25 |
|
return preg_split( |
182
|
25 |
|
$this->getTokenSplitPattern(), |
183
|
25 |
|
$rawValue, |
184
|
25 |
|
-1, |
185
|
25 |
|
PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY |
186
|
25 |
|
); |
187
|
|
|
} |
188
|
|
|
|
189
|
|
|
/** |
190
|
|
|
* Returns true if the passed string token marks the beginning marker for |
191
|
|
|
* the current consumer. |
192
|
|
|
* |
193
|
|
|
* @param string $token the current token |
194
|
|
|
*/ |
195
|
|
|
abstract protected function isStartToken($token); |
196
|
|
|
|
197
|
|
|
/** |
198
|
|
|
* Returns true if the passed string token marks the end marker for the |
199
|
|
|
* current consumer. |
200
|
|
|
* |
201
|
|
|
* @param string $token the current token |
202
|
|
|
*/ |
203
|
|
|
abstract protected function isEndToken($token); |
204
|
|
|
|
205
|
|
|
/** |
206
|
|
|
* Constructs and returns a \ZBateson\MailMimeParser\Header\Part\HeaderPart |
207
|
|
|
* for the passed string token. If the token should be ignored, the |
208
|
|
|
* function must return null. |
209
|
|
|
* |
210
|
|
|
* The default created part uses the instance's partFactory->newInstance |
211
|
|
|
* method. |
212
|
|
|
* |
213
|
|
|
* @param string $token the token |
214
|
|
|
* @param bool $isLiteral set to true if the token represents a literal - |
215
|
|
|
* e.g. an escaped token |
216
|
|
|
* @return \ZBateson\MailMimeParser\Header\Part\HeaderPart the constructed |
217
|
|
|
* header part or null if the token should be ignored |
218
|
|
|
*/ |
219
|
13 |
|
protected function getPartForToken($token, $isLiteral) |
220
|
|
|
{ |
221
|
13 |
|
if ($isLiteral) { |
222
|
1 |
|
return $this->partFactory->newLiteralPart($token); |
223
|
13 |
|
} elseif (preg_match('/^\s+$/', $token)) { |
224
|
10 |
|
return $this->partFactory->newToken(' '); |
225
|
|
|
} |
226
|
13 |
|
return $this->partFactory->newInstance($token); |
227
|
|
|
} |
228
|
|
|
|
229
|
|
|
/** |
230
|
|
|
* Iterates through this consumer's sub-consumers checking if the current |
231
|
|
|
* token triggers a sub-consumer's start token and passes control onto that |
232
|
|
|
* sub-consumer's parseTokenIntoParts. If no sub-consumer is responsible |
233
|
|
|
* for the current token, calls getPartForToken and returns it in an array. |
234
|
|
|
* |
235
|
|
|
* @param Iterator $tokens |
236
|
|
|
* @return \ZBateson\MailMimeParser\Header\Part\HeaderPart[]|array |
237
|
|
|
*/ |
238
|
25 |
|
private function getConsumerTokenParts(Iterator $tokens) |
239
|
|
|
{ |
240
|
25 |
|
$token = $tokens->current(); |
241
|
25 |
|
$subConsumers = $this->getSubConsumers(); |
242
|
25 |
|
foreach ($subConsumers as $consumer) { |
243
|
22 |
|
if ($consumer->isStartToken($token)) { |
244
|
13 |
|
$this->advanceToNextToken($tokens, true); |
245
|
13 |
|
return $consumer->parseTokensIntoParts($tokens); |
246
|
|
|
} |
247
|
25 |
|
} |
248
|
25 |
|
return [$this->getPartForToken($token, false)]; |
249
|
|
|
} |
250
|
|
|
|
251
|
|
|
/** |
252
|
|
|
* Returns an array of \ZBateson\MailMimeParser\Header\Part\HeaderPart for |
253
|
|
|
* the current token on the iterator. |
254
|
|
|
* |
255
|
|
|
* If the current token is a start token from a sub-consumer, the sub- |
256
|
|
|
* consumer's parseTokensIntoParts method is called. |
257
|
|
