4 * @file src/Content/Text/BBCode.php
7 namespace Friendica\Content\Text;
12 use Friendica\BaseObject;
13 use Friendica\Content\OEmbed;
14 use Friendica\Content\Smilies;
15 use Friendica\Core\Addon;
16 use Friendica\Core\Cache;
17 use Friendica\Core\Config;
18 use Friendica\Core\L10n;
19 use Friendica\Core\PConfig;
20 use Friendica\Core\Protocol;
21 use Friendica\Core\System;
22 use Friendica\Model\Contact;
23 use Friendica\Network\Probe;
24 use Friendica\Object\Image;
25 use Friendica\Util\Map;
26 use Friendica\Util\Network;
27 use Friendica\Util\ParseUrl;
28 use League\HTMLToMarkdown\HtmlConverter;
30 require_once "include/event.php";
31 require_once "mod/proxy.php";
33 class BBCode extends BaseObject
36 * @brief Fetches attachment data that were generated the old way
38 * @param string $body Message body
40 * 'type' -> Message type ("link", "video", "photo")
41 * 'text' -> Text before the shared message
42 * 'after' -> Text after the shared message
43 * 'image' -> Preview image of the message
44 * 'url' -> Url to the attached message
45 * 'title' -> Title of the attachment
46 * 'description' -> Description of the attachment
48 private static function getOldAttachmentData($body)
52 // Simplify image codes
53 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
55 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
56 foreach ($attached as $data) {
57 if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
61 $post["type"] = substr($data[1], 5);
63 $pos = strpos($body, $data[0]);
65 $post["text"] = trim(substr($body, 0, $pos));
66 $post["after"] = trim(substr($body, $pos + strlen($data[0])));
68 $post["text"] = trim(str_replace($data[0], "", $body));
71 $attacheddata = $data[2];
73 $URLSearchString = "^\[\]";
75 if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
77 $picturedata = Image::getInfoFromURL($matches[1]);
79 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
80 $post["image"] = $matches[1];
82 $post["preview"] = $matches[1];
86 if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
87 $post["url"] = $matches[1];
88 $post["title"] = $matches[2];
90 if (($post["url"] == "") && (in_array($post["type"], ["link", "video"]))
91 && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
92 $post["url"] = $matches[1];
95 // Search for description
96 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
97 $post["description"] = $matches[1];
105 * @brief Fetches attachment data that were generated with the "attachment" element
107 * @param string $body Message body
109 * 'type' -> Message type ("link", "video", "photo")
110 * 'text' -> Text before the shared message
111 * 'after' -> Text after the shared message
112 * 'image' -> Preview image of the message
113 * 'url' -> Url to the attached message
114 * 'title' -> Title of the attachment
115 * 'description' -> Description of the attachment
117 public static function getAttachmentData($body)
121 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
122 return self::getOldAttachmentData($body);
125 $attributes = $match[2];
127 $data["text"] = trim($match[1]);
130 preg_match("/type='(.*?)'/ism", $attributes, $matches);
131 if (x($matches, 1)) {
132 $type = strtolower($matches[1]);
135 preg_match('/type="(.*?)"/ism', $attributes, $matches);
136 if (x($matches, 1)) {
137 $type = strtolower($matches[1]);
144 if (!in_array($type, ["link", "audio", "photo", "video"])) {
149 $data["type"] = $type;
153 preg_match("/url='(.*?)'/ism", $attributes, $matches);
154 if (x($matches, 1)) {
158 preg_match('/url="(.*?)"/ism', $attributes, $matches);
159 if (x($matches, 1)) {
164 $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
168 preg_match("/title='(.*?)'/ism", $attributes, $matches);
169 if (x($matches, 1)) {
170 $title = $matches[1];
173 preg_match('/title="(.*?)"/ism', $attributes, $matches);
174 if (x($matches, 1)) {
175 $title = $matches[1];
179 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
180 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
181 $title = str_replace(["[", "]"], ["[", "]"], $title);
182 $data["title"] = $title;
186 preg_match("/image='(.*?)'/ism", $attributes, $matches);
187 if (x($matches, 1)) {
188 $image = $matches[1];
191 preg_match('/image="(.*?)"/ism', $attributes, $matches);
192 if (x($matches, 1)) {
193 $image = $matches[1];
197 $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
201 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
202 if (x($matches, 1)) {
203 $preview = $matches[1];
206 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
207 if (x($matches, 1)) {
208 $preview = $matches[1];
211 if ($preview != "") {
212 $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
215 $data["description"] = trim($match[3]);
217 $data["after"] = trim($match[4]);
222 public static function getAttachedData($body, $item = [])
226 - type: link, video, photo
234 $has_title = !empty($item['title']);
235 $plink = (!empty($item['plink']) ? $item['plink'] : '');
236 $post = self::getAttachmentData($body);
238 // if nothing is found, it maybe having an image.
