1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
/** |
4
|
|
|
* @package s9e\TextFormatter |
5
|
|
|
* @copyright Copyright (c) The s9e authors |
6
|
|
|
* @license http://www.opensource.org/licenses/mit-license.php The MIT License |
7
|
|
|
*/ |
8
|
|
|
namespace s9e\TextFormatter\Plugins\FancyPants; |
9
|
|
|
|
10
|
|
|
use s9e\TextFormatter\Plugins\ParserBase; |
11
|
|
|
|
12
|
|
|
class Parser extends ParserBase |
13
|
|
|
{ |
14
|
|
|
/** |
15
|
|
|
* @var bool Whether currrent test contains a double quote character |
16
|
|
|
*/ |
17
|
|
|
protected $hasDoubleQuote; |
18
|
|
|
|
19
|
|
|
/** |
20
|
|
|
* @var bool Whether currrent test contains a single quote character |
21
|
|
|
*/ |
22
|
|
|
protected $hasSingleQuote; |
23
|
|
|
|
24
|
|
|
/** |
25
|
|
|
* @var string Text being parsed |
26
|
|
|
*/ |
27
|
|
|
protected $text; |
28
|
|
|
|
29
|
|
|
/** |
30
|
|
|
* {@inheritdoc} |
31
|
|
|
*/ |
32
|
78 |
|
public function parse($text, array $matches) |
33
|
|
|
{ |
34
|
78 |
|
$this->text = $text; |
35
|
78 |
|
$this->hasSingleQuote = (strpos($text, "'") !== false); |
36
|
78 |
|
$this->hasDoubleQuote = (strpos($text, '"') !== false); |
37
|
|
|
|
38
|
78 |
|
if (empty($this->config['disableQuotes'])) |
39
|
|
|
{ |
40
|
76 |
|
$this->parseSingleQuotes(); |
41
|
76 |
|
$this->parseSingleQuotePairs(); |
42
|
76 |
|
$this->parseDoubleQuotePairs(); |
43
|
|
|
} |
44
|
78 |
|
if (empty($this->config['disableGuillemets'])) |
45
|
|
|
{ |
46
|
77 |
|
$this->parseGuillemets(); |
47
|
|
|
} |
48
|
78 |
|
if (empty($this->config['disableMathSymbols'])) |
49
|
|
|
{ |
50
|
77 |
|
$this->parseNotEqualSign(); |
51
|
77 |
|
$this->parseSymbolsAfterDigits(); |
52
|
77 |
|
$this->parseFractions(); |
53
|
|
|
} |
54
|
78 |
|
if (empty($this->config['disablePunctuation'])) |
55
|
|
|
{ |
56
|
77 |
|
$this->parseDashesAndEllipses(); |
57
|
|
|
} |
58
|
78 |
|
if (empty($this->config['disableSymbols'])) |
59
|
|
|
{ |
60
|
77 |
|
$this->parseSymbolsInParentheses(); |
61
|
|
|
} |
62
|
|
|
|
63
|
78 |
|
unset($this->text); |
64
|
|
|
} |
65
|
|
|
|
66
|
|
|
/** |
67
|
|
|
* Add a fancy replacement tag |
68
|
|
|
* |
69
|
|
|
* @param integer $tagPos Position of the tag in the text |
70
|
|
|
* @param integer $tagLen Length of text consumed by the tag |
71
|
|
|
* @param string $chr Replacement character |
72
|
|
|
* @param integer $prio Tag's priority |
73
|
|
|
* @return \s9e\TextFormatter\Parser\Tag |
74
|
|
|
*/ |
75
|
72 |
|
protected function addTag($tagPos, $tagLen, $chr, $prio = 0) |
76
|
|
|
{ |
77
|
72 |
|
$tag = $this->parser->addSelfClosingTag($this->config['tagName'], $tagPos, $tagLen, $prio); |
78
|
72 |
|
$tag->setAttribute($this->config['attrName'], $chr); |
79
|
|
|
|
80
|
72 |
|
return $tag; |
81
|
|
|
} |
82
|
|
|
|
83
|
|
|
/** |
84
|
|
|
* Parse dashes and ellipses |
85
|
|
|
* |
86
|
|
|
* Does en dash –, em dash — and ellipsis … |
87
|
|
|
* |
88
|
|
|
* @return void |
89
|
|
|
*/ |
90
|
77 |
