]>
Commit | Line | Data |
---|---|---|
1 | <?php | |
2 | /** | |
3 | * poche, a read it later open source system | |
4 | * | |
5 | * @category poche | |
6 | * @author Nicolas Lœuillet <support@inthepoche.com> | |
7 | * @copyright 2013 | |
8 | * @license http://www.wtfpl.net/ see COPYING file | |
9 | */ | |
10 | ||
11 | /** | |
12 | * Permet de générer l'URL de poche pour le bookmarklet | |
13 | */ | |
14 | function get_poche_url() | |
15 | { | |
16 | $protocol = "http"; | |
17 | if(isset($_SERVER['HTTPS'])) { | |
18 | if($_SERVER['HTTPS'] != "off" && $_SERVER['HTTPS'] != "") { | |
19 | $protocol = "https"; | |
20 | } | |
21 | } | |
22 | ||
23 | return $protocol . "://" . $_SERVER['HTTP_HOST'] . $_SERVER['REQUEST_URI']; | |
24 | } | |
25 | ||
26 | // function define to retrieve url content | |
27 | function get_external_file($url) | |
28 | { | |
29 | $timeout = 15; | |
30 | // spoofing FireFox 18.0 | |
31 | $useragent="Mozilla/5.0 (Windows NT 5.1; rv:18.0) Gecko/20100101 Firefox/18.0"; | |
32 | ||
33 | if (in_array ('curl', get_loaded_extensions())) { | |
34 | // Fetch feed from URL | |
35 | $curl = curl_init(); | |
36 | curl_setopt($curl, CURLOPT_URL, $url); | |
37 | curl_setopt($curl, CURLOPT_TIMEOUT, $timeout); | |
38 | curl_setopt($curl, CURLOPT_FOLLOWLOCATION, true); | |
39 | curl_setopt($curl, CURLOPT_RETURNTRANSFER, true); | |
40 | curl_setopt($curl, CURLOPT_HEADER, false); | |
41 | ||
42 | // FeedBurner requires a proper USER-AGENT... | |
43 | curl_setopt($curl, CURL_HTTP_VERSION_1_1, true); | |
44 | curl_setopt($curl, CURLOPT_ENCODING, "gzip, deflate"); | |
45 | curl_setopt($curl, CURLOPT_USERAGENT, $useragent); | |
46 | ||
47 | $data = curl_exec($curl); | |
48 | ||
49 | $httpcode = curl_getinfo($curl, CURLINFO_HTTP_CODE); | |
50 | ||
51 | $httpcodeOK = isset($httpcode) and ($httpcode == 200 or $httpcode == 301); | |
52 | ||
53 | curl_close($curl); | |
54 | } else { | |
55 | ||
56 | // create http context and add timeout and user-agent | |
57 | $context = stream_context_create(array('http'=>array('timeout' => $timeout,'header'=> "User-Agent: ".$useragent,/*spoot Mozilla Firefox*/'follow_location' => true))); | |
58 | ||
59 | // only download page lesser than 4MB | |
60 | $data = @file_get_contents($url, false, $context, -1, 4000000); // We download at most 4 MB from source. | |
61 | ||
62 | if(isset($http_response_header) and isset($http_response_header[0])) { | |
63 | $httpcodeOK = isset($http_response_header) and isset($http_response_header[0]) and ((strpos($http_response_header[0], '200 OK') !== FALSE) or (strpos($http_response_header[0], '301 Moved Permanently') !== FALSE)); | |
64 | } | |
65 | } | |
66 | ||
67 | // if response is not empty and response is OK | |
68 | if (isset($data) and isset($httpcodeOK) and $httpcodeOK ) { | |
69 | ||
70 | // take charset of page and get it | |
71 | preg_match('#<meta .*charset=.*>#Usi', $data, $meta); | |
72 | ||
73 | // if meta tag is found | |
74 | if (!empty($meta[0])) { | |
75 | // retrieve encoding in $enc | |
76 | preg_match('#charset="?(.*)"#si', $meta[0], $enc); | |
77 | ||
78 | // if charset is found set it otherwise, set it to utf-8 | |
79 | $html_charset = (!empty($enc[1])) ? strtolower($enc[1]) : 'utf-8'; | |
80 | ||
81 | } else { | |
82 | $html_charset = 'utf-8'; | |
83 | $enc[1] = ''; | |
84 | } | |
85 | ||
86 | // replace charset of url to charset of page | |
87 | $data = str_replace('charset='.$enc[1], 'charset='.$html_charset, $data); | |
88 | ||
89 | return $data; | |
90 | } | |
91 | else { | |
92 | return FALSE; | |
93 | } | |
94 | } | |
95 | ||
96 | /** | |
97 | * Préparation de l'URL avec récupération du contenu avant insertion en base | |
98 | */ | |
99 | function prepare_url($url) | |
100 | { | |
101 | $parametres = array(); | |
102 | $url = html_entity_decode(trim($url)); | |
103 | ||
104 | // We remove the annoying parameters added by FeedBurner and GoogleFeedProxy (?utm_source=...) | |
105 | // from shaarli, by sebsauvage | |
106 | $i=strpos($url,'&utm_source='); if ($i!==false) $url=substr($url,0,$i); | |
