Friendica Communications Platform (please note that this is a clone of the repository at github, issues are handled there) https://friendi.ca
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

1992 lines
69 KiB

3 years ago
3 years ago
  1. <?php
  2. /**
  3. * @file src/Content/Text/BBCode.php
  4. */
  5. namespace Friendica\Content\Text;
  6. use DOMDocument;
  7. use DOMXPath;
  8. use Exception;
  9. use Friendica\BaseObject;
  10. use Friendica\Content\OEmbed;
  11. use Friendica\Content\Smilies;
  12. use Friendica\Core\Addon;
  13. use Friendica\Core\Cache;
  14. use Friendica\Core\Config;
  15. use Friendica\Core\L10n;
  16. use Friendica\Core\Protocol;
  17. use Friendica\Core\System;
  18. use Friendica\Model\Contact;
  19. use Friendica\Model\Event;
  20. use Friendica\Network\Probe;
  21. use Friendica\Object\Image;
  22. use Friendica\Util\Map;
  23. use Friendica\Util\Network;
  24. use Friendica\Util\ParseUrl;
  25. use Friendica\Util\Proxy as ProxyUtils;
  26. use League\HTMLToMarkdown\HtmlConverter;
  27. class BBCode extends BaseObject
  28. {
  29. /**
  30. * @brief Fetches attachment data that were generated the old way
  31. *
  32. * @param string $body Message body
  33. * @return array
  34. * 'type' -> Message type ("link", "video", "photo")
  35. * 'text' -> Text before the shared message
  36. * 'after' -> Text after the shared message
  37. * 'image' -> Preview image of the message
  38. * 'url' -> Url to the attached message
  39. * 'title' -> Title of the attachment
  40. * 'description' -> Description of the attachment
  41. */
  42. private static function getOldAttachmentData($body)
  43. {
  44. $post = [];
  45. // Simplify image codes
  46. $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
  47. if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
  48. foreach ($attached as $data) {
  49. if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
  50. continue;
  51. }
  52. $post["type"] = substr($data[1], 5);
  53. $pos = strpos($body, $data[0]);
  54. if ($pos > 0) {
  55. $post["text"] = trim(substr($body, 0, $pos));
  56. $post["after"] = trim(substr($body, $pos + strlen($data[0])));
  57. } else {
  58. $post["text"] = trim(str_replace($data[0], "", $body));
  59. }
  60. $attacheddata = $data[2];
  61. $URLSearchString = "^\[\]";
  62. if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
  63. $picturedata = Image::getInfoFromURL($matches[1]);
  64. if ($picturedata) {
  65. if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
  66. $post["image"] = $matches[1];
  67. } else {
  68. $post["preview"] = $matches[1];
  69. }
  70. }
  71. }
  72. if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
  73. $post["url"] = $matches[1];
  74. $post["title"] = $matches[2];
  75. }
  76. if (!empty($post["url"]) && (in_array($post["type"], ["link", "video"]))
  77. && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
  78. $post["url"] = $matches[1];
  79. }
  80. // Search for description
  81. if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
  82. $post["description"] = $matches[1];
  83. }
  84. }
  85. }
  86. return $post;
  87. }
  88. /**
  89. * @brief Fetches attachment data that were generated with the "attachment" element
  90. *
  91. * @param string $body Message body
  92. * @return array
  93. * 'type' -> Message type ("link", "video", "photo")
  94. * 'text' -> Text before the shared message
  95. * 'after' -> Text after the shared message
  96. * 'image' -> Preview image of the message
  97. * 'url' -> Url to the attached message
  98. * 'title' -> Title of the attachment
  99. * 'description' -> Description of the attachment
  100. */
  101. public static function getAttachmentData($body)
  102. {
  103. $data = [];
  104. if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
  105. return self::getOldAttachmentData($body);
  106. }
  107. $attributes = $match[2];
  108. $data["text"] = trim($match[1]);
  109. $type = "";
  110. preg_match("/type='(.*?)'/ism", $attributes, $matches);
  111. if (x($matches, 1)) {
  112. $type = strtolower($matches[1]);
  113. }
  114. preg_match('/type="(.*?)"/ism', $attributes, $matches);
  115. if (x($matches, 1)) {
  116. $type = strtolower($matches[1]);
  117. }
  118. if ($type == "") {
  119. return [];
  120. }
  121. if (!in_array($type, ["link", "audio", "photo", "video"])) {
  122. return [];
  123. }
  124. if ($type != "") {
  125. $data["type"] = $type;
  126. }
  127. $url = "";
  128. preg_match("/url='(.*?)'/ism", $attributes, $matches);
  129. if (x($matches, 1)) {
  130. $url = $matches[1];
  131. }
  132. preg_match('/url="(.*?)"/ism', $attributes, $matches);
  133. if (x($matches, 1)) {
  134. $url = $matches[1];
  135. }
  136. if ($url != "") {
  137. $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
  138. }
  139. $title = "";
  140. preg_match("/title='(.*?)'/ism", $attributes, $matches);
  141. if (x($matches, 1)) {
  142. $title = $matches[1];
  143. }
  144. preg_match('/title="(.*?)"/ism', $attributes, $matches);
  145. if (x($matches, 1)) {
  146. $title = $matches[1];
  147. }
  148. if ($title != "") {
  149. $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
  150. $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
  151. $title = str_replace(["[", "]"], ["&#91;", "&#93;"], $title);
  152. $data["title"] = $title;
  153. }
  154. $image = "";
  155. preg_match("/image='(.*?)'/ism", $attributes, $matches);
  156. if (x($matches, 1)) {
  157. $image = $matches[1];
  158. }
  159. preg_match('/image="(.*?)"/ism', $attributes, $matches);
  160. if (x($matches, 1)) {
  161. $image = $matches[1];
  162. }
  163. if ($image != "") {
  164. $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
  165. }
  166. $preview = "";
  167. preg_match("/preview='(.*?)'/ism", $attributes, $matches);
  168. if (x($matches, 1)) {
  169. $preview = $matches[1];
  170. }
  171. preg_match('/preview="(.*?)"/ism', $attributes, $matches);
  172. if (x($matches, 1)) {
  173. $preview = $matches[1];
  174. }
  175. if ($preview != "") {
  176. $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
  177. }
  178. $data["description"] = trim($match[3]);
  179. $data["after"] = trim($match[4]);
  180. return $data;
  181. }
  182. public static function getAttachedData($body, $item = [])
  183. {
  184. /*
  185. - text:
  186. - type: link, video, photo
  187. - title:
  188. - url:
  189. - image:
  190. - description:
  191. - (thumbnail)
  192. */
  193. $has_title = !empty($item['title']);
  194. $plink = (!empty($item['plink']) ? $item['plink'] : '');
  195. $post = self::getAttachmentData($body);
  196. // if nothing is found, it maybe having an image.
  197. if (!isset($post["type"])) {
  198. // Simplify image codes
  199. $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
  200. $URLSearchString = "^\[\]";
  201. $body = preg_replace("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
  202. if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
  203. if ((count($pictures) == 1) && !$has_title) {
  204. // Checking, if the link goes to a picture
  205. $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
  206. // Workaround:
  207. // Sometimes photo posts to the own album are not detected at the start.
  208. // So we seem to cannot use the cache for these cases. That's strange.