|
protected function parseDashesAndEllipses() |
91
|
|
|
{ |
92
|
77 |
|
if (strpos($this->text, '...') === false && strpos($this->text, '--') === false) |
93
|
|
|
{ |
94
|
64 |
|
return; |
95
|
|
|
} |
96
|
|
|
|
97
|
|
|
$chrs = [ |
98
|
13 |
|
'--' => "\xE2\x80\x93", |
99
|
|
|
'---' => "\xE2\x80\x94", |
100
|
|
|
'...' => "\xE2\x80\xA6" |
101
|
|
|
]; |
102
|
13 |
|
$regexp = '/---?|\\.\\.\\./S'; |
103
|
13 |
|
preg_match_all($regexp, $this->text, $matches, PREG_OFFSET_CAPTURE); |
104
|
13 |
|
foreach ($matches[0] as $m) |
105
|
|
|
{ |
106
|
13 |
|
$this->addTag($m[1], strlen($m[0]), $chrs[$m[0]]); |
107
|
|
|
} |
108
|
|
|
} |
109
|
|
|
|
110
|
|
|
/** |
111
|
|
|
* Parse pairs of double quotes |
112
|
|
|
* |
113
|
|
|
* Does quote pairs “” -- must be done separately to handle nesting |
114
|
|
|
* |
115
|
|
|
* @return void |
116
|
|
|
*/ |
117
|
76 |
|
protected function parseDoubleQuotePairs() |
118
|
|
|
{ |
119
|
76 |
|
if ($this->hasDoubleQuote) |
120
|
|
|
{ |
121
|
11 |
|
$this->parseQuotePairs( |
122
|
11 |
|
'/(?<![0-9\\pL])"[^"\\n]+"(?![0-9\\pL])/uS', |
123
|
11 |
|
"\xE2\x80\x9C", |
124
|
11 |
|
"\xE2\x80\x9D" |
125
|
|
|
); |
126
|
|
|
} |
127
|
|
|
} |
128
|
|
|
|
129
|
|
|
/** |
130
|
|
|
* Parse vulgar fractions |
131
|
|
|
* |
132
|
|
|
* @return void |
133
|
|
|
*/ |
134
|
77 |
|
protected function parseFractions() |
135
|
|
|
{ |
136
|
77 |
|
if (strpos($this->text, '/') === false) |
137
|
|
|
{ |
138
|
74 |
|
return; |
139
|
|
|
} |
140
|
|
|
|
141
|
|
|
$map = [ |
142
|
3 |
|
'1/4' => "\xC2\xBC", |
143
|
|
|
'1/2' => "\xC2\xBD", |
144
|
|
|
'3/4' => "\xC2\xBE", |
145
|
|
|
'1/7' => "\xE2\x85\x90", |
146
|
|
|
'1/9' => "\xE2\x85\x91", |
147
|
|
|
'1/10' => "\xE2\x85\x92", |
148
|
|
|
'1/3' => "\xE2\x85\x93", |
149
|
|
|
'2/3' => "\xE2\x85\x94", |
150
|
|
|
'1/5' => "\xE2\x85\x95", |
151
|
|
|
'2/5' => "\xE2\x85\x96", |
152
|
|
|
'3/5' => "\xE2\x85\x97", |
153
|
|
|
'4/5' => "\xE2\x85\x98", |
154
|
|
|
'1/6' => "\xE2\x85\x99", |
155
|
|
|
'5/6' => "\xE2\x85\x9A", |
156
|
|
|
'1/8' => "\xE2\x85\x9B", |
157
|
|
|
'3/8' => "\xE2\x85\x9C", |
158
|
|
|
'5/8' => "\xE2\x85\x9D", |
159
|
|
|
'7/8' => "\xE2\x85\x9E", |
160
|
|
|
'0/3' => "\xE2\x86\x89" |
161
|
|
|
]; |
162
|
|
|
|
163
|
3 |
|
$regexp = '/\\b(?:0\\/3|1\\/(?:[2-9]|10)|2\\/[35]|3\\/[458]|4\\/5|5\\/[68]|7\\/8)\\b/S'; |
164
|
3 |
|
preg_match_all($regexp, $this->text, $matches, PREG_OFFSET_CAPTURE); |
165
|
3 |
|
foreach ($matches[0] as $m) |
166
|
|
|
{ |
167
|
2 |
|
$this->addTag($m[1], strlen($m[0]), $map[$m[0]]); |
168
|
|
|
} |
169
|
|
|
} |
170
|
|
|
|
171
|
|
|
/** |
172
|
|
|
* Parse guillemets-style quotation marks |
173
|
|
|
* |
174
|
|
|
* @return void |
175
|
|
|
*/ |
176
|
77 |
|
protected function parseGuillemets() |
177
|
|
|
{ |
178
|
77 |
|
if (strpos($this->text, '<<') === false) |
179
|
|
|
{ |
180
|
74 |
|
return; |
181
|
|
|
} |
182
|
|
|
|
183
|
3 |
|
$regexp = '/<<( ?)(?! )[^\\n<>]*?[^\\n <>]\\1>>(?!>)/'; |