107 | $i=strpos($url,'?utm_source='); if ($i!==false) $url=substr($url,0,$i); | |
108 | $i=strpos($url,'#xtor=RSS-'); if ($i!==false) $url=substr($url,0,$i); | |
109 | ||
110 | $title = $url; | |
111 | if (!preg_match('!^https?://!i', $url)) | |
112 | $url = 'http://' . $url; | |
113 | ||
114 | $html = Encoding::toUTF8(get_external_file($url,15)); | |
115 | if (isset($html) and strlen($html) > 0) | |
116 | { | |
117 | $r = new Readability($html, $url); | |
118 | $r->convertLinksToFootnotes = CONVERT_LINKS_FOOTNOTES; | |
119 | if($r->init()) | |
120 | { | |
121 | $content = $r->articleContent->innerHTML; | |
122 | $parametres['title'] = $r->articleTitle->innerHTML; | |
123 | $parametres['content'] = $content; | |
124 | return $parametres; | |
125 | } | |
126 | } | |
127 | ||
128 | $msg->add('e', 'error during url preparation'); | |
129 | logm('error during url preparation'); | |
130 | return FALSE; | |
131 | } | |
132 | ||
133 | /** | |
134 | * On modifie les URLS des images dans le corps de l'article | |
135 | */ | |
136 | function filtre_picture($content, $url, $id) | |
137 | { | |
138 | $matches = array(); | |
139 | preg_match_all('#<\s*(img)[^>]+src="([^"]*)"[^>]*>#Si', $content, $matches, PREG_SET_ORDER); | |
140 | foreach($matches as $i => $link) | |
141 | { | |
142 | $link[1] = trim($link[1]); | |
143 | if (!preg_match('#^(([a-z]+://)|(\#))#', $link[1]) ) | |
144 | { | |
145 | $absolute_path = get_absolute_link($link[2],$url); | |
146 | $filename = basename(parse_url($absolute_path, PHP_URL_PATH)); | |
147 | $directory = create_assets_directory($id); | |
148 | $fullpath = $directory . '/' . $filename; | |
149 | download_pictures($absolute_path, $fullpath); | |
150 | $content = str_replace($matches[$i][2], $fullpath, $content); | |
151 | } | |
152 | ||
153 | } | |
154 | ||
155 | return $content; | |
156 | } | |
157 | ||
158 | /** | |
159 | * Retourne le lien absolu | |
160 | */ | |
161 | function get_absolute_link($relative_link, $url) | |
162 | { | |
163 | /* return if already absolute URL */ | |
164 | if (parse_url($relative_link, PHP_URL_SCHEME) != '') return $relative_link; | |
165 | ||
166 | /* queries and anchors */ | |
167 | if ($relative_link[0]=='#' || $relative_link[0]=='?') return $url . $relative_link; | |
168 | ||
169 | /* parse base URL and convert to local variables: | |
170 | $scheme, $host, $path */ | |
171 | extract(parse_url($url)); | |
172 | ||
173 | /* remove non-directory element from path */ | |
174 | $path = preg_replace('#/[^/]*$#', '', $path); | |
175 | ||
176 | /* destroy path if relative url points to root */ | |
177 | if ($relative_link[0] == '/') $path = ''; | |
178 | ||
179 | /* dirty absolute URL */ | |
180 | $abs = $host . $path . '/' . $relative_link; | |
181 | ||
182 | /* replace '//' or '/./' or '/foo/../' with '/' */ | |
183 | $re = array('#(/\.?/)#', '#/(?!\.\.)[^/]+/\.\./#'); | |
184 | for($n=1; $n>0; $abs=preg_replace($re, '/', $abs, -1, $n)) {} | |
185 | ||
186 | /* absolute URL is ready! */ | |
187 | return $scheme.'://'.$abs; | |
188 | } | |
189 | ||
190 | /** | |
191 | * Téléchargement des images | |
192 | */ | |
193 | ||
194 | function download_pictures($absolute_path, $fullpath) | |
195 | { | |
196 | $rawdata = get_external_file($absolute_path); | |
197 | ||
198 | if(file_exists($fullpath)) { | |
199 | unlink($fullpath); | |
200 | } | |
201 | $fp = fopen($fullpath, 'x'); | |
202 | fwrite($fp, $rawdata); | |
203 | fclose($fp); | |
204 | } | |
205 | ||
206 | /** | |
207 | * Crée un répertoire de médias pour l'article | |
208 | */ | |
209 | function create_assets_directory($id) | |
210 | { | |
211 | $assets_path = ABS_PATH; | |
212 | if(!is_dir($assets_path)) { | |
213 | mkdir($assets_path, 0705); | |
214 | } | |
215 | ||
216 | $article_directory = $assets_path . $id; | |
217 | if(!is_dir($article_directory)) { | |
218 | mkdir($article_directory, 0705); | |
219 | } | |
220 | ||
221 | return $article_directory; | |
222 | } | |
223 | ||
224 | /** | |
225 | * Suppression du répertoire d'images | |
226 | */ | |