  209. if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
  210. $data = ParseUrl::getSiteinfo($pictures[0][1], true);
  211. }
  212. if ($data["type"] == "photo") {
  213. $post["type"] = "photo";
  214. if (isset($data["images"][0])) {
  215. $post["image"] = $data["images"][0]["src"];
  216. $post["url"] = $data["url"];
  217. } else {
  218. $post["image"] = $data["url"];
  219. }
  220. $post["preview"] = $pictures[0][2];
  221. $post["text"] = str_replace($pictures[0][0], "", $body);
  222. } else {
  223. $imgdata = Image::getInfoFromURL($pictures[0][1]);
  224. if ($imgdata && substr($imgdata["mime"], 0, 6) == "image/") {
  225. $post["type"] = "photo";
  226. $post["image"] = $pictures[0][1];
  227. $post["preview"] = $pictures[0][2];
  228. $post["text"] = str_replace($pictures[0][0], "", $body);
  229. }
  230. }
  231. } elseif (count($pictures) > 0) {
  232. $post["type"] = "link";
  233. $post["url"] = $plink;
  234. $post["image"] = $pictures[0][2];
  235. $post["text"] = $body;
  236. }
  237. } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
  238. if ((count($pictures) == 1) && !$has_title) {
  239. $post["type"] = "photo";
  240. $post["image"] = $pictures[0][1];
  241. $post["text"] = str_replace($pictures[0][0], "", $body);
  242. } elseif (count($pictures) > 0) {
  243. $post["type"] = "link";
  244. $post["url"] = $plink;
  245. $post["image"] = $pictures[0][1];
  246. $post["text"] = $body;
  247. }
  248. }
  249. // Test for the external links
  250. preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
  251. preg_match_all("(\[url\=([$URLSearchString]*)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
  252. $links = array_merge($links1, $links2);
  253. // If there is only a single one, then use it.
  254. // This should cover link posts via API.
  255. if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
  256. $post["type"] = "link";
  257. $post["text"] = trim($body);
  258. $post["url"] = $links[0][1];
  259. }
  260. // Now count the number of external media links
  261. preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
  262. preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
  263. preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
  264. preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
  265. // Add them to the other external links
  266. $links = array_merge($links, $links1, $links2, $links3, $links4);
  267. // Are there more than one?
  268. if (count($links) > 1) {
  269. // The post will be the type "text", which means a blog post
  270. unset($post["type"]);
  271. $post["url"] = $plink;
  272. }
  273. if (!isset($post["type"])) {
  274. $post["type"] = "text";
  275. $post["text"] = trim($body);
  276. }
  277. } elseif (isset($post["url"]) && ($post["type"] == "video")) {
  278. $data = ParseUrl::getSiteinfoCached($post["url"], true);
  279. if (isset($data["images"][0])) {
  280. $post["image"] = $data["images"][0]["src"];
  281. }
  282. }
  283. return $post;
  284. }
  285. /**
  286. * @brief Converts a BBCode text into plaintext
  287. *
  288. * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
  289. *
  290. * @return string
  291. */
  292. public static function toPlaintext($text, $keep_urls = true)
  293. {
  294. $naked_text = preg_replace('/\[(.+?)\]/','', $text);
  295. if (!$keep_urls) {
  296. $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text);
  297. }
  298. return $naked_text;
  299. }
  300. private static function proxyUrl($image, $simplehtml = false)
  301. {
  302. // Only send proxied pictures to API and for internal display
  303. if (in_array($simplehtml, [false, 2])) {
  304. return ProxyUtils::proxifyUrl($image);
  305. } else {
  306. return $image;
  307. }
  308. }
  309. public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
  310. {
  311. // Suppress "view full size"
  312. if (intval(Config::get('system', 'no_view_full_size'))) {
  313. $include_link = false;
  314. }
  315. // Picture addresses can contain special characters
  316. $s = htmlspecialchars_decode($srctext);
  317. $matches = null;
  318. $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
  319. if ($c) {
  320. foreach ($matches as $mtch) {
  321. logger('scale_external_image: ' . $mtch[1]);
  322. $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
  323. if (stristr($mtch[1], $hostname)) {
  324. continue;
  325. }
  326. // $scale_replace, if passed, is an array of two elements. The
  327. // first is the name of the full-size image. The second is the
  328. // name of a remote, scaled-down version of the full size image.
  329. // This allows Friendica to display the smaller remote image if
  330. // one exists, while still linking to the full-size image
  331. if ($scale_replace) {
  332. $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
  333. } else {
  334. $scaled = $mtch[1];
  335. }
  336. $i = Network::fetchUrl($scaled);
  337. if (!$i) {
  338. return $srctext;
  339. }
  340. // guess mimetype from headers or filename
  341. $type = Image::guessType($mtch[1], true);
  342. if ($i) {
  343. $Image = new Image($i, $type);
  344. if ($Image->isValid()) {
  345. $orig_width = $Image->getWidth();
  346. $orig_height = $Image->getHeight();
  347. if ($orig_width > 640 || $orig_height > 640) {
  348. $Image->scaleDown(640);
  349. $new_width = $Image->getWidth();
  350. $new_height = $Image->getHeight();
  351. logger('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], LOGGER_DEBUG);
  352. $s = str_replace(
  353. $mtch[0],
  354. '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
  355. . "\n" . (($include_link)
  356. ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
  357. : ''),
  358. $s
  359. );
  360. logger('scale_external_images: new string: ' . $s, LOGGER_DEBUG);
  361. }
  362. }
  363. }
  364. }
  365. }
  366. // replace the special char encoding
  367. $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
  368. return $s;
  369. }
  370. /**
  371. * The purpose of this function is to apply system message length limits to
  372. * imported messages without including any embedded photos in the length
  373. *
  374. * @brief Truncates imported message body string length to max_import_size
  375. * @param string $body
  376. * @return string
  377. */
  378. public static function limitBodySize($body)
  379. {
  380. $maxlen = Config::get('config', 'max_import_size', 0);
  381. // If the length of the body, including the embedded images, is smaller
  382. // than the maximum, then don't waste time looking for the images
  383. if ($maxlen && (strlen($body) > $maxlen)) {
  384. logger('the total body length exceeds the limit', LOGGER_DEBUG);
  385. $orig_body = $body;
  386. $new_body = '';
  387. $textlen = 0;
  388. $img_start = strpos($orig_body, '[img');
  389. $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
  390. $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
  391. while (($img_st_close !== false) && ($img_end !== false)) {
  392. $img_st_close++; // make it point to AFTER the closing bracket
  393. $img_end += $img_start;
  394. $img_end += strlen('[/img]');
