]>
Commit | Line | Data |
---|---|---|
558d9aab JB |
1 | <?php |
2 | ||
3 | namespace Wallabag\CoreBundle\Helper; | |
4 | ||
5 | use Graby\Graby; | |
45fd7e09 | 6 | use Psr\Log\LoggerInterface; |
8d7b4f0e | 7 | use Symfony\Component\HttpFoundation\File\MimeType\MimeTypeExtensionGuesser; |
42f3bb2c | 8 | use Symfony\Component\Validator\Constraints\Locale as LocaleConstraint; |
0d349ea6 JB |
9 | use Symfony\Component\Validator\Constraints\Url as UrlConstraint; |
10 | use Symfony\Component\Validator\Validator\ValidatorInterface; | |
f808b016 JB |
11 | use Wallabag\CoreBundle\Entity\Entry; |
12 | use Wallabag\CoreBundle\Tools\Utils; | |
558d9aab JB |
13 | |
14 | /** | |
15 | * This kind of proxy class take care of getting the content from an url | |
f1e29e69 | 16 | * and update the entry with what it found. |
558d9aab JB |
17 | */ |
18 | class ContentProxy | |
19 | { | |
20 | protected $graby; | |
c3510620 | 21 | protected $tagger; |
be54dfe4 | 22 | protected $validator; |
1c9cd2a7 | 23 | protected $logger; |
8d7b4f0e | 24 | protected $mimeGuesser; |
29dca432 | 25 | protected $fetchingErrorMessage; |
6bc6fb1f | 26 | protected $eventDispatcher; |
8a219854 | 27 | protected $storeArticleHeaders; |
558d9aab | 28 | |
709e21a3 | 29 | public function __construct(Graby $graby, RuleBasedTagger $tagger, ValidatorInterface $validator, LoggerInterface $logger, $fetchingErrorMessage, $storeArticleHeaders = false) |
558d9aab | 30 | { |
347fa6be | 31 | $this->graby = $graby; |
c3510620 | 32 | $this->tagger = $tagger; |
0d349ea6 | 33 | $this->validator = $validator; |
1c9cd2a7 | 34 | $this->logger = $logger; |
8d7b4f0e | 35 | $this->mimeGuesser = new MimeTypeExtensionGuesser(); |
29dca432 | 36 | $this->fetchingErrorMessage = $fetchingErrorMessage; |
8a219854 | 37 | $this->storeArticleHeaders = $storeArticleHeaders; |
558d9aab JB |
38 | } |
39 | ||
40 | /** | |
6acadf8e | 41 | * Update entry using either fetched or provided content. |
4d0ec0e7 | 42 | * |
6acadf8e JB |
43 | * @param Entry $entry Entry to update |
44 | * @param string $url Url of the content | |
45 | * @param array $content Array with content provided for import with AT LEAST keys title, html, url to skip the fetchContent from the url | |
46 | * @param bool $disableContentUpdate Whether to skip trying to fetch content using Graby | |
558d9aab | 47 | */ |
6acadf8e | 48 | public function updateEntry(Entry $entry, $url, array $content = [], $disableContentUpdate = false) |
558d9aab | 49 | { |
6acadf8e JB |
50 | if (!empty($content['html'])) { |
51 | $content['html'] = $this->graby->cleanupHtml($content['html'], $url); | |
d5c2cc54 | 52 | } |
e668a812 | 53 | |
6acadf8e | 54 | if ((empty($content) || false === $this->validateContent($content)) && false === $disableContentUpdate) { |
ec970721 | 55 | $fetchedContent = $this->graby->fetchContent($url); |
106bdbcd JB |
56 | |
57 | // when content is imported, we have information in $content | |
58 | // in case fetching content goes bad, we'll keep the imported information instead of overriding them | |
6acadf8e | 59 | if (empty($content) || $fetchedContent['html'] !== $this->fetchingErrorMessage) { |
29dca432 JC |
60 | $content = $fetchedContent; |
61 | } | |
4d0ec0e7 | 62 | } |
558d9aab | 63 | |
6acadf8e JB |
64 | // be sure to keep the url in case of error |
65 | // so we'll be able to refetch it in the future | |
66 | $content['url'] = !empty($content['url']) ? $content['url'] : $url; | |
67 | ||
3362b3e3 KD |
68 | // In one case (at least in tests), url is empty here |
69 | // so we set it using $url provided in the updateEntry call. | |
c801d03e KD |
70 | // Not sure what are the other possible cases where this property is empty |
71 | if (empty($entry->getUrl()) && !empty($url)) { | |
3362b3e3 KD |
