3 * @file src/Content/Text/BBCode.php
5 namespace Friendica\Content\Text;
10 use Friendica\Content\OEmbed;
11 use Friendica\Content\Smilies;
12 use Friendica\Content\Text\Plaintext;
13 use Friendica\Core\Addon;
14 use Friendica\Core\Cache;
15 use Friendica\Core\Config;
16 use Friendica\Core\L10n;
17 use Friendica\Core\Protocol;
18 use Friendica\Core\PConfig;
19 use Friendica\Core\System;
20 use Friendica\Model\Contact;
21 use Friendica\Object\Image;
22 use Friendica\Util\Map;
23 use Friendica\Util\Network;
24 use Friendica\Util\ParseUrl;
26 require_once "include/event.php";
27 require_once "include/html2plain.php";
28 require_once "mod/proxy.php";
33 * @brief Fetches attachment data that were generated the old way
35 * @param string $body Message body
37 * 'type' -> Message type ("link", "video", "photo")
38 * 'text' -> Text before the shared message
39 * 'after' -> Text after the shared message
40 * 'image' -> Preview image of the message
41 * 'url' -> Url to the attached message
42 * 'title' -> Title of the attachment
43 * 'description' -> Description of the attachment
45 private static function getOldAttachmentData($body)
49 // Simplify image codes
50 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
52 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
53 foreach ($attached as $data) {
54 if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
58 $post["type"] = substr($data[1], 5);
60 $pos = strpos($body, $data[0]);
62 $post["text"] = trim(substr($body, 0, $pos));
63 $post["after"] = trim(substr($body, $pos + strlen($data[0])));
65 $post["text"] = trim(str_replace($data[0], "", $body));
68 $attacheddata = $data[2];
70 $URLSearchString = "^\[\]";
72 if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
74 $picturedata = Image::getInfoFromURL($matches[1]);
76 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
77 $post["image"] = $matches[1];
79 $post["preview"] = $matches[1];
83 if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
84 $post["url"] = $matches[1];
85 $post["title"] = $matches[2];
87 if (($post["url"] == "") && (in_array($post["type"], ["link", "video"]))
88 && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
89 $post["url"] = $matches[1];
92 // Search for description
93 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
94 $post["description"] = $matches[1];
102 * @brief Fetches attachment data that were generated with the "attachment" element
104 * @param string $body Message body
106 * 'type' -> Message type ("link", "video", "photo")
107 * 'text' -> Text before the shared message
108 * 'after' -> Text after the shared message
109 * 'image' -> Preview image of the message
110 * 'url' -> Url to the attached message
111 * 'title' -> Title of the attachment
112 * 'description' -> Description of the attachment
114 public static function getAttachmentData($body)
118 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
119 return self::getOldAttachmentData($body);
122 $attributes = $match[2];
124 $data["text"] = trim($match[1]);
127 preg_match("/type='(.*?)'/ism", $attributes, $matches);
128 if (x($matches, 1)) {
129 $type = strtolower($matches[1]);
132 preg_match('/type="(.*?)"/ism', $attributes, $matches);
133 if (x($matches, 1)) {
134 $type = strtolower($matches[1]);
141 if (!in_array($type, ["link", "audio", "photo", "video"])) {
146 $data["type"] = $type;
150 preg_match("/url='(.*?)'/ism", $attributes, $matches);
151 if (x($matches, 1)) {
155 preg_match('/url="(.*?)"/ism', $attributes, $matches);
156 if (x($matches, 1)) {
161 $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
165 preg_match("/title='(.*?)'/ism", $attributes, $matches);
166 if (x($matches, 1)) {
167 $title = $matches[1];
170 preg_match('/title="(.*?)"/ism', $attributes, $matches);
171 if (x($matches, 1)) {
172 $title = $matches[1];
176 $title = BBCode::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
177 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
178 $title = str_replace(["[", "]"], ["[", "]"], $title);
179 $data["title"] = $title;
183 preg_match("/image='(.*?)'/ism", $attributes, $matches);
184 if (x($matches, 1)) {
185 $image = $matches[1];
188 preg_match('/image="(.*?)"/ism', $attributes, $matches);
189 if (x($matches, 1)) {
190 $image = $matches[1];
194 $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
198 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
199 if (x($matches, 1)) {
200 $preview = $matches[1];
203 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
204 if (x($matches, 1)) {
205 $preview = $matches[1];
208 if ($preview != "") {
209 $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
212 $data["description"] = trim($match[3]);
214 $data["after"] = trim($match[4]);
219 public static function getAttachedData($body, $item = [])
223 - type: link, video, photo
231 $has_title = !empty($item['title']);
232 $plink = (!empty($item['plink']) ? $item['plink'] : '');
233 $post = self::getAttachmentData($body);
235 // if nothing is found, it maybe having an image.
236 if (!isset($post["type"])) {
237 // Simplify image codes
238 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
240 $URLSearchString = "^\[\]";
241 if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
242 if ((count($pictures) == 1) && !$has_title) {
243 // Checking, if the link goes to a picture
244 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
247 // Sometimes photo posts to the own album are not detected at the start.
248 // So we seem to cannot use the cache for these cases. That's strange.