  395. if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
  396. // This is an embedded image
  397. if (($textlen + $img_start) > $maxlen) {
  398. if ($textlen < $maxlen) {
  399. logger('the limit happens before an embedded image', LOGGER_DEBUG);
  400. $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
  401. $textlen = $maxlen;
  402. }
  403. } else {
  404. $new_body = $new_body . substr($orig_body, 0, $img_start);
  405. $textlen += $img_start;
  406. }
  407. $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
  408. } else {
  409. if (($textlen + $img_end) > $maxlen) {
  410. if ($textlen < $maxlen) {
  411. logger('the limit happens before the end of a non-embedded image', LOGGER_DEBUG);
  412. $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
  413. $textlen = $maxlen;
  414. }
  415. } else {
  416. $new_body = $new_body . substr($orig_body, 0, $img_end);
  417. $textlen += $img_end;
  418. }
  419. }
  420. $orig_body = substr($orig_body, $img_end);
  421. if ($orig_body === false) {
  422. // in case the body ends on a closing image tag
  423. $orig_body = '';
  424. }
  425. $img_start = strpos($orig_body, '[img');
  426. $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
  427. $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
  428. }
  429. if (($textlen + strlen($orig_body)) > $maxlen) {
  430. if ($textlen < $maxlen) {
  431. logger('the limit happens after the end of the last image', LOGGER_DEBUG);
  432. $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
  433. }
  434. } else {
  435. logger('the text size with embedded images extracted did not violate the limit', LOGGER_DEBUG);
  436. $new_body = $new_body . $orig_body;
  437. }
  438. return $new_body;
  439. } else {
  440. return $body;
  441. }
  442. }
  443. /**
  444. * Processes [attachment] tags
  445. *
  446. * Note: Can produce a [bookmark] tag in the returned string
  447. *
  448. * @brief Processes [attachment] tags
  449. * @param string $return
  450. * @param bool|int $simplehtml
  451. * @param bool $tryoembed
  452. * @return string
  453. */
  454. private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
  455. {
  456. $data = self::getAttachmentData($return);
  457. if (empty($data) || empty($data["url"])) {
  458. return $return;
  459. }
  460. if (isset($data["title"])) {
  461. $data["title"] = strip_tags($data["title"]);
  462. $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
  463. } else {
  464. $data["title"] = null;
  465. }
  466. if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data["image"])) {
  467. $data["preview"] = $data["image"];
  468. $data["image"] = "";
  469. }
  470. $return = '';
  471. if ($simplehtml == 7) {
  472. $return = self::convertUrlForOStatus($data["url"]);
  473. } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
  474. $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data["url"], $data["title"]);
  475. } else {
  476. try {
  477. if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
  478. $return = OEmbed::getHTML($data['url'], $data['title']);
  479. } else {
  480. throw new Exception('OEmbed is disabled for this attachment.');
  481. }
  482. } catch (Exception $e) {
  483. $data["title"] = defaults($data, 'title', $data['url']);
  484. if ($simplehtml != 4) {
  485. $return = sprintf('<div class="type-%s">', $data["type"]);
  486. }
  487. if (!empty($data["image"])) {
  488. $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
  489. } elseif (!empty($data["preview"])) {
  490. $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], self::proxyUrl($data["preview"], $simplehtml), $data["title"]);
  491. }
  492. if (($data["type"] == "photo") && !empty($data["url"]) && !empty($data["image"])) {
  493. $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
  494. } else {
  495. $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
  496. }
  497. if (!empty($data["description"]) && $data["description"] != $data["title"]) {
  498. // Sanitize the HTML by converting it to BBCode
  499. $bbcode = HTML::toBBCode($data["description"]);
  500. $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
  501. }
  502. if ($data["type"] == "link") {
  503. $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
  504. }
  505. if ($simplehtml != 4) {
  506. $return .= '</div>';
  507. }
  508. }
  509. }
  510. return trim(defaults($data, 'text', '') . ' ' . $return . ' ' . defaults($data, 'after', ''));
  511. }
  512. public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
  513. {
  514. $data = self::getAttachmentData($Text);
  515. if (!$data) {
  516. return $Text;
  517. } elseif ($nolink) {
  518. return $data["text"] . defaults($data, 'after', '');
  519. }
  520. $title = htmlentities(defaults($data, 'title', ''), ENT_QUOTES, 'UTF-8', false);
  521. $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
  522. if ($plaintext || (($title != "") && strstr($text, $title))) {
  523. $data["title"] = $data["url"];
  524. } elseif (($text != "") && strstr($title, $text)) {
  525. $data["text"] = $data["title"];
  526. $data["title"] = $data["url"];
  527. }
  528. if (($data["text"] == "") && ($data["title"] != "") && ($data["url"] == "")) {
  529. return $data["title"] . $data["after"];
  530. }
  531. // If the link already is included in the post, don't add it again
  532. if (!empty($data["url"]) && strpos($data["text"], $data["url"])) {
  533. return $data["text"] . $data["after"];
  534. }
  535. $text = $data["text"];
  536. if (!empty($data["url"]) && !empty($data["title"])) {
  537. $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
  538. } elseif (!empty($data["url"])) {
  539. $text .= "\n[url]" . $data["url"] . "[/url]";
  540. }
  541. return $text . "\n" . $data["after"];
  542. }
  543. /**
  544. * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
  545. *
  546. * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
  547. * @param array $match Array with the matching values
  548. * @return string reformatted link including HTML codes
  549. */
  550. private static function convertUrlForOStatusCallback($match)
  551. {
  552. $url = $match[1];
  553. if (isset($match[2]) && ($match[1] != $match[2])) {
  554. return $match[0];
  555. }
  556. $parts = parse_url($url);
  557. if (!isset($parts['scheme'])) {
  558. return $match[0];
  559. }
  560. return self::convertUrlForOStatus($url);
  561. }
  562. /**
  563. * @brief Converts [url] BBCodes in a format that looks fine on OStatus systems.
  564. * @param string $url URL that is about to be reformatted
  565. * @return string reformatted link including HTML codes
  566. */
  567. private static function convertUrlForOStatus($url)
  568. {
  569. $parts = parse_url($url);
  570. $scheme = $parts['scheme'] . '://';
  571. $styled_url = str_replace($scheme, '', $url);
  572. if (strlen($styled_url) > 30) {
  573. $styled_url = substr($styled_url, 0, 30) . "";
  574. }
  575. $html = '<a href="%s" target="_blank">%s</a>';
  576. return sprintf($html, $url, $styled_url);
  577. }
  578. /*
  579. * [noparse][i]italic[/i][/noparse] turns into
  580. * [noparse][ i ]italic[ /i ][/noparse],
  581. * to hide them from parser.