227 | function remove_directory($directory) | |
228 | { | |
229 | if(is_dir($directory)) { | |
230 | $files = array_diff(scandir($directory), array('.','..')); | |
231 | foreach ($files as $file) { | |
232 | (is_dir("$directory/$file")) ? remove_directory("$directory/$file") : unlink("$directory/$file"); | |
233 | } | |
234 | return rmdir($directory); | |
235 | } | |
236 | } | |
237 | ||
238 | function display_view($view, $id = 0, $full_head = 'yes') | |
239 | { | |
240 | global $tpl, $store, $msg; | |
241 | ||
242 | switch ($view) | |
243 | { | |
244 | case 'export': | |
245 | $entries = $store->retrieveAll(); | |
246 | $tpl->assign('export', myTool::renderJson($entries)); | |
247 | $tpl->draw('export'); | |
248 | logm('export view'); | |
249 | break; | |
250 | case 'config': | |
251 | $tpl->assign('load_all_js', 0); | |
252 | $tpl->draw('head'); | |
253 | $tpl->draw('home'); | |
254 | $tpl->draw('config'); | |
255 | $tpl->draw('js'); | |
256 | $tpl->draw('footer'); | |
257 | logm('config view'); | |
258 | break; | |
259 | case 'view': | |
260 | $entry = $store->retrieveOneById($id); | |
261 | ||
262 | if ($entry != NULL) { | |
263 | $tpl->assign('id', $entry['id']); | |
264 | $tpl->assign('url', $entry['url']); | |
265 | $tpl->assign('title', $entry['title']); | |
266 | $tpl->assign('content', $entry['content']); | |
267 | $tpl->assign('is_fav', $entry['is_fav']); | |
268 | $tpl->assign('is_read', $entry['is_read']); | |
269 | $tpl->assign('load_all_js', 0); | |
270 | $tpl->draw('view'); | |
271 | } | |
272 | else { | |
273 | logm('error in view call : entry is NULL'); | |
274 | } | |
275 | ||
276 | logm('view link #' . $id); | |
277 | break; | |
278 | default: # home view | |
279 | $entries = $store->getEntriesByView($view); | |
280 | ||
281 | $tpl->assign('entries', $entries); | |
282 | ||
283 | if ($full_head == 'yes') { | |
284 | $tpl->assign('load_all_js', 1); | |
285 | $tpl->draw('head'); | |
286 | $tpl->draw('home'); | |
287 | } | |
288 | ||
289 | $tpl->draw('entries'); | |
290 | ||
291 | if ($full_head == 'yes') { | |
292 | $tpl->draw('js'); | |
293 | $tpl->draw('footer'); | |
294 | } | |
295 | break; | |
296 | } | |
297 | } | |
298 | ||
299 | /** | |
300 | * Appel d'une action (mark as fav, archive, delete) | |
301 | */ | |
302 | function action_to_do($action, $url, $id = 0) | |
303 | { | |
304 | global $store, $msg; | |
305 | ||
306 | switch ($action) | |
307 | { | |
308 | case 'add': | |
309 | if ($url == '') | |
310 | continue; | |
311 | ||
312 | if (MyTool::isUrl($url)) { | |
313 | if($parametres_url = prepare_url($url)) { | |
314 | $store->add($url, $parametres_url['title'], $parametres_url['content']); | |
315 | $last_id = $store->getLastId(); | |
316 | if (DOWNLOAD_PICTURES) { | |
317 | $content = filtre_picture($parametres_url['content'], $url, $last_id); | |
318 | } | |
319 | $msg->add('s', 'the link has been added successfully'); | |
320 | } | |
321 | } | |
322 | else { | |
323 | $msg->add('e', 'the link has been added successfully'); | |
324 | logm($url . ' is not a valid url'); | |
325 | } | |
326 | ||
327 | logm('add link ' . $url); | |
328 | break; | |
329 | case 'delete': | |
330 | remove_directory(ABS_PATH . $id); | |
331 | $store->deleteById($id); | |
332 | $msg->add('s', 'the link has been deleted successfully'); | |
333 | logm('delete link #' . $id); | |
334 | break; | |
335 | case 'toggle_fav' : | |
336 | $store->favoriteById($id); | |
337 | $msg->add('s', 'the favorite toggle has been done successfully'); | |
338 | logm('mark as favorite link #' . $id); | |
339 | break; | |
340 | case 'toggle_archive' : | |
341 | $store->archiveById($id); | |
342 | $msg->add('s', 'the archive toggle has been done successfully'); | |
343 | logm('archive link #' . $id); | |
344 | break; | |
345 | default: | |
346 | break; | |
347 | } | |
348 | } | |
349 | ||
350 | function logm($message) | |
351 | { | |
352 | $t = strval(date('Y/m/d_H:i:s')).' - '.$_SERVER["REMOTE_ADDR"].' - '.strval($message)."\n"; | |
353 | file_put_contents('./log.txt',$t,FILE_APPEND); | |
354 | } |