249 if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
250 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
253 if ($data["type"] == "photo") {
254 $post["type"] = "photo";
255 if (isset($data["images"][0])) {
256 $post["image"] = $data["images"][0]["src"];
257 $post["url"] = $data["url"];
259 $post["image"] = $data["url"];
262 $post["preview"] = $pictures[0][2];
263 $post["text"] = str_replace($pictures[0][0], "", $body);
265 $imgdata = Image::getInfoFromURL($pictures[0][1]);
266 if (substr($imgdata["mime"], 0, 6) == "image/") {
267 $post["type"] = "photo";
268 $post["image"] = $pictures[0][1];
269 $post["preview"] = $pictures[0][2];
270 $post["text"] = str_replace($pictures[0][0], "", $body);
273 } elseif (count($pictures) > 0) {
274 $post["type"] = "link";
275 $post["url"] = $plink;
276 $post["image"] = $pictures[0][2];
277 $post["text"] = $body;
279 } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
280 if ((count($pictures) == 1) && !$has_title) {
281 $post["type"] = "photo";
282 $post["image"] = $pictures[0][1];
283 $post["text"] = str_replace($pictures[0][0], "", $body);
284 } elseif (count($pictures) > 0) {
285 $post["type"] = "link";
286 $post["url"] = $plink;
287 $post["image"] = $pictures[0][1];
288 $post["text"] = $body;
292 // Test for the external links
293 preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
294 preg_match_all("(\[url\=([$URLSearchString]*)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
296 $links = array_merge($links1, $links2);
298 // If there is only a single one, then use it.
299 // This should cover link posts via API.
300 if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
301 $post["type"] = "link";
302 $post["text"] = trim($body);
303 $post["url"] = $links[0][1];
306 // Now count the number of external media links
307 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
308 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
309 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
310 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
312 // Add them to the other external links
313 $links = array_merge($links, $links1, $links2, $links3, $links4);
315 // Are there more than one?
316 if (count($links) > 1) {
317 // The post will be the type "text", which means a blog post
318 unset($post["type"]);
319 $post["url"] = $plink;
322 if (!isset($post["type"])) {
323 $post["type"] = "text";
324 $post["text"] = trim($body);
326 } elseif (isset($post["url"]) && ($post["type"] == "video")) {
327 $data = ParseUrl::getSiteinfoCached($post["url"], true);
329 if (isset($data["images"][0])) {
330 $post["image"] = $data["images"][0]["src"];
338 * @brief Convert a message into plaintext for connectors to other networks
340 * @param array $b The message array that is about to be posted
341 * @param int $limit The maximum number of characters when posting to that network
342 * @param bool $includedlinks Has an attached link to be included into the message?
343 * @param int $htmlmode This triggers the behaviour of the bbcode conversion
344 * @param string $target_network Name of the network where the post should go to.
346 * @return string The converted message
348 public static function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "")
350 // Remove the hash tags
351 $URLSearchString = "^\[\]";
352 $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]);
354 // Add an URL element if the text contains a raw link
355 $body = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2[/url]', $body);
357 // Remove the abstract
358 $body = self::stripAbstract($body);
360 // At first look at data that is attached via "type-..." stuff
361 // This will hopefully replaced with a dedicated bbcode later
362 //$post = self::getAttachedData($b["body"]);
363 $post = self::getAttachedData($body, $b);
365 if (($b["title"] != "") && ($post["text"] != "")) {
366 $post["text"] = trim($b["title"]."\n\n".$post["text"]);
367 } elseif ($b["title"] != "") {
368 $post["text"] = trim($b["title"]);
373 // Fetch the abstract from the given target network
374 if ($target_network != "") {
375 $default_abstract = self::getAbstract($b["body"]);
376 $abstract = self::getAbstract($b["body"], $target_network);
378 // If we post to a network with no limit we only fetch
379 // an abstract exactly for this network
380 if (($limit == 0) && ($abstract == $default_abstract)) {
383 } else {// Try to guess the correct target network
386 $abstract = self::getAbstract($b["body"], NETWORK_TWITTER);
389 $abstract = self::getAbstract($b["body"], NETWORK_STATUSNET);
392 $abstract = self::getAbstract($b["body"], NETWORK_APPNET);
394 default: // We don't know the exact target.
395 // We fetch an abstract since there is a posting limit.
397 $abstract = self::getAbstract($b["body"]);
402 if ($abstract != "") {
403 $post["text"] = $abstract;
405 if ($post["type"] == "text") {
406 $post["type"] = "link";
407 $post["url"] = $b["plink"];
411 $html = BBCode::convert($post["text"].$post["after"], false, $htmlmode);
412 $msg = html2plain($html, 0, true);
413 $msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8'));
416 if ($includedlinks) {
417 if ($post["type"] == "link") {
418 $link = $post["url"];
419 } elseif ($post["type"] == "text") {
420 $link = $post["url"];
421 } elseif ($post["type"] == "video") {
422 $link = $post["url"];
423 } elseif ($post["type"] == "photo") {
424 $link = $post["image"];
427 if (($msg == "") && isset($post["title"])) {
428 $msg = trim($post["title"]);
431 if (($msg == "") && isset($post["description"])) {
432 $msg = trim($post["description"]);
435 // If the link is already contained in the post, then it neeedn't to be added again
436 // But: if the link is beyond the limit, then it has to be added.
437 if (($link != "") && strstr($msg, $link)) {
438 $pos = strpos($msg, $link);
440 // Will the text be shortened in the link?
441 // Or is the link the last item in the post?