  582. */
  583. private static function escapeNoparseCallback($match)
  584. {
  585. $whole_match = $match[0];
  586. $captured = $match[1];
  587. $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
  588. $new_str = str_replace($captured, $spacefied, $whole_match);
  589. return $new_str;
  590. }
  591. /*
  592. * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
  593. * now turns back and the [noparse] tags are trimed
  594. * returning [i]italic[/i]
  595. */
  596. private static function unescapeNoparseCallback($match)
  597. {
  598. $captured = $match[1];
  599. $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
  600. return $unspacefied;
  601. }
  602. /**
  603. * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
  604. * occurrences
  605. *
  606. * @param string $text Text to search
  607. * @param string $name Tag name
  608. * @param int $occurrences Number of first occurrences to skip
  609. * @return boolean|array
  610. */
  611. public static function getTagPosition($text, $name, $occurrences = 0)
  612. {
  613. if ($occurrences < 0) {
  614. $occurrences = 0;
  615. }
  616. $start_open = -1;
  617. for ($i = 0; $i <= $occurrences; $i++) {
  618. if ($start_open !== false) {
  619. $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
  620. }
  621. }
  622. if ($start_open === false) {
  623. return false;
  624. }
  625. $start_equal = strpos($text, '=', $start_open);
  626. $start_close = strpos($text, ']', $start_open);
  627. if ($start_close === false) {
  628. return false;
  629. }
  630. $start_close++;
  631. $end_open = strpos($text, '[/' . $name . ']', $start_close);
  632. if ($end_open === false) {
  633. return false;
  634. }
  635. $res = [
  636. 'start' => [
  637. 'open' => $start_open,
  638. 'close' => $start_close
  639. ],
  640. 'end' => [
  641. 'open' => $end_open,
  642. 'close' => $end_open + strlen('[/' . $name . ']')
  643. ],
  644. ];
  645. if ($start_equal !== false) {
  646. $res['start']['equal'] = $start_equal + 1;
  647. }
  648. return $res;
  649. }
  650. /**
  651. * Performs a preg_replace within the boundaries of all named BBCode tags in a text
  652. *
  653. * @param type $pattern Preg pattern string
  654. * @param type $replace Preg replace string
  655. * @param type $name BBCode tag name
  656. * @param type $text Text to search
  657. * @return string
  658. */
  659. public static function pregReplaceInTag($pattern, $replace, $name, $text)
  660. {
  661. $occurrences = 0;
  662. $pos = self::getTagPosition($text, $name, $occurrences);
  663. while ($pos !== false && $occurrences++ < 1000) {
  664. $start = substr($text, 0, $pos['start']['open']);
  665. $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
  666. $end = substr($text, $pos['end']['close']);
  667. if ($end === false) {
  668. $end = '';
  669. }
  670. $subject = preg_replace($pattern, $replace, $subject);
  671. $text = $start . $subject . $end;
  672. $pos = self::getTagPosition($text, $name, $occurrences);
  673. }
  674. return $text;
  675. }
  676. private static function extractImagesFromItemBody($body)
  677. {
  678. $saved_image = [];
  679. $orig_body = $body;
  680. $new_body = '';
  681. $cnt = 0;
  682. $img_start = strpos($orig_body, '[img');
  683. $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
  684. $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
  685. while (($img_st_close !== false) && ($img_end !== false)) {
  686. $img_st_close++; // make it point to AFTER the closing bracket
  687. $img_end += $img_start;
  688. if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
  689. // This is an embedded image
  690. $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
  691. $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
  692. $cnt++;
  693. } else {
  694. $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
  695. }
  696. $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
  697. if ($orig_body === false) {
  698. // in case the body ends on a closing image tag
  699. $orig_body = '';
  700. }
  701. $img_start = strpos($orig_body, '[img');
  702. $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
  703. $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
  704. }
  705. $new_body = $new_body . $orig_body;
  706. return ['body' => $new_body, 'images' => $saved_image];
  707. }
  708. private static function interpolateSavedImagesIntoItemBody($body, array $images)
  709. {
  710. $newbody = $body;
  711. $cnt = 0;
  712. foreach ($images as $image) {
  713. // We're depending on the property of 'foreach' (specified on the PHP website) that
  714. // it loops over the array starting from the first element and going sequentially
  715. // to the last element
  716. $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
  717. '<img src="' . self::proxyUrl($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
  718. $cnt++;
  719. }
  720. return $newbody;
  721. }
  722. /**
  723. * Processes [share] tags
  724. *
  725. * Note: Can produce a [bookmark] tag in the output
  726. *
  727. * @brief Processes [share] tags
  728. * @param array $share preg_match_callback result array
  729. * @param bool|int $simplehtml
  730. * @return string
  731. */
  732. private static function convertShare($share, $simplehtml)
  733. {
  734. $attributes = $share[2];
  735. $author = "";
  736. preg_match("/author='(.*?)'/ism", $attributes, $matches);
  737. if (x($matches, 1)) {
  738. $author = html_entity_decode($matches[1], ENT_QUOTES, 'UTF-8');
  739. }
  740. preg_match('/author="(.*?)"/ism', $attributes, $matches);
  741. if (x($matches, 1)) {
  742. $author = $matches[1];
  743. }
  744. $profile = "";
  745. preg_match("/profile='(.*?)'/ism", $attributes, $matches);
  746. if (x($matches, 1)) {
  747. $profile = $matches[1];
  748. }
  749. preg_match('/profile="(.*?)"/ism', $attributes, $matches);
  750. if (x($matches, 1)) {
  751. $profile = $matches[1];
  752. }
  753. $avatar = "";
  754. preg_match("/avatar='(.*?)'/ism", $attributes, $matches);
  755. if (x($matches, 1)) {
  756. $avatar = $matches[1];
  757. }
  758. preg_match('/avatar="(.*?)"/ism', $attributes, $matches);
  759. if (x($matches, 1)) {
  760. $avatar = $matches[1];
  761. }
  762. $link = "";
  763. preg_match("/link='(.*?)'/ism", $attributes, $matches);
  764. if (x($matches, 1)) {
  765. $link = $matches[1];
  766. }
  767. preg_match('/link="(.*?)"/ism', $attributes, $matches);
  768. if (x($matches, 1)) {
  769. $link = $matches[1];
  770. }
  771. $posted = "";
  772. preg_match("/posted='(.*?)'/ism", $attributes, $matches);
  773. if (x($matches, 1)) {
  774. $posted = $matches[1];
  775. }
  776. preg_match('/posted="(.*?)"/ism', $attributes, $matches);
  777. if (x($matches, 1)) {
  778. $posted = $matches[1];
  779. }
  780. // We only call this so that a previously unknown contact can be added.
  781. // This is important for the function "Model\Contact::getDetailsByURL()".
  782. // This function then can fetch an entry from the contact table.
