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\MediaEmbed; |
9
|
|
|
|
10
|
|
|
use InvalidArgumentException; |
11
|
|
|
use RuntimeException; |
12
|
|
|
use s9e\TextFormatter\Configurator\Helpers\FilterHelper; |
13
|
|
|
use s9e\TextFormatter\Configurator\Items\Regexp; |
14
|
|
|
use s9e\TextFormatter\Configurator\Items\Tag; |
15
|
|
|
use s9e\TextFormatter\Configurator\JavaScript\Dictionary; |
16
|
|
|
use s9e\TextFormatter\Plugins\ConfiguratorBase; |
17
|
|
|
use s9e\TextFormatter\Plugins\MediaEmbed\Configurator\Collections\CachedDefinitionCollection; |
18
|
|
|
use s9e\TextFormatter\Plugins\MediaEmbed\Configurator\TemplateBuilder; |
19
|
|
|
|
20
|
|
|
class Configurator extends ConfiguratorBase |
21
|
|
|
{ |
22
|
|
|
/** |
23
|
|
|
* @var array List of filters that are explicitly allowed in attribute definitions |
24
|
|
|
*/ |
25
|
|
|
public $allowedFilters = ['htmlspecialchars_decode', 'stripslashes', 'str_replace', 'urldecode']; |
26
|
|
|
|
27
|
|
|
/** |
28
|
|
|
* @var bool Whether to create the MEDIA BBCode |
29
|
|
|
*/ |
30
|
|
|
protected $createMediaBBCode = true; |
31
|
|
|
|
32
|
|
|
/** |
33
|
|
|
* @var Configurator\Collections\SiteDefinitionCollection Default sites |
34
|
|
|
*/ |
35
|
|
|
public $defaultSites; |
36
|
|
|
|
37
|
|
|
/** |
38
|
|
|
* {@inheritdoc} |
39
|
|
|
*/ |
40
|
|
|
protected $quickMatch = '://'; |
41
|
|
|
|
42
|
|
|
/** |
43
|
|
|
* {@inheritdoc} |
44
|
|
|
*/ |
45
|
|
|
protected $regexp = '/\\bhttps?:\\/\\/[^["\'\\s]+/Si'; |
46
|
|
|
|
47
|
|
|
/** |
48
|
|
|
* @var array Configured sites |
49
|
|
|
*/ |
50
|
|
|
protected $sites = []; |
51
|
|
|
|
52
|
|
|
/** |
53
|
|
|
* @var string Name of the tag used to handle embeddable URLs |
54
|
|
|
*/ |
55
|
|
|
protected $tagName = 'MEDIA'; |
56
|
|
|
|
57
|
|
|
/** |
58
|
|
|
* @var TemplateBuilder |
59
|
|
|
*/ |
60
|
|
|
protected $templateBuilder; |
61
|
|
|
|
62
|
|
|
/** |
63
|
|
|
* {@inheritdoc} |
64
|
|
|
*/ |
65
|
25 |
|
protected function setUp() |
66
|
|
|
{ |
67
|
25 |
|
$this->defaultSites = new CachedDefinitionCollection; |
68
|
25 |
|
$this->templateBuilder = new TemplateBuilder; |
69
|
|
|
|
70
|
25 |
|
$this->configurator->registeredVars['MediaEmbed.hosts'] = new Dictionary; |
71
|
25 |
|
$this->configurator->registeredVars['MediaEmbed.sites'] = new Dictionary; |
72
|
|
|
|
73
|
|
|
// Create a MEDIA tag |
74
|
25 |
|
$this->createMediaTag(); |
75
|
|
|
|
76
|
|
|
// Create a [MEDIA] BBCode if applicable |
77
|
25 |
|
if ($this->createMediaBBCode) |
78
|
|
|
{ |
79
|
24 |
|
$this->configurator->BBCodes->set($this->tagName, ['contentAttributes' => ['url']]); |
80
|
|
|
} |
81
|
|
|
} |
82
|
|
|
|
83
|
|
|
/** |
84
|
|
|
* {@inheritdoc} |
85
|
|
|
*/ |
86
|
4 |
|
public function asConfig() |
87
|
|
|
{ |
88
|
4 |
|
if (empty($this->sites)) |
89
|
|
|
{ |
90
|
1 |
|
return; |