442 if (($limit > 0) && ($pos < $limit) && (($pos + 23 > $limit) || ($pos + strlen($link) == strlen($msg)))) {
443 $msg = trim(str_replace($link, "", $msg));
444 } elseif (($limit == 0) || ($pos < $limit)) {
445 // The limit has to be increased since it will be shortened - but not now
446 // Only do it with Twitter (htmlmode = 8)
447 if (($limit > 0) && (strlen($link) > 23) && ($htmlmode == 8)) {
448 $limit = $limit - 23 + strlen($link);
453 if ($post["type"] == "text") {
461 // Reduce multiple spaces
462 // When posted to a network with limited space, we try to gain space where possible
463 while (strpos($msg, " ") !== false) {
464 $msg = str_replace(" ", " ", $msg);
467 // Twitter is using its own limiter, so we always assume that shortened links will have this length
468 if (iconv_strlen($link, "UTF-8") > 0) {
469 $limit = $limit - 23;
472 if (iconv_strlen($msg, "UTF-8") > $limit) {
473 if (($post["type"] == "text") && isset($post["url"])) {
474 $post["url"] = $b["plink"];
475 } elseif (!isset($post["url"])) {
476 $limit = $limit - 23;
477 $post["url"] = $b["plink"];
478 // Which purpose has this line? It is now uncommented, but left as a reminder
479 //} elseif (strpos($b["body"], "[share") !== false) {
480 // $post["url"] = $b["plink"];
481 } elseif (PConfig::get($b["uid"], "system", "no_intelligent_shortening")) {
482 $post["url"] = $b["plink"];
484 $msg = Plaintext::shorten($msg, $limit);
488 $post["text"] = trim($msg);
493 public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
495 // Suppress "view full size"
496 if (intval(Config::get('system', 'no_view_full_size'))) {
497 $include_link = false;
500 // Picture addresses can contain special characters
501 $s = htmlspecialchars_decode($srctext);
504 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
506 foreach ($matches as $mtch) {
507 logger('scale_external_image: ' . $mtch[1]);
509 $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
510 if (stristr($mtch[1], $hostname)) {
514 // $scale_replace, if passed, is an array of two elements. The
515 // first is the name of the full-size image. The second is the
516 // name of a remote, scaled-down version of the full size image.
517 // This allows Friendica to display the smaller remote image if
518 // one exists, while still linking to the full-size image
519 if ($scale_replace) {
520 $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
524 $i = Network::fetchUrl($scaled);
529 // guess mimetype from headers or filename
530 $type = Image::guessType($mtch[1], true);
533 $Image = new Image($i, $type);
534 if ($Image->isValid()) {
535 $orig_width = $Image->getWidth();
536 $orig_height = $Image->getHeight();
538 if ($orig_width > 640 || $orig_height > 640) {
539 $Image->scaleDown(640);
540 $new_width = $Image->getWidth();
541 $new_height = $Image->getHeight();
542 logger('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], LOGGER_DEBUG);
545 '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
546 . "\n" . (($include_link)
547 ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
551 logger('scale_external_images: new string: ' . $s, LOGGER_DEBUG);
558 // replace the special char encoding
559 $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
564 * The purpose of this function is to apply system message length limits to
565 * imported messages without including any embedded photos in the length
567 * @brief Truncates imported message body string length to max_import_size
568 * @param string $body
571 public static function limitBodySize($body)
573 $maxlen = get_max_import_size();
575 // If the length of the body, including the embedded images, is smaller
576 // than the maximum, then don't waste time looking for the images
577 if ($maxlen && (strlen($body) > $maxlen)) {
579 logger('the total body length exceeds the limit', LOGGER_DEBUG);
585 $img_start = strpos($orig_body, '[img');
586 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
587 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
588 while (($img_st_close !== false) && ($img_end !== false)) {
590 $img_st_close++; // make it point to AFTER the closing bracket
591 $img_end += $img_start;
592 $img_end += strlen('[/img]');
594 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
595 // This is an embedded image
597 if (($textlen + $img_start) > $maxlen) {
598 if ($textlen < $maxlen) {
599 logger('the limit happens before an embedded image', LOGGER_DEBUG);
600 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
604 $new_body = $new_body . substr($orig_body, 0, $img_start);
605 $textlen += $img_start;
608 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
611 if (($textlen + $img_end) > $maxlen) {
612 if ($textlen < $maxlen) {
613 logger('the limit happens before the end of a non-embedded image', LOGGER_DEBUG);
614 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
618 $new_body = $new_body . substr($orig_body, 0, $img_end);
619 $textlen += $img_end;
622 $orig_body = substr($orig_body, $img_end);
624 if ($orig_body === false) {
625 // in case the body ends on a closing image tag
629 $img_start = strpos($orig_body, '[img');
630 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
631 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
634 if (($textlen + strlen($orig_body)) > $maxlen) {
635 if ($textlen < $maxlen) {
636 logger('the limit happens after the end of the last image', LOGGER_DEBUG);
637 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
640 logger('the text size with embedded images extracted did not violate the limit', LOGGER_DEBUG);
641 $new_body = $new_body . $orig_body;
651 * Processes [attachment] tags
653 * Note: Can produce a [bookmark] tag in the returned string
655 * @brief Processes [attachment] tags
656 * @param string $return
657 * @param bool|int $simplehtml
658 * @param bool $tryoembed
661 private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
663 $data = self::getAttachmentData($return);
668 if (isset($data["title"])) {
669 $data["title"] = strip_tags($data["title"]);
670 $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
673 if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) {
674 $data["preview"] = $data["image"];
679 if ($simplehtml == 7) {
680 $return = self::convertUrlForMastodon($data["url"]);
681 } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
682 $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data["url"], $data["title"]);
685 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
686 $return = OEmbed::getHTML($data['url'], $data['title']);