  783. Contact::getIdForURL($profile, 0, true);
  784. $data = Contact::getDetailsByURL($profile);
  785. if (x($data, "name") && x($data, "addr")) {
  786. $userid_compact = $data["name"] . " (" . $data["addr"] . ")";
  787. } else {
  788. $userid_compact = Protocol::getAddrFromProfileUrl($profile, $author);
  789. }
  790. if (x($data, "addr")) {
  791. $userid = $data["addr"];
  792. } else {
  793. $userid = Protocol::formatMention($profile, $author);
  794. }
  795. if (x($data, "name")) {
  796. $author = $data["name"];
  797. }
  798. if (x($data, "micro")) {
  799. $avatar = $data["micro"];
  800. }
  801. $preshare = trim($share[1]);
  802. if ($preshare != "") {
  803. $preshare .= "<br />";
  804. }
  805. switch ($simplehtml) {
  806. case 1:
  807. $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . ' <a href="' . $profile . '">' . $userid . "</a>: <br />»" . $share[3] . "«";
  808. break;
  809. case 2:
  810. $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
  811. break;
  812. case 3: // Diaspora
  813. $headline = '<b>' . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . $userid . ':</b><br />';
  814. $text = trim($share[1]);
  815. if ($text != "") {
  816. $text .= "<hr />";
  817. }
  818. if (stripos(normalise_link($link), 'http://twitter.com/') === 0) {
  819. $text .= '<br /><a href="' . $link . '">' . $link . '</a>';
  820. } else {
  821. $text .= $headline . '<blockquote>' . trim($share[3]) . "</blockquote><br />";
  822. if ($link != "") {
  823. $text .= '<br /><a href="' . $link . '">[l]</a>';
  824. }
  825. }
  826. break;
  827. case 4:
  828. $headline = '<br /><b>' . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8');
  829. $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $link, $userid, $posted);
  830. $headline .= ":</b><br />";
  831. $text = trim($share[1]);
  832. if ($text != "") {
  833. $text .= "<hr />";
  834. }
  835. $text .= $headline . '<blockquote class="shared_content">' . trim($share[3]) . "</blockquote><br />";
  836. break;
  837. case 5:
  838. $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
  839. break;
  840. case 7: // statusnet/GNU Social
  841. $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3];
  842. break;
  843. case 8: // twitter
  844. $text = $preshare . "RT @" . $userid_compact . ": " . $share[3];
  845. break;
  846. case 9: // Google+
  847. $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
  848. if ($link != "") {
  849. $text .= "<br /><br />" . $link;
  850. }
  851. break;
  852. default:
  853. // Transforms quoted tweets in rich attachments to avoid nested tweets
  854. if (stripos(normalise_link($link), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($link)) {
  855. try {
  856. $oembed = OEmbed::getHTML($link, $preshare);
  857. } catch (Exception $e) {
  858. $oembed = sprintf('[bookmark=%s]%s[/bookmark]', $link, $preshare);
  859. }
  860. $text = $preshare . $oembed;
  861. } else {
  862. $text = trim($share[1]) . "\n";
  863. $avatar = ProxyUtils::proxifyUrl($avatar, false, ProxyUtils::SIZE_THUMB);
  864. $tpl = get_markup_template('shared_content.tpl');
  865. $text .= replace_macros($tpl, [
  866. '$profile' => $profile,
  867. '$avatar' => $avatar,
  868. '$author' => $author,
  869. '$link' => $link,
  870. '$posted' => $posted,
  871. '$content' => trim($share[3])
  872. ]);
  873. }
  874. break;
  875. }
  876. return $text;
  877. }
  878. private static function removePictureLinksCallback($match)
  879. {
  880. $text = Cache::get($match[1]);
  881. if (is_null($text)) {
  882. $a = self::getApp();
  883. $stamp1 = microtime(true);
  884. $ch = @curl_init($match[1]);
  885. @curl_setopt($ch, CURLOPT_NOBODY, true);
  886. @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  887. @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
  888. @curl_exec($ch);
  889. $curl_info = @curl_getinfo($ch);
  890. $a->save_timestamp($stamp1, "network");
  891. if (substr($curl_info["content_type"], 0, 6) == "image/") {
  892. $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
  893. } else {
  894. $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
  895. // if its not a picture then look if its a page that contains a picture link
  896. $body = Network::fetchUrl($match[1]);
  897. $doc = new DOMDocument();
  898. @$doc->loadHTML($body);
  899. $xpath = new DOMXPath($doc);
  900. $list = $xpath->query("//meta[@name]");
  901. foreach ($list as $node) {
  902. $attr = [];
  903. if ($node->attributes->length) {
  904. foreach ($node->attributes as $attribute) {
  905. $attr[$attribute->name] = $attribute->value;
  906. }
  907. }
  908. if (strtolower($attr["name"]) == "twitter:image") {
  909. $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
  910. }
  911. }
  912. }
  913. Cache::set($match[1], $text);
  914. }
  915. return $text;
  916. }
  917. private static function expandLinksCallback($match)
  918. {
  919. if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
  920. return ($match[1] . "[url]" . $match[2] . "[/url]");
  921. } else {
  922. return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
  923. }
  924. }
  925. private static function cleanPictureLinksCallback($match)
  926. {
  927. $text = Cache::get($match[1]);
  928. if (is_null($text)) {
  929. $a = self::getApp();
  930. $stamp1 = microtime(true);
  931. $ch = @curl_init($match[1]);
  932. @curl_setopt($ch, CURLOPT_NOBODY, true);
  933. @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  934. @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
  935. @curl_exec($ch);
  936. $curl_info = @curl_getinfo($ch);
  937. $a->save_timestamp($stamp1, "network");
  938. // if its a link to a picture then embed this picture
  939. if (substr($curl_info["content_type"], 0, 6) == "image/") {
  940. $text = "[img]" . $match[1] . "[/img]";
  941. } else {
  942. $text = "[img]" . $match[2] . "[/img]";
  943. // if its not a picture then look if its a page that contains a picture link
  944. $body = Network::fetchUrl($match[1]);
  945. $doc = new DOMDocument();
  946. @$doc->loadHTML($body);
  947. $xpath = new DOMXPath($doc);
  948. $list = $xpath->query("//meta[@name]");
  949. foreach ($list as $node) {
  950. $attr = [];
  951. if ($node->attributes->length) {
  952. foreach ($node->attributes as $attribute) {
  953. $attr[$attribute->name] = $attribute->value;
  954. }
  955. }
  956. if (strtolower($attr["name"]) == "twitter:image") {
  957. $text = "[img]" . $attr["content"] . "[/img]";
  958. }
  959. }
  960. }
  961. Cache::set($match[1], $text);
  962. }
  963. return $text;
  964. }
  965. public static function cleanPictureLinks($text)
  966. {
  967. $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
  968. return $return;
  969. }
  970. private static function textHighlightCallback($match)
  971. {
  972. // Fallback in case the language doesn't exist
  973. $return = '[code]' . $match[2] . '[/code]';
  974. if (in_array(strtolower($match[1]),
  975. ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby',
  976. 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh', 'bash'])
  977. ) {
  978. $return = text_highlight($match[2], strtolower($match[1]));
  979. }
  980. return $return;
  981. }
  982. /**
  983. * @brief Converts a BBCode message to HTML message
  984. *
  985. * BBcode 2 HTML was written by WAY2WEB.net
  986. * extended to work with Mistpark/Friendica - Mike Macgirvin
  987. *
  988. * Simple HTML values meaning:
  989. * - 0: Friendica display
  990. * - 1: Unused
  991. * - 2: Used for Google+, Windows Phone push, Friendica API
  992. * - 3: Used before converting to Markdown in bb2diaspora.php
  993. * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
  994. * - 5: Unused
  995. * - 6: Unused
  996. * - 7: Used for dfrn, OStatus
  997. * - 8: Used for WP backlink text setting
  998. *
  999. * @param string $text
  1000. * @param bool $try_oembed
  1001. * @param int $simple_html
  1002. * @param bool $for_plaintext
  1003. * @return string
  1004. */
  1005. public static function convert($text, $try_oembed = true, $simple_html = false, $for_plaintext = false)
  1006. {
  1007. $a = self::getApp();
  1008. /*
  1009. * preg_match_callback function to replace potential Oembed tags with Oembed content
  1010. *
  1011. * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
  1012. * $match[1] = $url
  1013. * $match[2] = $title or absent
  1014. */
  1015. $try_oembed_callback = function ($match)
  1016. {
  1017. $url = $match[1];
  1018. $title = defaults($match, 2, null);
  1019. try {
  1020. $return = OEmbed::getHTML($url, $title);
  1021. } catch (Exception $ex) {
  1022. $return = $match[0];
  1023. }
  1024. return $return;
  1025. };
  1026. // Hide all [noparse] contained bbtags by spacefying them
  1027. // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
  1028. $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
  1029. $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
  1030. $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
  1031. // Remove the abstract element. It is a non visible element.
  1032. $text = self::stripAbstract($text);
  1033. // Move all spaces out of the tags
  1034. $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
  1035. $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
  1036. // Extract the private images which use data urls since preg has issues with
  1037. // large data sizes. Stash them away while we do bbcode conversion, and then put them back
  1038. // in after we've done all the regex matching. We cannot use any preg functions to do this.
  1039. $extracted = self::extractImagesFromItemBody($text);
  1040. $text = $extracted['body'];
  1041. $saved_image = $extracted['images'];
  1042. // If we find any event code, turn it into an event.