91
|
|
|
} |
92
|
|
|
|
93
|
|
|
return [ |
94
|
3 |
|
'quickMatch' => $this->quickMatch, |
95
|
3 |
|
'regexp' => $this->regexp, |
96
|
3 |
|
'tagName' => $this->tagName |
97
|
|
|
]; |
98
|
|
|
} |
99
|
|
|
|
100
|
|
|
/** |
101
|
|
|
* Add a media site |
102
|
|
|
* |
103
|
|
|
* @param string $siteId Site's ID |
104
|
|
|
* @param array $siteConfig Site's config |
105
|
|
|
* @return Tag Tag created for this site |
106
|
|
|
*/ |
107
|
21 |
|
public function add($siteId, ?array $siteConfig = null) |
108
|
|
|
{ |
109
|
|
|
// Normalize or retrieve the site definition |
110
|
21 |
|
$siteId = $this->normalizeId($siteId); |
111
|
20 |
|
if (isset($siteConfig)) |
112
|
|
|
{ |
113
|
15 |
|
$siteConfig = $this->defaultSites->normalizeValue($siteConfig); |
114
|
|
|
} |
115
|
|
|
else |
116
|
|
|
{ |
117
|
5 |
|
$siteConfig = $this->defaultSites->get($siteId); |
118
|
|
|
} |
119
|
19 |
|
$siteConfig['extract'] = $this->convertRegexps($siteConfig['extract']); |
120
|
19 |
|
$siteConfig['scrape'] = $this->convertScrapes($siteConfig['scrape']); |
121
|
|
|
|
122
|
|
|
// Check the safety of attribute filters |
123
|
19 |
|
$this->checkAttributeFilters($siteConfig['attributes']); |
124
|
|
|
|
125
|
|
|
// Create the tag for this site |
126
|
18 |
|
$tag = $this->addTag($siteId, $siteConfig); |
127
|
|
|
|
128
|
|
|
// Update the configurator's data |
129
|
17 |
|
$this->sites[$siteId] = $siteConfig; |
130
|
17 |
|
foreach ($siteConfig['host'] as $host) |
131
|
|
|
{ |
132
|
17 |
|
$this->configurator->registeredVars['MediaEmbed.hosts'][$host] = $siteId; |
133
|
|
|
} |
134
|
17 |
|
$this->configurator->registeredVars['MediaEmbed.sites'][$siteId] = [$siteConfig['extract'], $siteConfig['scrape']]; |
135
|
|
|
|
136
|
17 |
|
return $tag; |
137
|
|
|
} |
138
|
|
|
|
139
|
|
|
/** |
140
|
|
|
* Return the list of configured sites |
141
|
|
|
* |
142
|
|
|
* @return array Site's ID as keys, site's config as values |
143
|
|
|
*/ |
144
|
1 |
|
public function getSites() |
145
|
|
|
{ |
146
|
1 |
|
return $this->sites; |
147
|
|
|
} |
148
|
|
|
|
149
|
|
|
/** |
150
|
|
|
* Create and return a tag that handles given media site |
151
|
|
|
* |
152
|
|
|
* @param string $siteId |
153
|
|
|
* @param array $siteConfig |
154
|
|
|
* @return Tag |
155
|
|
|
*/ |
156
|
18 |
|
protected function addTag($siteId, array $siteConfig) |
157
|
|
|
{ |
158
|
18 |
|
$tag = new Tag([ |
159
|
18 |
|
'attributes' => $this->getAttributesConfig($siteConfig), |
160
|
|
|
'rules' => [ |
161
|
18 |
|
'allowChild' => 'URL', |
162
|
|
|
'autoClose' => true, |
163
|
18 |
|
'denyChild' => [$siteId, $this->tagName] |
164
|
|
|
], |
165
|
18 |
|
'template' => $this->templateBuilder->build($siteId, $siteConfig) |
166
|
|
|
]); |
167
|
|
|
|
168
|
18 |
|
$this->configurator->templateNormalizer->normalizeTag($tag); |
169
|
18 |