688 throw new Exception('OEmbed is disabled for this attachment.');
690 } catch (Exception $e) {
691 if ($simplehtml != 4) {
692 $return = sprintf('<div class="type-%s">', $data["type"]);
695 if ($data["image"] != "") {
696 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], proxy_url($data["image"]), $data["title"]);
697 } elseif ($data["preview"] != "") {
698 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], proxy_url($data["preview"]), $data["title"]);
701 if (($data["type"] == "photo") && ($data["url"] != "") && ($data["image"] != "")) {
702 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], proxy_url($data["image"]), $data["title"]);
704 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
707 if ($data["description"] != "" && $data["description"] != $data["title"]) {
708 $return .= sprintf('<blockquote>%s</blockquote>', trim(BBCode::convert($data["description"])));
711 if ($data["type"] == "link") {
712 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
715 if ($simplehtml != 4) {
721 return trim($data["text"] . ' ' . $return . ' ' . $data["after"]);
724 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
726 $data = self::getAttachmentData($Text);
731 return $data["text"] . $data["after"];
734 $title = htmlentities($data["title"], ENT_QUOTES, 'UTF-8', false);
735 $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
736 if ($plaintext || (($title != "") && strstr($text, $title))) {
737 $data["title"] = $data["url"];
738 } elseif (($text != "") && strstr($title, $text)) {
739 $data["text"] = $data["title"];
740 $data["title"] = $data["url"];
743 if (($data["text"] == "") && ($data["title"] != "") && ($data["url"] == "")) {
744 return $data["title"] . $data["after"];
747 // If the link already is included in the post, don't add it again
748 if (($data["url"] != "") && strpos($data["text"], $data["url"])) {
749 return $data["text"] . $data["after"];
752 $text = $data["text"];
754 if (($data["url"] != "") && ($data["title"] != "")) {
755 $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
756 } elseif (($data["url"] != "")) {
757 $text .= "\n[url]" . $data["url"] . "[/url]";
760 return $text . "\n" . $data["after"];
763 private static function cleanCss($input)
767 $input = strtolower($input);
769 for ($i = 0; $i < strlen($input); $i++) {
770 $char = substr($input, $i, 1);
772 if (($char >= "a") && ($char <= "z")) {
776 if (!(strpos(" #;:0123456789-_.%", $char) === false)) {
785 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
787 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
788 * @param array $match Array with the matching values
789 * @return string reformatted link including HTML codes
791 private static function convertUrlForMastodonCallback($match)
795 if (isset($match[2]) && ($match[1] != $match[2])) {
799 $parts = parse_url($url);
800 if (!isset($parts['scheme'])) {
804 return self::convertUrlForMastodon($url);
808 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon and GNU Social.
809 * @param string $url URL that is about to be reformatted
810 * @return string reformatted link including HTML codes
812 private static function convertUrlForMastodon($url)
814 $parts = parse_url($url);
815 $scheme = $parts['scheme'] . '://';
816 $styled_url = str_replace($scheme, '', $url);
818 $html = '<a href="%s" class="attachment" rel="nofollow noopener" target="_blank">' .
819 '<span class="invisible">%s</span>';
821 if (strlen($styled_url) > 30) {
822 $html .= '<span class="ellipsis">%s</span>' .
823 '<span class="invisible">%s</span></a>';
825 $ellipsis = substr($styled_url, 0, 30);
826 $rest = substr($styled_url, 30);
827 return sprintf($html, $url, $scheme, $ellipsis, $rest);
830 return sprintf($html, $url, $scheme, $styled_url);
835 * @brief Shortens [url] BBCodes in a format that looks less ugly than the full address. (callback function)
836 * @param array $match Array with the matching values
837 * @return string reformatted link including HTML codes
839 private static function shortenVisibleUrlCallback($match)
843 if (isset($match[2]) && ($match[1] != $match[2])) {
847 $parts = parse_url($url);
848 if (!isset($parts['scheme'])) {
852 return self::shortenVisibleUrl($url);
856 * @brief Shortens [url] BBCodes in a format that looks less ugly than the full address.
857 * @param string $url URL that is about to be reformatted
858 * @return string reformatted link including HTML codes
860 private static function shortenVisibleUrl($url)
862 $parts = parse_url($url);
863 $scheme = $parts['scheme'] . '://';
864 $styled_url = str_replace($scheme, '', $url);
866 if (strlen($styled_url) > 30) {
867 $styled_url = substr($styled_url, 0, 30) . "…";
870 $html = '<a href="%s" target="_blank">%s</a>';
872 return sprintf($html, $url, $styled_url);
876 * [noparse][i]italic[/i][/noparse] turns into
877 * [noparse][ i ]italic[ /i ][/noparse],
878 * to hide them from parser.
880 private static function escapeNoparseCallback($match)
882 $whole_match = $match[0];
883 $captured = $match[1];
884 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
885 $new_str = str_replace($captured, $spacefied, $whole_match);
890 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
891 * now turns back and the [noparse] tags are trimed
892 * returning [i]italic[/i]
894 private static function unescapeNoparseCallback($match)
896 $captured = $match[1];
897 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
902 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
905 * @param string $text Text to search
906 * @param string $name Tag name
907 * @param int $occurrences Number of first occurrences to skip
908 * @return boolean|array
910 public static function getTagPosition($text, $name, $occurrences = 0)
912 if ($occurrences < 0) {
917 for ($i = 0; $i <= $occurrences; $i++) {
918 if ($start_open !== false) {
919 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
923 if ($start_open === false) {
927 $start_equal = strpos($text, '=', $start_open);
928 $start_close = strpos($text, ']', $start_open);
930 if ($start_close === false) {
936 $end_open = strpos($text, '[/' . $name . ']', $start_close);
938 if ($end_open === false) {
944 'open' => $start_open,
945 'close' => $start_close
949 'close' => $end_open + strlen('[/' . $name . ']')
953 if ($start_equal !== false) {
954 $res['start']['equal'] = $start_equal + 1;
961 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
963 * @param type $pattern Preg pattern string
964 * @param type $replace Preg replace string