|
* |
258
|
|
|
* @param Iterator $tokens |
259
|
|
|
* @return \ZBateson\MailMimeParser\Header\Part\HeaderPart[]|array |
260
|
|
|
*/ |
261
|
25 |
|
protected function getTokenParts(Iterator $tokens) |
262
|
|
|
{ |
263
|
25 |
|
$token = $tokens->current(); |
264
|
25 |
|
if (strlen($token) === 2 && $token[0] === '\\') { |
265
|
3 |
|
return [$this->getPartForToken(substr($token, 1), true)]; |
266
|
|
|
} |
267
|
25 |
|
return $this->getConsumerTokenParts($tokens); |
268
|
|
|
} |
269
|
|
|
|
270
|
|
|
/** |
271
|
|
|
* Determines if the iterator should be advanced to the next token after |
272
|
|
|
* reading tokens or finding a start token. |
273
|
|
|
* |
274
|
|
|
* The default implementation will advance for a start token, but not |
275
|
|
|
* advance on the end token of the current consumer, allowing the end token |
276
|
|
|
* to be passed up to a higher-level consumer. |
277
|
|
|
* |
278
|
|
|
* @param Iterator $tokens |
279
|
|
|
* @param bool $isStartToken |
280
|
|
|
*/ |
281
|
22 |
|
protected function advanceToNextToken(Iterator $tokens, $isStartToken) |
282
|
|
|
{ |
283
|
22 |
|
if (($isStartToken) || ($tokens->valid() && !$this->isEndToken($tokens->current()))) { |
284
|
22 |
|
$tokens->next(); |
285
|
22 |
|
} |
286
|
22 |
|
} |
287
|
|
|
|
288
|
|
|
/** |
289
|
|
|
* Iterates over the passed token Iterator and returns an array of parsed |
290
|
|
|
* \ZBateson\MailMimeParser\Header\Part\HeaderPart objects. |
291
|
|
|
* |
292
|
|
|
* The method checks each token to see if the token matches a sub-consumer's |
293
|
|
|
* start token, or if it matches the current consumer's end token to stop |
294
|
|
|
* processing. |
295
|
|
|
* |
296
|
|
|
* If a sub-consumer's start token is matched, the sub-consumer is invoked |
297
|
|
|
* and its returned parts are merged to the current consumer's header parts. |
298
|
|
|
* |
299
|
|
|
* After all tokens are read and an array of Header\Parts are constructed, |
300
|
|
|
* the array is passed to AbstractConsumer::processParts for any final |
301
|
|
|
* processing. |
302
|
|
|
* |
303
|
|
|
* @param Iterator $tokens an iterator over a string of tokens |
304
|
|
|
* @return \ZBateson\MailMimeParser\Header\Part\HeaderPart[] an array of parsed |
305
|
|
|
* parts |
306
|
|
|
*/ |
307
|
25 |
|
protected function parseTokensIntoParts(Iterator $tokens) |
308
|
|
|
{ |
309
|
25 |
|
$parts = []; |
310
|
25 |
|
while ($tokens->valid() && !$this->isEndToken($tokens->current())) { |
311
|
25 |
|
$parts = array_merge($parts, $this->getTokenParts($tokens)); |
312
|
25 |
|
$this->advanceToNextToken($tokens, false); |
313
|
25 |
|
} |
314
|
25 |
|
return $this->processParts($parts); |
315
|
|
|
} |
316
|
|
|
|
317
|
|
|
/** |
318
|
|
|
* Performs any final processing on the array of parsed parts before |
319
|
|
|
* returning it to the consumer client. |
320
|
|
|
* |
321
|
|
|
* The default implementation simply returns the passed array. |
322
|
|
|
* |
323
|
|
|
* @param \ZBateson\MailMimeParser\Header\Part\HeaderPart[] $parts |
324
|
|
|
* @return \ZBateson\MailMimeParser\Header\Part\HeaderPart[] |
325
|
|
|
*/ |
326
|
5 |
|
protected function processParts(array $parts) |
327
|
|
|
{ |
328
|
5 |
|
return $parts; |
329
|
|
|
} |
330
|
|
|
} |
331
|
|
|
|