|
$this->configurator->templateChecker->checkTag($tag); |
170
|
17 |
|
$this->configurator->tags->add($siteId, $tag); |
171
|
|
|
|
172
|
17 |
|
return $tag; |
173
|
|
|
} |
174
|
|
|
|
175
|
|
|
/** |
176
|
|
|
* Check the safety of given attributes |
177
|
|
|
* |
178
|
|
|
* @param array $attributes |
179
|
|
|
* @return void |
180
|
|
|
*/ |
181
|
19 |
|
protected function checkAttributeFilters(array $attributes) |
182
|
|
|
{ |
183
|
19 |
|
foreach ($attributes as $attrConfig) |
184
|
|
|
{ |
185
|
7 |
|
if (empty($attrConfig['filterChain'])) |
186
|
|
|
{ |
187
|
3 |
|
continue; |
188
|
|
|
} |
189
|
6 |
|
foreach ($attrConfig['filterChain'] as $filter) |
190
|
|
|
{ |
191
|
6 |
|
if (!FilterHelper::isAllowed($filter, $this->allowedFilters)) |
192
|
|
|
{ |
193
|
1 |
|
throw new RuntimeException("Filter '$filter' is not allowed in media sites"); |
194
|
|
|
} |
195
|
|
|
} |
196
|
|
|
} |
197
|
|
|
} |
198
|
|
|
|
199
|
|
|
/** |
200
|
|
|
* Convert given regexp to a [regexp, map] pair |
201
|
|
|
* |
202
|
|
|
* @param string $regexp Original regexp |
203
|
|
|
* @return array [regexp, [list of captures' names]] |
204
|
|
|
*/ |
205
|
17 |
|
protected function convertRegexp($regexp) |
206
|
|
|
{ |
207
|
17 |
|
$regexp = new Regexp($regexp); |
208
|
|
|
|
209
|
17 |
|
return [$regexp, $regexp->getCaptureNames()]; |
210
|
|
|
} |
211
|
|
|
|
212
|
|
|
/** |
213
|
|
|
* Convert a list of regexps |
214
|
|
|
* |
215
|
|
|
* @param string[] $regexps Original list |
216
|
|
|
* @return array[] Converted list |
217
|
|
|
*/ |
218
|
19 |
|
protected function convertRegexps(array $regexps) |
219
|
|
|
{ |
220
|
19 |
|
return array_map($this->convertRegexp(...), $regexps); |
221
|
|
|
} |
222
|
|
|
|
223
|
|
|
/** |
224
|
|
|
* Convert all regexps in a scraping config |
225
|
|
|
* |
226
|
|
|
* @param array $config Original config |
227
|
|
|
* @return array Converted config |
228
|
|
|
*/ |
229
|
1 |
|
protected function convertScrapeConfig(array $config) |
230
|
|
|
{ |
231
|
1 |
|
$config['extract'] = $this->convertRegexps($config['extract']); |
232
|
1 |
|
$config['match'] = $this->convertRegexps($config['match']); |
233
|
|
|
|
234
|
1 |
|
return $config; |
235
|
|
|
} |
236
|
|
|
|
237
|
|
|
/** |
238
|
|
|
* Convert all regexps in a list of scraping configs |
239
|
|
|
* |
240
|
|
|
* @param array[] $scrapes Original config |
241
|
|
|
* @return array[] Converted config |
242
|
|
|
*/ |
243
|
19 |
|
protected function convertScrapes(array $scrapes) |
244
|
|
|
{ |
245
|
19 |
|
return array_map($this->convertScrapeConfig(...), $scrapes); |
246
|
|
|
} |
247
|
|
|
|
248
|
|
|
/** |
249
|
|
|
* Create the default MEDIA tag |
250
|
|
|
* |
251
|
|
|
* @return void |
252
|
|
|
*/ |
253
|
25 |
|
protected function createMediaTag() |
254
|
|
|
{ |
255
|
25 |
|