965 * @param type $name BBCode tag name
966 * @param type $text Text to search
969 public static function pregReplaceInTag($pattern, $replace, $name, $text)
972 $pos = self::getTagPosition($text, $name, $occurrences);
973 while ($pos !== false && $occurrences++ < 1000) {
974 $start = substr($text, 0, $pos['start']['open']);
975 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
976 $end = substr($text, $pos['end']['close']);
977 if ($end === false) {
981 $subject = preg_replace($pattern, $replace, $subject);
982 $text = $start . $subject . $end;
984 $pos = self::getTagPosition($text, $name, $occurrences);
990 private static function extractImagesFromItemBody($body)
997 $img_start = strpos($orig_body, '[img');
998 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
999 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
1000 while (($img_st_close !== false) && ($img_end !== false)) {
1001 $img_st_close++; // make it point to AFTER the closing bracket
1002 $img_end += $img_start;
1004 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
1005 // This is an embedded image
1006 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
1007 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
1011 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
1014 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
1016 if ($orig_body === false) {
1017 // in case the body ends on a closing image tag
1021 $img_start = strpos($orig_body, '[img');
1022 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
1023 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
1026 $new_body = $new_body . $orig_body;
1028 return ['body' => $new_body, 'images' => $saved_image];
1031 private static function interpolateSavedImagesIntoItemBody($body, array $images)
1036 foreach ($images as $image) {
1037 // We're depending on the property of 'foreach' (specified on the PHP website) that
1038 // it loops over the array starting from the first element and going sequentially
1039 // to the last element
1040 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
1041 '<img src="' . proxy_url($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
1049 * Processes [share] tags
1051 * Note: Can produce a [bookmark] tag in the output
1053 * @brief Processes [share] tags
1054 * @param array $share preg_match_callback result array
1055 * @param bool|int $simplehtml
1058 private static function convertShare($share, $simplehtml)
1060 $attributes = $share[2];
1063 preg_match("/author='(.*?)'/ism", $attributes, $matches);
1064 if (x($matches, 1)) {
1065 $author = html_entity_decode($matches[1], ENT_QUOTES, 'UTF-8');
1068 preg_match('/author="(.*?)"/ism', $attributes, $matches);
1069 if (x($matches, 1)) {
1070 $author = $matches[1];
1074 preg_match("/profile='(.*?)'/ism", $attributes, $matches);
1075 if (x($matches, 1)) {
1076 $profile = $matches[1];
1079 preg_match('/profile="(.*?)"/ism', $attributes, $matches);
1080 if (x($matches, 1)) {
1081 $profile = $matches[1];
1085 preg_match("/avatar='(.*?)'/ism", $attributes, $matches);
1086 if (x($matches, 1)) {
1087 $avatar = $matches[1];
1090 preg_match('/avatar="(.*?)"/ism', $attributes, $matches);
1091 if (x($matches, 1)) {
1092 $avatar = $matches[1];
1096 preg_match("/link='(.*?)'/ism", $attributes, $matches);
1097 if (x($matches, 1)) {
1098 $link = $matches[1];
1101 preg_match('/link="(.*?)"/ism', $attributes, $matches);
1102 if (x($matches, 1)) {
1103 $link = $matches[1];
1108 preg_match("/posted='(.*?)'/ism", $attributes, $matches);
1109 if (x($matches, 1)) {
1110 $posted = $matches[1];
1113 preg_match('/posted="(.*?)"/ism', $attributes, $matches);
1114 if (x($matches, 1)) {
1115 $posted = $matches[1];
1118 // We only call this so that a previously unknown contact can be added.
1119 // This is important for the function "Model\Contact::getDetailsByURL()".
1120 // This function then can fetch an entry from the contact table.
1121 Contact::getIdForURL($profile, 0, true);
1123 $data = Contact::getDetailsByURL($profile);
1125 if (x($data, "name") && x($data, "addr")) {
1126 $userid_compact = $data["name"] . " (" . $data["addr"] . ")";
1128 $userid_compact = Protocol::getAddrFromProfileUrl($profile, $author);
1131 if (x($data, "addr")) {
1132 $userid = $data["addr"];
1134 $userid = Protocol::formatMention($profile, $author);
1137 if (x($data, "name")) {
1138 $author = $data["name"];
1141 if (x($data, "micro")) {
1142 $avatar = $data["micro"];
1145 $preshare = trim($share[1]);
1146 if ($preshare != "") {
1147 $preshare .= "<br />";
1150 switch ($simplehtml) {
1152 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' <a href="' . $profile . '">' . $userid . "</a>: <br />»" . $share[3] . "«";
1155 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1158 $headline = '<b>' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . $userid . ':</b><br />';
1160 $text = trim($share[1]);
1166 if (stripos(normalise_link($link), 'http://twitter.com/') === 0) {
1167 $text .= '<br /><a href="' . $link . '">' . $link . '</a>';
1169 $text .= $headline . '<blockquote>' . trim($share[3]) . "</blockquote><br />";
1172 $text .= '<br /><a href="' . $link . '">[l]</a>';
1178 $headline = '<br /><b>' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8');
1179 $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $link, $userid, $posted);
1180 $headline .= ":</b><br />";
1182 $text = trim($share[1]);
1188 $text .= $headline . '<blockquote class="shared_content">' . trim($share[3]) . "</blockquote><br />";
1192 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1195 $text = $preshare . ">> @" . $userid_compact . ": <br />" . $share[3];
1197 case 7: // statusnet/GNU Social
1198 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3];
1201 $text = $preshare . "RT @" . $userid_compact . ": " . $share[3];
1203 case 9: // Google+/Facebook
1204 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1207 $text .= "<br /><br />" . $link;
1211 // Transforms quoted tweets in rich attachments to avoid nested tweets
1212 if (stripos(normalise_link($link), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($link)) {
1214 $oembed = OEmbed::getHTML($link, $preshare);
1215 } catch (Exception $e) {
1216 $oembed = sprintf('[bookmark=%s]%s[/bookmark]', $link, $preshare);
1219 $text = $preshare . $oembed;
1221 $text = trim($share[1]) . "\n";
1223 $avatar = proxy_url($avatar, false, PROXY_SIZE_THUMB);
1225 $tpl = get_markup_template('shared_content.tpl');
1226 $text .= replace_macros($tpl, [
1227 '$profile' => $profile,