184
|
3 |
|
preg_match_all($regexp, $this->text, $matches, PREG_OFFSET_CAPTURE); |
185
|
3 |
|
foreach ($matches[0] as $m) |
186
|
|
|
{ |
187
|
2 |
|
$left = $this->addTag($m[1], 2, "\xC2\xAB"); |
188
|
2 |
|
$right = $this->addTag($m[1] + strlen($m[0]) - 2, 2, "\xC2\xBB"); |
189
|
|
|
|
190
|
2 |
|
$left->cascadeInvalidationTo($right); |
191
|
|
|
} |
192
|
|
|
} |
193
|
|
|
|
194
|
|
|
/** |
195
|
|
|
* Parse the not equal sign |
196
|
|
|
* |
197
|
|
|
* Supports != and =/= |
198
|
|
|
* |
199
|
|
|
* @return void |
200
|
|
|
*/ |
201
|
77 |
|
protected function parseNotEqualSign() |
202
|
|
|
{ |
203
|
77 |
|
if (strpos($this->text, '!=') === false && strpos($this->text, '=/=') === false) |
204
|
|
|
{ |
205
|
75 |
|
return; |
206
|
|
|
} |
207
|
|
|
|
208
|
2 |
|
$regexp = '/\\b (?:!|=\\/)=(?= \\b)/'; |
209
|
2 |
|
preg_match_all($regexp, $this->text, $matches, PREG_OFFSET_CAPTURE); |
210
|
2 |
|
foreach ($matches[0] as $m) |
211
|
|
|
{ |
212
|
2 |
|
$this->addTag($m[1] + 1, strlen($m[0]) - 1, "\xE2\x89\xA0"); |
213
|
|
|
} |
214
|
|
|
} |
215
|
|
|
|
216
|
|
|
/** |
217
|
|
|
* Parse pairs of quotes |
218
|
|
|
* |
219
|
|
|
* @param string $regexp Regexp used to identify quote pairs |
220
|
|
|
* @param string $leftQuote Fancy replacement for left quote |
221
|
|
|
* @param string $rightQuote Fancy replacement for right quote |
222
|
|
|
* @return void |
223
|
|
|
*/ |
224
|
37 |
|
protected function parseQuotePairs($regexp, $leftQuote, $rightQuote) |
225
|
|
|
{ |
226
|
37 |
|
preg_match_all($regexp, $this->text, $matches, PREG_OFFSET_CAPTURE); |
227
|
37 |
|
foreach ($matches[0] as $m) |
228
|
|
|
{ |
229
|
10 |
|
$left = $this->addTag($m[1], 1, $leftQuote); |
230
|
10 |
|
$right = $this->addTag($m[1] + strlen($m[0]) - 1, 1, $rightQuote); |
231
|
|
|
|
232
|
|
|
// Cascade left tag's invalidation to the right so that if we skip the left quote, |
233
|
|
|
// the right quote remains untouched |
234
|
10 |
|
$left->cascadeInvalidationTo($right); |
235
|
|
|
} |
236
|
|
|
} |
237
|
|
|
|
238
|
|
|
/** |
239
|
|
|
* Parse pairs of single quotes |
240
|
|
|
* |
241
|
|
|
* Does quote pairs ‘’ must be done separately to handle nesting |
242
|
|
|
* |
243
|
|
|
* @return void |
244
|
|
|
*/ |
245
|
76 |
|
protected function parseSingleQuotePairs() |
246
|
|
|
{ |
247
|
76 |
|
if ($this->hasSingleQuote) |
248
|
|
|
{ |
249
|
30 |
|
$this->parseQuotePairs( |
250
|
30 |
|
"/(?<![0-9\\pL])'[^'\\n]+'(?![0-9\\pL])/uS", |
251
|
30 |
|
"\xE2\x80\x98", |
252
|
30 |
|
"\xE2\x80\x99" |
253
|
|
|
); |
254
|
|
|
} |
255
|
|
|
} |
256
|
|
|
|
257
|
|
|
/** |
258
|
|
|
* Parse single quotes in general |
259
|
|
|
* |
260
|
|
|
* Does apostrophes ’ after a letter or at the beginning of a word or a couple of digits |
261
|
|
|
* |
262
|
|
|
* @return void |
263
|
|
|
*/ |
264
|
76 |
|
protected function parseSingleQuotes() |
265
|
|
|
{ |
266
|
76 |
|
if (!$this->hasSingleQuote) |
267
|
|
|
{ |
268
|
46 |
|
return; |
269
|