72 | $entry->setUrl($url); |
73 | } | |
74 | ||
d0e9b3d6 JC |
75 | $this->stockEntry($entry, $content); |
76 | } | |
77 | ||
c18a2476 JB |
78 | /** |
79 | * Use a Symfony validator to ensure the language is well formatted. | |
80 | * | |
81 | * @param Entry $entry | |
82 | * @param string $value Language to validate and save | |
83 | */ | |
84 | public function updateLanguage(Entry $entry, $value) | |
85 | { | |
86 | // some lang are defined as fr-FR, es-ES. | |
87 | // replacing - by _ might increase language support | |
88 | $value = str_replace('-', '_', $value); | |
89 | ||
90 | $errors = $this->validator->validate( | |
91 | $value, | |
92 | (new LocaleConstraint()) | |
93 | ); | |
94 | ||
95 | if (0 === count($errors)) { | |
96 | $entry->setLanguage($value); | |
97 | ||
98 | return; | |
99 | } | |
100 | ||
101 | $this->logger->warning('Language validation failed. ' . (string) $errors); | |
102 | } | |
103 | ||
104 | /** | |
105 | * Use a Symfony validator to ensure the preview picture is a real url. | |
106 | * | |
107 | * @param Entry $entry | |
108 | * @param string $value URL to validate and save | |
109 | */ | |
110 | public function updatePreviewPicture(Entry $entry, $value) | |
111 | { | |
112 | $errors = $this->validator->validate( | |
113 | $value, | |
114 | (new UrlConstraint()) | |
115 | ); | |
116 | ||
117 | if (0 === count($errors)) { | |
118 | $entry->setPreviewPicture($value); | |
119 | ||
120 | return; | |
121 | } | |
122 | ||
123 | $this->logger->warning('PreviewPicture validation failed. ' . (string) $errors); | |
124 | } | |
125 | ||
126 | /** | |
127 | * Update date. | |
128 | * | |
129 | * @param Entry $entry | |
130 | * @param string $value Date to validate and save | |
131 | */ | |
132 | public function updatePublishedAt(Entry $entry, $value) | |
133 | { | |
ff9f89fd | 134 | $date = $value; |
c18a2476 JB |
135 | |
136 | // is it a timestamp? | |
3ef055ce | 137 | if (false !== filter_var($date, FILTER_VALIDATE_INT)) { |
ff9f89fd | 138 | $date = '@' . $date; |
c18a2476 JB |
139 | } |
140 | ||
141 | try { | |
ff9f89fd JB |
142 | // is it already a DateTime? |
143 | // (it's inside the try/catch in case of fail to be parse time string) | |
144 | if (!$date instanceof \DateTime) { | |
145 | $date = new \DateTime($date); | |
146 | } | |
147 | ||
148 | $entry->setPublishedAt($date); | |
c18a2476 JB |
149 | } catch (\Exception $e) { |
150 | $this->logger->warning('Error while defining date', ['e' => $e, 'url' => $entry->getUrl(), 'date' => $value]); | |
151 | } | |
152 | } | |
153 | ||
af29e1bf KD |
154 | /** |
155 | * Helper to extract and save host from entry url. | |
156 | * | |
157 | * @param Entry $entry | |
158 | */ | |
159 | public function setEntryDomainName(Entry $entry) | |
160 | { | |
161 | $domainName = parse_url($entry->getUrl(), PHP_URL_HOST); | |
162 | if (false !== $domainName) { | |
163 | $entry->setDomainName($domainName); | |
164 | } | |
165 | } | |
166 | ||
167 | /** | |
168 | * Helper to set a default title using: | |
169 | * - url basename, if applicable | |
170 | * - hostname. | |
171 | * | |
172 | * @param Entry $entry | |
173 | */ | |
174 | public function setDefaultEntryTitle(Entry $entry) | |
175 | { | |
176 | $url = parse_url($entry->getUrl()); | |
177 | $path = pathinfo($url['path'], PATHINFO_BASENAME); | |
178 | ||
179 | if (empty($path)) { | |
180 | $path = $url['host']; | |
181 | } | |
182 | ||
183 | $entry->setTitle($path); | |
184 | } | |
185 | ||
d0e9b3d6 JC |
186 | /** |
187 | * Stock entry with fetched or imported content. | |
188 | * Will fall back to OpenGraph data if available. | |
189 | * | |
d5c2cc54 | 190 | * @param Entry $entry Entry to stock |
ec970721 | 191 | * @param array $content Array with at least title, url & html |
d0e9b3d6 JC |