1228 '$avatar' => $avatar,
1229 '$author' => $author,
1231 '$posted' => $posted,
1232 '$content' => trim($share[3])
1241 private static function removePictureLinksCallback($match)
1243 $text = Cache::get($match[1]);
1245 if (is_null($text)) {
1248 $stamp1 = microtime(true);
1250 $ch = @curl_init($match[1]);
1251 @curl_setopt($ch, CURLOPT_NOBODY, true);
1252 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1253 @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
1255 $curl_info = @curl_getinfo($ch);
1257 $a->save_timestamp($stamp1, "network");
1259 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1260 $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
1262 $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
1264 // if its not a picture then look if its a page that contains a picture link
1265 $body = Network::fetchUrl($match[1]);
1267 $doc = new DOMDocument();
1268 @$doc->loadHTML($body);
1269 $xpath = new DomXPath($doc);
1270 $list = $xpath->query("//meta[@name]");
1271 foreach ($list as $node) {
1274 if ($node->attributes->length) {
1275 foreach ($node->attributes as $attribute) {
1276 $attr[$attribute->name] = $attribute->value;
1280 if (strtolower($attr["name"]) == "twitter:image") {
1281 $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
1285 Cache::set($match[1], $text);
1291 private static function expandLinksCallback($match)
1293 if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1294 return ($match[1] . "[url]" . $match[2] . "[/url]");
1296 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1300 private static function cleanPictureLinksCallback($match)
1302 $text = Cache::get($match[1]);
1304 if (is_null($text)) {
1307 $stamp1 = microtime(true);
1309 $ch = @curl_init($match[1]);
1310 @curl_setopt($ch, CURLOPT_NOBODY, true);
1311 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1312 @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
1314 $curl_info = @curl_getinfo($ch);
1316 $a->save_timestamp($stamp1, "network");
1318 // if its a link to a picture then embed this picture
1319 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1320 $text = "[img]" . $match[1] . "[/img]";
1322 $text = "[img]" . $match[2] . "[/img]";
1324 // if its not a picture then look if its a page that contains a picture link
1325 $body = Network::fetchUrl($match[1]);
1327 $doc = new DOMDocument();
1328 @$doc->loadHTML($body);
1329 $xpath = new DomXPath($doc);
1330 $list = $xpath->query("//meta[@name]");
1331 foreach ($list as $node) {
1333 if ($node->attributes->length) {
1334 foreach ($node->attributes as $attribute) {
1335 $attr[$attribute->name] = $attribute->value;
1339 if (strtolower($attr["name"]) == "twitter:image") {
1340 $text = "[img]" . $attr["content"] . "[/img]";
1344 Cache::set($match[1], $text);
1350 public static function cleanPictureLinks($text)
1352 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1356 private static function textHighlightCallback($match)
1358 if (in_array(strtolower($match[1]),
1359 ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby',
1360 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh'])
1362 return text_highlight($match[2], strtolower($match[1]));
1368 * @brief Converts a BBCode message to HTML message
1370 * BBcode 2 HTML was written by WAY2WEB.net
1371 * extended to work with Mistpark/Friendica - Mike Macgirvin
1373 * Simple HTML values meaning:
1374 * - 0: Friendica display
1376 * - 2: Used for Facebook, Google+, Windows Phone push, Friendica API
1377 * - 3: Used before converting to Markdown in bb2diaspora.php
1378 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1380 * - 6: Used for Appnet
1381 * - 7: Used for dfrn, OStatus
1382 * - 8: Used for WP backlink text setting
1384 * @param string $text
1385 * @param bool $try_oembed
1386 * @param int $simple_html
1387 * @param bool $for_plaintext
1390 public static function convert($text, $try_oembed = true, $simple_html = false, $for_plaintext = false)
1395 * preg_match_callback function to replace potential Oembed tags with Oembed content
1397 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1399 * $match[2] = $title or absent
1401 $try_oembed_callback = function ($match)
1404 $title = defaults($match, 2, null);
1407 $return = OEmbed::getHTML($url, $title);
1408 } catch (Exception $ex) {
1409 $return = $match[0];
1415 // Hide all [noparse] contained bbtags by spacefying them
1416 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1418 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1419 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1420 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1422 // Remove the abstract element. It is a non visible element.
1423 $text = self::stripAbstract($text);
1425 // Move all spaces out of the tags
1426 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1427 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1429 // Extract the private images which use data urls since preg has issues with
1430 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1431 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1433 $extracted = self::extractImagesFromItemBody($text);
1434 $text = $extracted['body'];
1435 $saved_image = $extracted['images'];
1437 // If we find any event code, turn it into an event.
1438 // After we're finished processing the bbcode we'll
1439 // replace all of the event code with a reformatted version.
1441 $ev = bbtoevent($text);
1443 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1444 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1446 $text = str_replace("<", "<", $text);
1447 $text = str_replace(">", ">", $text);
1449 // remove some newlines before the general conversion
1450 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1451 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1453 $text = preg_replace("/\n\[code\]/ism", "[code]", $text);
1454 $text = preg_replace("/\[\/code\]\n/ism", "[/code]", $text);
1456 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1458 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1461 // Check for [code] text here, before the linefeeds are messed with.
1462 // The highlighter will unescape and re-escape the content.
1463 if (strpos($text, '[code=') !== false) {
1464 $text = preg_replace_callback("/\[code=(.*?)\](.*?)\[\/code\]/ism", 'self::textHighlightCallback', $text);
1466 // Convert new line chars to html <br /> tags
1468 // nlbr seems to be hopelessly messed up
1469 // $Text = nl2br($Text);
1471 // We'll emulate it.