$tag = $this->configurator->tags->add($this->tagName); |
256
|
|
|
|
257
|
|
|
// This tag should not need to be closed and should not contain itself |
258
|
25 |
|
$tag->rules->autoClose(); |
259
|
25 |
|
$tag->rules->denyChild($this->tagName); |
260
|
|
|
|
261
|
|
|
// Empty this tag's filter chain and add our tag filter |
262
|
25 |
|
$tag->filterChain->clear(); |
263
|
25 |
|
$tag->filterChain |
264
|
25 |
|
->append(__NAMESPACE__ . '\\Parser::filterTag') |
265
|
25 |
|
->resetParameters() |
266
|
25 |
|
->addParameterByName('tag') |
267
|
25 |
|
->addParameterByName('parser') |
268
|
25 |
|
->addParameterByName('MediaEmbed.hosts') |
269
|
25 |
|
->addParameterByName('MediaEmbed.sites') |
270
|
25 |
|
->addParameterByName('cacheDir') |
271
|
25 |
|
->setJS(file_get_contents(__DIR__ . '/Parser/tagFilter.js')); |
272
|
|
|
} |
273
|
|
|
|
274
|
|
|
/** |
275
|
|
|
* Return the list of named captures from a list of [regexp, map] pairs |
276
|
|
|
* |
277
|
|
|
* @param array[] $regexps List of [regexp, map] pairs |
278
|
|
|
* @return string[] |
279
|
|
|
*/ |
280
|
18 |
|
protected function getAttributeNamesFromRegexps(array $regexps) |
281
|
|
|
{ |
282
|
18 |
|
$attrNames = []; |
283
|
18 |
|
foreach ($regexps as list($regexp, $map)) |
284
|
|
|
{ |
285
|
16 |
|
$attrNames += array_flip(array_filter($map)); |
286
|
|
|
} |
287
|
|
|
|
288
|
18 |
|
return $attrNames; |
289
|
|
|
} |
290
|
|
|
|
291
|
|
|
/** |
292
|
|
|
* Get the attributes config for given site config |
293
|
|
|
* |
294
|
|
|
* @param array $siteConfig Site's config |
295
|
|
|
* @return array Map of [attrName => attrConfig] |
296
|
|
|
*/ |
297
|
18 |
|
protected function getAttributesConfig(array $siteConfig) |
298
|
|
|
{ |
299
|
18 |
|
$attrNames = $this->getAttributeNamesFromRegexps($siteConfig['extract']); |
300
|
18 |
|
foreach ($siteConfig['scrape'] as $scrapeConfig) |
301
|
|
|
{ |
302
|
1 |
|
$attrNames += $this->getAttributeNamesFromRegexps($scrapeConfig['extract']); |
303
|
|
|
} |
304
|
|
|
|
305
|
18 |
|
$attributes = $siteConfig['attributes'] + array_fill_keys(array_keys($attrNames), []); |
306
|
18 |
|
foreach ($attributes as &$attrConfig) |
307
|
|
|
{ |
308
|
16 |
|
$attrConfig += ['required' => false]; |
309
|
|
|
} |
310
|
18 |
|
unset($attrConfig); |
311
|
|
|
|
312
|
18 |
|
return $attributes; |
313
|
|
|
} |
314
|
|
|
|
315
|
|
|
/** |
316
|
|
|
* Validate and normalize a site ID |
317
|
|
|
* |
318
|
|
|
* @param string $siteId |
319
|
|
|
* @return string |
320
|
|
|
*/ |
321
|
21 |
|
protected function normalizeId($siteId) |
322
|
|
|
{ |
323
|
21 |
|
$siteId = strtolower($siteId); |
324
|
|
|
|
325
|
21 |
|
if (!preg_match('(^[a-z0-9]+$)', $siteId)) |
326
|
|
|
{ |
327
|
1 |
|
throw new InvalidArgumentException('Invalid site ID'); |
328
|
|
|
} |
329
|
|
|
|
330
|
20 |
|
return $siteId; |
331
|
|
|
} |
332
|
|
|
} |