  1043. // After we're finished processing the bbcode we'll
  1044. // replace all of the event code with a reformatted version.
  1045. $ev = Event::fromBBCode($text);
  1046. // Replace any html brackets with HTML Entities to prevent executing HTML or script
  1047. // Don't use strip_tags here because it breaks [url] search by replacing & with amp
  1048. $text = str_replace("<", "&lt;", $text);
  1049. $text = str_replace(">", "&gt;", $text);
  1050. // remove some newlines before the general conversion
  1051. $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
  1052. $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
  1053. $text = preg_replace("/\n\[code\]/ism", "[code]", $text);
  1054. $text = preg_replace("/\[\/code\]\n/ism", "[/code]", $text);
  1055. // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
  1056. if (!$try_oembed) {
  1057. $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
  1058. }
  1059. // Check for [code] text here, before the linefeeds are messed with.
  1060. // The highlighter will unescape and re-escape the content.
  1061. if (strpos($text, '[code=') !== false) {
  1062. $text = preg_replace_callback("/\[code=(.*?)\](.*?)\[\/code\]/ism", 'self::textHighlightCallback', $text);
  1063. }
  1064. // Convert new line chars to html <br /> tags
  1065. // nlbr seems to be hopelessly messed up
  1066. // $Text = nl2br($Text);
  1067. // We'll emulate it.
  1068. $text = trim($text);
  1069. $text = str_replace("\r\n", "\n", $text);
  1070. // removing multiplicated newlines
  1071. if (Config::get("system", "remove_multiplicated_lines")) {
  1072. $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
  1073. "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
  1074. $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
  1075. "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
  1076. do {
  1077. $oldtext = $text;
  1078. $text = str_replace($search, $replace, $text);
  1079. } while ($oldtext != $text);
  1080. }
  1081. // Set up the parameters for a URL search string
  1082. $URLSearchString = "^\[\]";
  1083. // Set up the parameters for a MAIL search string
  1084. $MAILSearchString = $URLSearchString;
  1085. // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text
  1086. if (!$for_plaintext) {
  1087. // Autolink feature (thanks to http://code.seebz.net/p/autolink-php/)
  1088. // Currently disabled, since the function is too greedy
  1089. // $autolink_regex = "`([^\]\=\"']|^)(https?\://[^\s<]+[^\s<\.\)])`ism";
  1090. $autolink_regex = "/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism";
  1091. $text = preg_replace($autolink_regex, '$1[url]$2[/url]', $text);
  1092. if ($simple_html == 7) {
  1093. $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
  1094. $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
  1095. }
  1096. } else {
  1097. $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text);
  1098. $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
  1099. }
  1100. // Handle attached links or videos
  1101. $text = self::convertAttachment($text, $simple_html, $try_oembed);
  1102. $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
  1103. // Remove all hashtag addresses
  1104. if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7])) {
  1105. $text = preg_replace("/([#@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $text);
  1106. } elseif ($simple_html == 3) {
  1107. // The ! is converted to @ since Diaspora only understands the @
  1108. $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
  1109. '@<a href="$2">$3</a>',
  1110. $text);
  1111. } elseif ($simple_html == 7) {
  1112. $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
  1113. '$1<span class="vcard"><a href="$2" class="url" title="$3"><span class="fn nickname mention">$3</span></a></span>',
  1114. $text);
  1115. } elseif (!$simple_html) {
  1116. $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
  1117. '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
  1118. $text);
  1119. }
  1120. // Bookmarks in red - will be converted to bookmarks in friendica
  1121. $text = preg_replace("/#\^\[url\]([$URLSearchString]*)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
  1122. $text = preg_replace("/#\^\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
  1123. $text = preg_replace("/#\[url\=[$URLSearchString]*\]\^\[\/url\]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i",
  1124. "[bookmark=$1]$2[/bookmark]", $text);
  1125. if (in_array($simple_html, [2, 6, 7, 8, 9])) {
  1126. $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
  1127. //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
  1128. $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
  1129. }
  1130. if ($simple_html == 5) {
  1131. $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
  1132. }
  1133. // Perform URL Search
  1134. if ($try_oembed) {
  1135. $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
  1136. }
  1137. if ($simple_html == 5) {
  1138. $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
  1139. } else {
  1140. $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
  1141. }
  1142. // Handle Diaspora posts
  1143. $text = preg_replace_callback(
  1144. "&\[url=/posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
  1145. function ($match) {
  1146. return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
  1147. }, $text
  1148. );
  1149. $text = preg_replace_callback(
  1150. "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
  1151. function ($match) {
  1152. return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
  1153. }, $text
  1154. );
  1155. // Server independent link to posts and comments
  1156. // See issue: https://github.com/diaspora/diaspora_federation/issues/75
  1157. $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
  1158. $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
  1159. $text = preg_replace("/([#])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
  1160. '$1<a href="' . System::baseUrl() . '/search?tag=$3" class="tag" title="$3">$3</a>', $text);
  1161. $text = preg_replace("/\[url\=([$URLSearchString]*)\]#(.*?)\[\/url\]/ism",
  1162. '#<a href="' . System::baseUrl() . '/search?tag=$2" class="tag" title="$2">$2</a>', $text);
  1163. $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
  1164. $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
  1165. //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $Text);
  1166. // Red compatibility, though the link can't be authenticated on Friendica
  1167. $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
  1168. // we may need to restrict this further if it picks up too many strays
  1169. // link acct:user@host to a webfinger profile redirector
  1170. $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', '<a href="' . System::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
  1171. // Perform MAIL Search
  1172. $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
  1173. $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
  1174. // leave open the posibility of [map=something]
  1175. // this is replaced in prepare_body() which has knowledge of the item location
  1176. if (strpos($text, '[/map]') !== false) {
  1177. $text = preg_replace_callback(
  1178. "/\[map\](.*?)\[\/map\]/ism",
  1179. function ($match) use ($simple_html) {
  1180. return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
  1181. },
  1182. $text
  1183. );
  1184. }
  1185. if (strpos($text, '[map=') !== false) {
  1186. $text = preg_replace_callback(
  1187. "/\[map=(.*?)\]/ism",
  1188. function ($match) use ($simple_html) {
  1189. return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
  1190. },
  1191. $text
  1192. );
  1193. }
  1194. if (strpos($text, '[map]') !== false) {
  1195. $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
  1196. }
  1197. // Check for headers
  1198. $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
  1199. $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
  1200. $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
  1201. $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
  1202. $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
  1203. $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
  1204. // Check for paragraph
  1205. $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
  1206. // Check for bold text
  1207. $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
  1208. // Check for Italics text
  1209. $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
  1210. // Check for Underline text
  1211. $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
  1212. // Check for strike-through text
  1213. $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
  1214. // Check for over-line text
  1215. $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
  1216. // Check for colored text
  1217. $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
  1218. // Check for sized text
  1219. // [size=50] --> font-size: 50px (with the unit).