192 | */ |
193 | private function stockEntry(Entry $entry, array $content) | |
194 | { | |
c801d03e KD |
195 | // When a redirection occurs while fetching an entry |
196 | // we move the original url in origin_url property if empty | |
197 | // and set the entry url with the final value | |
198 | if (!empty($content['url']) && $entry->getUrl() !== $content['url']) { | |
199 | if (empty($entry->getOriginUrl())) { | |
200 | $entry->setOriginUrl($entry->getUrl()); | |
201 | } | |
202 | $entry->setUrl($content['url']); | |
203 | } | |
a05b6115 | 204 | |
af29e1bf | 205 | $this->setEntryDomainName($entry); |
a05b6115 JB |
206 | |
207 | if (!empty($content['title'])) { | |
208 | $entry->setTitle($content['title']); | |
209 | } elseif (!empty($content['open_graph']['og_title'])) { | |
210 | $entry->setTitle($content['open_graph']['og_title']); | |
558d9aab JB |
211 | } |
212 | ||
213 | $html = $content['html']; | |
214 | if (false === $html) { | |
36e6ef52 | 215 | $html = $this->fetchingErrorMessage; |
558d9aab | 216 | |
e668a812 | 217 | if (!empty($content['open_graph']['og_description'])) { |
558d9aab JB |
218 | $html .= '<p><i>But we found a short description: </i></p>'; |
219 | $html .= $content['open_graph']['og_description']; | |
220 | } | |
221 | } | |
222 | ||
48656e0e | 223 | $entry->setContent($html); |
a05b6115 | 224 | $entry->setReadingTime(Utils::getReadingTime($html)); |
f0378b4d | 225 | |
a05b6115 JB |
226 | if (!empty($content['status'])) { |
227 | $entry->setHttpStatus($content['status']); | |
5e9009ce NL |
228 | } |
229 | ||
645291e8 | 230 | if (!empty($content['authors']) && is_array($content['authors'])) { |
7b0b3622 NL |
231 | $entry->setPublishedBy($content['authors']); |
232 | } | |
233 | ||
8a219854 | 234 | if (!empty($content['all_headers']) && $this->storeArticleHeaders) { |
dda6a6ad NL |
235 | $entry->setHeaders($content['all_headers']); |
236 | } | |
237 | ||
a05b6115 JB |
238 | if (!empty($content['date'])) { |
239 | $this->updatePublishedAt($entry, $content['date']); | |
240 | } | |
0d349ea6 | 241 | |
a05b6115 JB |
242 | if (!empty($content['language'])) { |
243 | $this->updateLanguage($entry, $content['language']); | |
244 | } | |
245 | ||
246 | if (!empty($content['open_graph']['og_image'])) { | |
247 | $this->updatePreviewPicture($entry, $content['open_graph']['og_image']); | |
248 | } | |
0d349ea6 | 249 | |
be54dfe4 | 250 | // if content is an image, define it as a preview too |
0d349ea6 | 251 | if (!empty($content['content_type']) && in_array($this->mimeGuesser->guess($content['content_type']), ['jpeg', 'jpg', 'gif', 'png'], true)) { |
d0ec2ddd | 252 | $this->updatePreviewPicture($entry, $content['url']); |
0d349ea6 JB |
253 | } |
254 | ||
a05b6115 JB |
255 | if (!empty($content['content_type'])) { |
256 | $entry->setMimetype($content['content_type']); | |
4d0ec0e7 | 257 | } |
558d9aab | 258 | |
1c9cd2a7 KG |
259 | try { |
260 | $this->tagger->tag($entry); | |
261 | } catch (\Exception $e) { | |
4094ea47 | 262 | $this->logger->error('Error while trying to automatically tag an entry.', [ |
d0e9b3d6 | 263 | 'entry_url' => $content['url'], |
1c9cd2a7 | 264 | 'error_msg' => $e->getMessage(), |
4094ea47 | 265 | ]); |
1c9cd2a7 | 266 | } |
558d9aab | 267 | } |
c2656f96 | 268 | |
4d0ec0e7 | 269 | /** |
d0e9b3d6 | 270 | * Validate that the given content has at least a title, an html and a url. |
4d0ec0e7 JB |
271 | * |
272 | * @param array $content | |
6acadf8e JB |
273 | * |
274 | * @return bool true if valid otherwise false | |
4d0ec0e7 JB |
275 | */ |
276 | private function validateContent(array $content) | |
277 | { | |
6acadf8e | 278 | return !empty($content['title']) && !empty($content['html']) && !empty($content['url']); |
4d0ec0e7 | 279 | } |
558d9aab | 280 | } |