|
|
} |
270
|
|
|
|
271
|
30 |
|
$regexp = "/(?<=\\pL)'|(?<!\\S)'(?=\\pL|[0-9]{2})/uS"; |
272
|
30 |
|
preg_match_all($regexp, $this->text, $matches, PREG_OFFSET_CAPTURE); |
273
|
30 |
|
foreach ($matches[0] as $m) |
274
|
|
|
{ |
275
|
|
|
// Give this tag a worse priority than default so that quote pairs take precedence |
276
|
20 |
|
$this->addTag($m[1], 1, "\xE2\x80\x99", 10); |
277
|
|
|
} |
278
|
|
|
} |
279
|
|
|
|
280
|
|
|
/** |
281
|
|
|
* Parse symbols found after digits |
282
|
|
|
* |
283
|
|
|
* Does symbols found after a digit: |
284
|
|
|
* - apostrophe ’ if it's followed by an "s" as in 80's |
285
|
|
|
* - prime ′ and double prime ″ |
286
|
|
|
* - multiply sign × if it's followed by an optional space and another digit |
287
|
|
|
* |
288
|
|
|
* @return void |
289
|
|
|
*/ |
290
|
77 |
|
protected function parseSymbolsAfterDigits() |
291
|
|
|
{ |
292
|
77 |
|
if (!$this->hasSingleQuote && !$this->hasDoubleQuote && strpos($this->text, 'x') === false) |
293
|
|
|
{ |
294
|
34 |
|
return; |
295
|
|
|
} |
296
|
|
|
|
297
|
|
|
$map = [ |
298
|
|
|
// 80's -- use an apostrophe |
299
|
43 |
|
"'s" => "\xE2\x80\x99", |
300
|
|
|
// 12' or 12" -- use a prime |
301
|
|
|
"'" => "\xE2\x80\xB2", |
302
|
|
|
"' " => "\xE2\x80\xB2", |
303
|
|
|
"'x" => "\xE2\x80\xB2", |
304
|
|
|
'"' => "\xE2\x80\xB3", |
305
|
|
|
'" ' => "\xE2\x80\xB3", |
306
|
|
|
'"x' => "\xE2\x80\xB3" |
307
|
|
|
]; |
308
|
|
|
|
309
|
43 |
|
$regexp = "/[0-9](?>'s|[\"']? ?x(?= ?[0-9])|[\"'])/S"; |
310
|
43 |
|
preg_match_all($regexp, $this->text, $matches, PREG_OFFSET_CAPTURE); |
311
|
43 |
|
foreach ($matches[0] as $m) |
312
|
|
|
{ |
313
|
|
|
// Test for a multiply sign at the end |
314
|
16 |
|
if (substr($m[0], -1) === 'x') |
315
|
|
|
{ |
316
|
10 |
|
$this->addTag($m[1] + strlen($m[0]) - 1, 1, "\xC3\x97"); |
317
|
|
|
} |
318
|
|
|
|
319
|
|
|
// Test for an apostrophe/prime right after the digit |
320
|
16 |
|
$str = substr($m[0], 1, 2); |
321
|
16 |
|
if (isset($map[$str])) |
322
|
|
|
{ |
323
|
11 |
|
$this->addTag($m[1] + 1, 1, $map[$str]); |
324
|
|
|
} |
325
|
|
|
} |
326
|
|
|
} |
327
|
|
|
|
328
|
|
|
/** |
329
|
|
|
* Parse symbols found in parentheses such as (c) |
330
|
|
|
* |
331
|
|
|
* Does symbols ©, ® and ™ |
332
|
|
|
* |
333
|
|
|
* @return void |
334
|
|
|
*/ |
335
|
77 |
|
protected function parseSymbolsInParentheses() |
336
|
|
|
{ |
337
|
77 |
|
if (strpos($this->text, '(') === false) |
338
|
|
|
{ |
339
|
65 |
|
return; |
340
|
|
|
} |
341
|
|
|
|
342
|
|
|
$chrs = [ |
343
|
12 |
|
'(c)' => "\xC2\xA9", |
344
|
|
|
'(r)' => "\xC2\xAE", |
345
|
|
|
'(tm)' => "\xE2\x84\xA2" |
346
|
|
|
]; |
347
|
12 |
|
$regexp = '/\\((?>c|r|tm)\\)/i'; |
348
|
12 |
|
preg_match_all($regexp, $this->text, $matches, PREG_OFFSET_CAPTURE); |
349
|
12 |
|
foreach ($matches[0] as $m) |
350
|
|
|
{ |
351
|
12 |
|
$this->addTag($m[1], strlen($m[0]), $chrs[strtr($m[0], 'CMRT', 'cmrt')]); |
352
|
|
|
} |
353
|
|
|
} |
354
|
|
|
} |