  1220. $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
  1221. $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
  1222. // Check for centered text
  1223. $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
  1224. // Check for list text
  1225. $text = str_replace("[*]", "<li>", $text);
  1226. // Check for style sheet commands
  1227. $text = preg_replace_callback(
  1228. "(\[style=(.*?)\](.*?)\[\/style\])ism",
  1229. function ($match) {
  1230. return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
  1231. },
  1232. $text
  1233. );
  1234. // Check for CSS classes
  1235. $text = preg_replace_callback(
  1236. "(\[class=(.*?)\](.*?)\[\/class\])ism",
  1237. function ($match) {
  1238. return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
  1239. },
  1240. $text
  1241. );
  1242. // handle nested lists
  1243. $endlessloop = 0;
  1244. while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
  1245. ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
  1246. ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
  1247. ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
  1248. $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
  1249. $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
  1250. $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
  1251. $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
  1252. $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
  1253. $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
  1254. $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
  1255. $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
  1256. $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
  1257. $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
  1258. }
  1259. $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
  1260. $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
  1261. $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
  1262. $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
  1263. $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
  1264. $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
  1265. $text = str_replace('[hr]', '<hr />', $text);
  1266. // This is actually executed in prepare_body()
  1267. $text = str_replace('[nosmile]', '', $text);
  1268. // Check for font change text
  1269. $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
  1270. // Declare the format for [code] layout
  1271. $CodeLayout = '<code>$1</code>';
  1272. // Check for [code] text
  1273. $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text);
  1274. // Declare the format for [spoiler] layout
  1275. $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
  1276. // Check for [spoiler] text
  1277. // handle nested quotes
  1278. $endlessloop = 0;
  1279. while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
  1280. $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
  1281. }
  1282. // Check for [spoiler=Author] text
  1283. $t_wrote = L10n::t('$1 wrote:');
  1284. // handle nested quotes
  1285. $endlessloop = 0;
  1286. while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
  1287. $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
  1288. "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
  1289. $text);
  1290. }
  1291. // Declare the format for [quote] layout
  1292. $QuoteLayout = '<blockquote>$1</blockquote>';
  1293. // Check for [quote] text
  1294. // handle nested quotes
  1295. $endlessloop = 0;
  1296. while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
  1297. $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
  1298. }
  1299. // Check for [quote=Author] text
  1300. $t_wrote = L10n::t('$1 wrote:');
  1301. // handle nested quotes
  1302. $endlessloop = 0;
  1303. while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
  1304. $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
  1305. "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
  1306. $text);
  1307. }
  1308. // [img=widthxheight]image source[/img]
  1309. $text = preg_replace_callback(
  1310. "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
  1311. function ($matches) use ($simple_html) {
  1312. if (strpos($matches[3], "data:image/") === 0) {
  1313. return $matches[0];
  1314. }
  1315. $matches[3] = self::proxyUrl($matches[3], $simple_html);
  1316. return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
  1317. },
  1318. $text
  1319. );
  1320. $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
  1321. $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
  1322. $text = preg_replace_callback("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism",
  1323. function ($matches) use ($simple_html) {
  1324. $matches[1] = self::proxyUrl($matches[1], $simple_html);
  1325. $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
  1326. return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '">';
  1327. },
  1328. $text);
  1329. // Images
  1330. // [img]pathtoimage[/img]
  1331. $text = preg_replace_callback(
  1332. "/\[img\](.*?)\[\/img\]/ism",
  1333. function ($matches) use ($simple_html) {
  1334. if (strpos($matches[1], "data:image/") === 0) {
  1335. return $matches[0];
  1336. }
  1337. $matches[1] = self::proxyUrl($matches[1], $simple_html);
  1338. return "[img]" . $matches[1] . "[/img]";
  1339. },
  1340. $text
  1341. );
  1342. $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
  1343. $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
  1344. // Shared content
  1345. $text = preg_replace_callback("/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
  1346. function ($match) use ($simple_html) {
  1347. return self::convertShare($match, $simple_html);
  1348. }, $text);
  1349. $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . L10n::t('Encrypted content') . '" /><br />', $text);
  1350. $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $text);
  1351. //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $Text);
  1352. // Try to Oembed
  1353. if ($try_oembed) {
  1354. $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism", '<video src="$1" controls="controls" width="' . $a->videowidth . '" height="' . $a->videoheight . '" loop="true"><a href="$1">$1</a></video>', $text);
  1355. $text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3).*?)\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
  1356. $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
  1357. $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
  1358. } else {
  1359. $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
  1360. '<a href="$1" target="_blank">$1</a>', $text);
  1361. $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
  1362. '<a href="$1" target="_blank">$1</a>', $text);
  1363. }
  1364. // html5 video and audio
  1365. if ($try_oembed) {
  1366. $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
  1367. } else {
  1368. $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
  1369. }
  1370. // Youtube extensions
  1371. if ($try_oembed) {
  1372. $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
  1373. $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
  1374. $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
  1375. }
  1376. $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
  1377. $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
  1378. $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
  1379. if ($try_oembed) {
  1380. $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '<iframe width="' . $a->videowidth . '" height="' . $a->videoheight . '" src="https://www.youtube.com/embed/$1" frameborder="0" ></iframe>', $text);
  1381. } else {
  1382. $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
  1383. '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
  1384. }
  1385. if ($try_oembed) {
  1386. $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
  1387. $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
  1388. }
  1389. $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
  1390. $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
  1391. if ($try_oembed) {
  1392. $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '<iframe width="' . $a->videowidth . '" height="' . $a->videoheight . '" src="https://player.vimeo.com/video/$1" frameborder="0" ></iframe>', $text);
  1393. } else {
  1394. $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
  1395. '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
  1396. }
  1397. // oembed tag
  1398. $text = OEmbed::BBCode2HTML($text);
  1399. // Avoid triple linefeeds through oembed
  1400. $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
  1401. // If we found an event earlier, strip out all the event code and replace with a reformatted version.
  1402. // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
  1403. // Summary (e.g. title) is required, earlier revisions only required description (in addition to
  1404. // start which is always required). Allow desc with a missing summary for compatibility.
  1405. if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) {
  1406. $sub = Event::getHTML($ev, $simple_html);
  1407. $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
  1408. $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
  1409. $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
  1410. $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
  1411. $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
  1412. $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
  1413. $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
  1414. }
  1415. // Replace non graphical smilies for external posts
  1416. if ($simple_html) {
  1417. $text = Smilies::replace($text, false, true);
  1418. }
  1419. // Replace inline code blocks
  1420. $text = preg_replace_callback("|(?!<br[^>]*>)<code>([^<]*)</code>(?!<br[^>]*>)|ism",
  1421. function ($match) use ($simple_html) {
  1422. $return = '<key>' . $match[1] . '</key>';
  1423. // Use <code> for Diaspora inline code blocks
  1424. if ($simple_html === 3) {
  1425. $return = '<code>' . $match[1] . '</code>';
  1426. }
  1427. return $return;
  1428. }
  1429. , $text);
  1430. // Unhide all [noparse] contained bbtags unspacefying them
  1431. // and triming the [noparse] tag.
  1432. $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
  1433. $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
  1434. $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
  1435. /// @todo What is the meaning of these lines?
  1436. $text = preg_replace('/\[\&amp\;([#a-z0-9]+)\;\]/', '&$1;', $text);
  1437. $text = preg_replace('/\&\#039\;/', '\'', $text);
  1438. // Currently deactivated, it made problems with " inside of alt texts.
  1439. //$text = preg_replace('/\&quot\;/', '"', $text);
  1440. // fix any escaped ampersands that may have been converted into links
  1441. $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&amp\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
  1442. // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
  1443. $allowed_src_protocols = ['http', 'redir', 'cid'];
  1444. $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
  1445. '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
  1446. // sanitize href attributes (only whitelisted protocols URLs)
  1447. // default value for backward compatibility
  1448. $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']);
  1449. // Always allowed protocol even if config isn't set or not including it
  1450. $allowed_link_protocols[] = 'http';
  1451. $allowed_link_protocols[] = 'redir/';
  1452. $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
  1453. $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
  1454. if ($saved_image) {
  1455. $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
  1456. }
  1457. // Clean up the HTML by loading and saving the HTML with the DOM.