1473 $text = trim($text);
1474 $text = str_replace("\r\n", "\n", $text);
1476 // removing multiplicated newlines
1477 if (Config::get("system", "remove_multiplicated_lines")) {
1478 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1479 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1480 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1481 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1484 $text = str_replace($search, $replace, $text);
1485 } while ($oldtext != $text);
1488 // Set up the parameters for a URL search string
1489 $URLSearchString = "^\[\]";
1490 // Set up the parameters for a MAIL search string
1491 $MAILSearchString = $URLSearchString;
1493 // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text
1494 if (!$for_plaintext) {
1495 // Autolink feature (thanks to http://code.seebz.net/p/autolink-php/)
1496 // Currently disabled, since the function is too greedy
1497 // $autolink_regex = "`([^\]\=\"']|^)(https?\://[^\s<]+[^\s<\.\)])`ism";
1498 $autolink_regex = "/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism";
1499 $text = preg_replace($autolink_regex, '$1[url]$2[/url]', $text);
1500 if ($simple_html == 7) {
1501 $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForMastodonCallback', $text);
1502 $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForMastodonCallback', $text);
1504 $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::shortenVisibleUrlCallback', $text);
1505 $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::shortenVisibleUrlCallback', $text);
1508 $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text);
1509 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1513 // Handle attached links or videos
1514 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1516 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1518 // Remove all hashtag addresses
1519 if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7])) {
1520 $text = preg_replace("/([#@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $text);
1521 } elseif ($simple_html == 3) {
1522 // The ! is converted to @ since Diaspora only understands the @
1523 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1524 '@<a href="$2">$3</a>',
1526 } elseif ($simple_html == 7) {
1527 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1528 '$1<span class="vcard"><a href="$2" class="url" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1530 } elseif (!$simple_html) {
1531 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1532 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1536 // Bookmarks in red - will be converted to bookmarks in friendica
1537 $text = preg_replace("/#\^\[url\]([$URLSearchString]*)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1538 $text = preg_replace("/#\^\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1539 $text = preg_replace("/#\[url\=[$URLSearchString]*\]\^\[\/url\]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i",
1540 "[bookmark=$1]$2[/bookmark]", $text);
1542 if (in_array($simple_html, [2, 6, 7, 8, 9])) {
1543 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1544 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1545 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1548 if ($simple_html == 5) {
1549 $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1552 // Perform URL Search
1554 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1557 if ($simple_html == 5) {
1558 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1560 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1563 // Handle Diaspora posts
1564 $text = preg_replace_callback(
1565 "&\[url=/posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1567 return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1571 // Server independent link to posts and comments
1572 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1573 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1574 $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
1576 $text = preg_replace("/([#])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1577 '$1<a href="' . System::baseUrl() . '/search?tag=$3" class="tag" title="$3">$3</a>', $text);
1579 $text = preg_replace("/\[url\=([$URLSearchString]*)\]#(.*?)\[\/url\]/ism",
1580 '#<a href="' . System::baseUrl() . '/search?tag=$2" class="tag" title="$2">$2</a>', $text);
1582 $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1583 $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1584 //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $Text);
1586 // Red compatibility, though the link can't be authenticated on Friendica
1587 $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1590 // we may need to restrict this further if it picks up too many strays
1591 // link acct:user@host to a webfinger profile redirector
1593 $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);
1595 // Perform MAIL Search
1596 $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1597 $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1599 // leave open the posibility of [map=something]
1600 // this is replaced in prepare_body() which has knowledge of the item location
1602 if (strpos($text, '[/map]') !== false) {
1603 $text = preg_replace_callback(
1604 "/\[map\](.*?)\[\/map\]/ism",
1606 // the extra space in the following line is intentional
1607 // Whyyy? - @MrPetovan
1608 return str_replace($match[0], '<div class="map" >' . Map::byLocation($match[1]) . '</div>', $match[0]);
1613 if (strpos($text, '[map=') !== false) {
1614 $text = preg_replace_callback(
1615 "/\[map=(.*?)\]/ism",
1617 // the extra space in the following line is intentional
1618 // Whyyy? - @MrPetovan
1619 return str_replace($match[0], '<div class="map" >' . Map::byCoordinates(str_replace('/', ' ', $match[1])) . '</div>', $match[0]);
1624 if (strpos($text, '[map]') !== false) {
1625 $text = preg_replace("/\[map\]/", '<div class="map"></div>', $text);
1628 // Check for headers
1629 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1630 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1631 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1632 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1633 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1634 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1636 // Check for paragraph
1637 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1639 // Check for bold text
1640 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1642 // Check for Italics text
1643 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1645 // Check for Underline text
1646 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1648 // Check for strike-through text
1649 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<strike>$1</strike>', $text);
1651 // Check for over-line text
1652 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1654 // Check for colored text
1655 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1657 // Check for sized text
1658 // [size=50] --> font-size: 50px (with the unit).