  1458. // Bad structured html can break a whole page.
  1459. // For performance reasons do it only with ativated item cache or at export.
  1460. if (!$try_oembed || (get_itemcachepath() != "")) {
  1461. $doc = new DOMDocument();
  1462. $doc->preserveWhiteSpace = false;
  1463. $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
  1464. $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
  1465. $encoding = '<?xml encoding="UTF-8">';
  1466. @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
  1467. $doc->encoding = 'UTF-8';
  1468. $text = $doc->saveHTML();
  1469. $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
  1470. $text = str_replace('<br></li>', '</li>', $text);
  1471. //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
  1472. }
  1473. // Clean up some useless linebreaks in lists
  1474. //$Text = str_replace('<br /><ul', '<ul ', $Text);
  1475. //$Text = str_replace('</ul><br />', '</ul>', $Text);
  1476. //$Text = str_replace('</li><br />', '</li>', $Text);
  1477. //$Text = str_replace('<br /><li>', '<li>', $Text);
  1478. //$Text = str_replace('<br /><ul', '<ul ', $Text);
  1479. Addon::callHooks('bbcode', $text);
  1480. return trim($text);
  1481. }
  1482. /**
  1483. * @brief Strips the "abstract" tag from the provided text
  1484. *
  1485. * @param string $text The text with BBCode
  1486. * @return string The same text - but without "abstract" element
  1487. */
  1488. public static function stripAbstract($text)
  1489. {
  1490. $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
  1491. $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
  1492. return $text;
  1493. }
  1494. /**
  1495. * @brief Returns the value of the "abstract" element
  1496. *
  1497. * @param string $text The text that maybe contains the element
  1498. * @param string $addon The addon for which the abstract is meant for
  1499. * @return string The abstract
  1500. */
  1501. public static function getAbstract($text, $addon = "")
  1502. {
  1503. $abstract = "";
  1504. $abstracts = [];
  1505. $addon = strtolower($addon);
  1506. if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
  1507. foreach ($results AS $result) {
  1508. $abstracts[strtolower($result[1])] = $result[2];
  1509. }
  1510. }
  1511. if (isset($abstracts[$addon])) {
  1512. $abstract = $abstracts[$addon];
  1513. }
  1514. if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
  1515. $abstract = $result[1];
  1516. }
  1517. return $abstract;
  1518. }
  1519. /**
  1520. * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
  1521. *
  1522. * @param array $match Matching values for the callback
  1523. * @return string Replaced mention
  1524. */
  1525. private static function bbCodeMention2DiasporaCallback($match)
  1526. {
  1527. $contact = Contact::getDetailsByURL($match[3]);
  1528. if (empty($contact['addr'])) {
  1529. $contact = Probe::uri($match[3]);
  1530. }
  1531. if (empty($contact['addr'])) {
  1532. return $match[0];
  1533. }
  1534. $mention = '@{' . $match[2] . '; ' . $contact['addr'] . '}';
  1535. return $mention;
  1536. }
  1537. /**
  1538. * @brief Converts a BBCode text into Markdown
  1539. *
  1540. * This function converts a BBCode item body to be sent to Markdown-enabled
  1541. * systems like Diaspora and Libertree
  1542. *
  1543. * @param string $text
  1544. * @param bool $for_diaspora Diaspora requires more changes than Libertree
  1545. * @return string
  1546. */
  1547. public static function toMarkdown($text, $for_diaspora = true)
  1548. {
  1549. $a = self::getApp();
  1550. $original_text = $text;
  1551. // Since Diaspora is creating a summary for links, this function removes them before posting
  1552. if ($for_diaspora) {
  1553. $text = self::removeShareInformation($text);
  1554. }
  1555. /**
  1556. * Transform #tags, strip off the [url] and replace spaces with underscore
  1557. */
  1558. $url_search_string = "^\[\]";
  1559. $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
  1560. function ($matches) {
  1561. return '#' . str_replace(' ', '_', $matches[2]);
  1562. },
  1563. $text
  1564. );
  1565. // Converting images with size parameters to simple images. Markdown doesn't know it.
  1566. $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
  1567. // Extracting multi-line code blocks before the whitespace processing/code highlighter in self::convert()
  1568. $codeblocks = [];
  1569. $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#is",
  1570. function ($matches) use (&$codeblocks) {
  1571. $return = $matches[0];
  1572. if (strpos($matches[2], "\n") !== false) {
  1573. $return = '#codeblock-' . count($codeblocks) . '#';
  1574. $prefix = '````' . $matches[1] . PHP_EOL;
  1575. $codeblocks[] = $prefix . trim($matches[2]) . PHP_EOL . '````';
  1576. }
  1577. return $return;
  1578. },
  1579. $text
  1580. );
  1581. // Convert it to HTML - don't try oembed
  1582. if ($for_diaspora) {
  1583. $text = self::convert($text, false, 3);
  1584. // Add all tags that maybe were removed
  1585. if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
  1586. $tagline = "";
  1587. foreach ($tags[2] as $tag) {
  1588. $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
  1589. if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
  1590. $tagline .= '#' . $tag . ' ';
  1591. }
  1592. }
  1593. $text = $text . " " . $tagline;
  1594. }
  1595. } else {
  1596. $text = self::convert($text, false, 4);
  1597. }
  1598. // mask some special HTML chars from conversation to markdown
  1599. $text = str_replace(['&lt;', '&gt;', '&amp;'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text);
  1600. // If a link is followed by a quote then there should be a newline before it
  1601. // Maybe we should make this newline at every time before a quote.
  1602. $text = str_replace(["</a><blockquote>"], ["</a><br><blockquote>"], $text);
  1603. $stamp1 = microtime(true);
  1604. // Now convert HTML to Markdown
  1605. $converter = new HtmlConverter();
  1606. $text = $converter->convert($text);
  1607. // unmask the special chars back to HTML
  1608. $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['&lt;', '&gt;', '&amp;'], $text);
  1609. $a->save_timestamp($stamp1, "parser");
  1610. // Libertree has a problem with escaped hashtags.
  1611. $text = str_replace(['\#'], ['#'], $text);
  1612. // Remove any leading or trailing whitespace, as this will mess up
  1613. // the Diaspora signature verification and cause the item to disappear
  1614. $text = trim($text);
  1615. if ($for_diaspora) {
  1616. $url_search_string = "^\[\]";
  1617. $text = preg_replace_callback(
  1618. "/([@]\[(.*?)\])\(([$url_search_string]*?)\)/ism",
  1619. ['self', 'bbCodeMention2DiasporaCallback'],
  1620. $text
  1621. );
  1622. }
  1623. // Restore code blocks
  1624. $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
  1625. function ($matches) use ($codeblocks) {
  1626. $return = '';
  1627. if (isset($codeblocks[intval($matches[1])])) {
  1628. $return = $codeblocks[$matches[1]];
  1629. }
  1630. return $return;
  1631. },
  1632. $text
  1633. );
  1634. Addon::callHooks('bb2diaspora', $text);
  1635. return $text;
  1636. }
  1637. }