1659 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1660 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1662 // Check for centered text
1663 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1665 // Check for list text
1666 $text = str_replace("[*]", "<li>", $text);
1668 // Check for style sheet commands
1669 $text = preg_replace_callback(
1670 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1672 return "<span style=\"" . self::cleanCss($match[1]) . ";\">" . $match[2] . "</span>";
1677 // Check for CSS classes
1678 $text = preg_replace_callback(
1679 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1681 return "<span class=\"" . self::cleanCss($match[1]) . "\">" . $match[2] . "</span>";
1686 // handle nested lists
1689 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1690 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1691 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1692 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1693 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1694 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1695 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1696 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1697 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1698 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1699 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1700 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1701 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1702 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1705 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1706 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1707 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1708 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1710 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1711 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1713 $text = str_replace('[hr]', '<hr />', $text);
1715 // This is actually executed in prepare_body()
1717 $text = str_replace('[nosmile]', '', $text);
1719 // Check for font change text
1720 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1722 // Declare the format for [code] layout
1724 $CodeLayout = '<code>$1</code>';
1725 // Check for [code] text
1726 $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text);
1728 // Declare the format for [spoiler] layout
1729 $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
1731 // Check for [spoiler] text
1732 // handle nested quotes
1734 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1735 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
1738 // Check for [spoiler=Author] text
1740 $t_wrote = L10n::t('$1 wrote:');
1742 // handle nested quotes
1744 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1745 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1746 "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
1750 // Declare the format for [quote] layout
1751 $QuoteLayout = '<blockquote>$1</blockquote>';
1753 // Check for [quote] text
1754 // handle nested quotes
1756 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1757 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1760 // Check for [quote=Author] text
1762 $t_wrote = L10n::t('$1 wrote:');
1764 // handle nested quotes
1766 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1767 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1768 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1773 // [img=widthxheight]image source[/img]
1774 $text = preg_replace_callback(
1775 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1776 function ($matches) {
1777 if (strpos($matches[3], "data:image/") === 0) {
1781 $matches[3] = proxy_url($matches[3]);
1782 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1787 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1788 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1791 // [img]pathtoimage[/img]
1792 $text = preg_replace_callback(
1793 "/\[img\](.*?)\[\/img\]/ism",
1794 function ($matches) {
1795 if (strpos($matches[1], "data:image/") === 0) {
1799 $matches[1] = proxy_url($matches[1]);
1800 return "[img]" . $matches[1] . "[/img]";
1805 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1806 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1809 $text = preg_replace_callback("/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
1810 function ($match) use ($simple_html) {
1811 return self::convertShare($match, $simple_html);
1814 $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);
1815 $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 />', $text);
1816 //$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);
1820 $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);
1821 $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);
1823 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1824 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1826 $text = preg_replace("/\[video\](.*?)\[\/video\]/",
1827 '<a href="$1" target="_blank">$1</a>', $text);
1828 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/",
1829 '<a href="$1" target="_blank">$1</a>', $text);
1832 // html5 video and audio
1836 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1838 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1841 // Youtube extensions
1843 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1844 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1845 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1848 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1849 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1850 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1853 $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);
1855 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1856 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1860 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1861 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1864 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1865 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1868 $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);
1870 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1871 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1875 $text = OEmbed::BBCode2HTML($text);
1877 // Avoid triple linefeeds through oembed
1878 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1880 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1881 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1882 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1883 // start which is always required). Allow desc with a missing summary for compatibility.
1885 if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) {
1886 $sub = format_event_html($ev, $simple_html);
1888 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1889 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1890 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1891 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1892 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1893 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1894 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1897 // Replace non graphical smilies for external posts
1899 $text = Smilies::replace($text, false, true);
1902 // Replace inline code blocks
1903 $text = preg_replace_callback("|(?!<br[^>]*>)<code>([^<]*)</code>(?!<br[^>]*>)|ism",
1904 function ($match) use ($simple_html) {
1905 $return = '<key>' . $match[1] . '</key>';
1906 // Use <code> for Diaspora inline code blocks
1907 if ($simple_html === 3) {
1908 $return = '<code>' . $match[1] . '</code>';
1914 // Unhide all [noparse] contained bbtags unspacefying them
1915 // and triming the [noparse] tag.
1917 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1918 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1919 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1922 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1923 $text = preg_replace('/\&\#039\;/', '\'', $text);
1924 $text = preg_replace('/\"\;/', '"', $text);
1926 // fix any escaped ampersands that may have been converted into links
1927 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1929 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1930 $allowed_src_protocols = ['http', 'redir', 'cid'];
1931 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1932 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
1934 // sanitize href attributes (only whitelisted protocols URLs)
1935 // default value for backward compatibility
1936 $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']);
1938 // Always allowed protocol even if config isn't set or not including it
1939 $allowed_link_protocols[] = 'http';
1940 $allowed_link_protocols[] = 'redir/';
1942 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1943 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
1946 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1949 // Clean up the HTML by loading and saving the HTML with the DOM.
1950 // Bad structured html can break a whole page.
1951 // For performance reasons do it only with ativated item cache or at export.
1952 if (!$try_oembed || (get_itemcachepath() != "")) {
1953 $doc = new DOMDocument();
1954 $doc->preserveWhiteSpace = false;
1956 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1958 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1959 $encoding = '<?xml encoding="UTF-8">';
1960 @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
1961 $doc->encoding = 'UTF-8';
1962 $text = $doc->saveHTML();
1963 $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
1965 $text = str_replace('<br></li>', '</li>', $text);
1967 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1970 // Clean up some useless linebreaks in lists
1971 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1972 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1973 //$Text = str_replace('</li><br />', '</li>', $Text);
1974 //$Text = str_replace('<br /><li>', '<li>', $Text);
1975 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1977 Addon::callHooks('bbcode', $text);
1983 * @brief Strips the "abstract" tag from the provided text
1985 * @param string $text The text with BBCode
1986 * @return string The same text - but without "abstract" element
1988 public static function stripAbstract($text)
1990 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1991 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1997 * @brief Returns the value of the "abstract" element
1999 * @param string $text The text that maybe contains the element
2000 * @param string $addon The addon for which the abstract is meant for
2001 * @return string The abstract
2003 private static function getAbstract($text, $addon = "")
2007 $addon = strtolower($addon);
2009 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
2010 foreach ($results AS $result) {
2011 $abstracts[strtolower($result[1])] = $result[2];
2015 if (isset($abstracts[$addon])) {
2016 $abstract = $abstracts[$addon];
2019 if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
2020 $abstract = $result[1];