239 if (!isset($post["type"])) {
240 // Simplify image codes
241 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
243 $URLSearchString = "^\[\]";
244 if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
245 if ((count($pictures) == 1) && !$has_title) {
246 // Checking, if the link goes to a picture
247 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
250 // Sometimes photo posts to the own album are not detected at the start.
251 // So we seem to cannot use the cache for these cases. That's strange.
252 if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
253 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
256 if ($data["type"] == "photo") {
257 $post["type"] = "photo";
258 if (isset($data["images"][0])) {
259 $post["image"] = $data["images"][0]["src"];
260 $post["url"] = $data["url"];
262 $post["image"] = $data["url"];
265 $post["preview"] = $pictures[0][2];
266 $post["text"] = str_replace($pictures[0][0], "", $body);
268 $imgdata = Image::getInfoFromURL($pictures[0][1]);
269 if (substr($imgdata["mime"], 0, 6) == "image/") {
270 $post["type"] = "photo";
271 $post["image"] = $pictures[0][1];
272 $post["preview"] = $pictures[0][2];
273 $post["text"] = str_replace($pictures[0][0], "", $body);
276 } elseif (count($pictures) > 0) {
277 $post["type"] = "link";
278 $post["url"] = $plink;
279 $post["image"] = $pictures[0][2];
280 $post["text"] = $body;
282 } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
283 if ((count($pictures) == 1) && !$has_title) {
284 $post["type"] = "photo";
285 $post["image"] = $pictures[0][1];
286 $post["text"] = str_replace($pictures[0][0], "", $body);
287 } elseif (count($pictures) > 0) {
288 $post["type"] = "link";
289 $post["url"] = $plink;
290 $post["image"] = $pictures[0][1];
291 $post["text"] = $body;
295 // Test for the external links
296 preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
297 preg_match_all("(\[url\=([$URLSearchString]*)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
299 $links = array_merge($links1, $links2);
301 // If there is only a single one, then use it.
302 // This should cover link posts via API.
303 if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
304 $post["type"] = "link";
305 $post["text"] = trim($body);
306 $post["url"] = $links[0][1];
309 // Now count the number of external media links
310 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
311 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
312 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
313 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
315 // Add them to the other external links
316 $links = array_merge($links, $links1, $links2, $links3, $links4);
318 // Are there more than one?
319 if (count($links) > 1) {
320 // The post will be the type "text", which means a blog post
321 unset($post["type"]);
322 $post["url"] = $plink;
325 if (!isset($post["type"])) {
326 $post["type"] = "text";
327 $post["text"] = trim($body);
329 } elseif (isset($post["url"]) && ($post["type"] == "video")) {
330 $data = ParseUrl::getSiteinfoCached($post["url"], true);
332 if (isset($data["images"][0])) {
333 $post["image"] = $data["images"][0]["src"];
341 * @brief Convert a message into plaintext for connectors to other networks
343 * @param array $b The message array that is about to be posted
344 * @param int $limit The maximum number of characters when posting to that network
345 * @param bool $includedlinks Has an attached link to be included into the message?
346 * @param int $htmlmode This triggers the behaviour of the bbcode conversion
347 * @param string $target_network Name of the network where the post should go to.
349 * @return string The converted message
351 public static function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "")
353 // Remove the hash tags
354 $URLSearchString = "^\[\]";
355 $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]);
357 // Add an URL element if the text contains a raw link
358 $body = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2[/url]', $body);
360 // Remove the abstract
361 $body = self::stripAbstract($body);
363 // At first look at data that is attached via "type-..." stuff
364 // This will hopefully replaced with a dedicated bbcode later
365 //$post = self::getAttachedData($b["body"]);
366 $post = self::getAttachedData($body, $b);
368 if (($b["title"] != "") && ($post["text"] != "")) {
369 $post["text"] = trim($b["title"]."\n\n".$post["text"]);
370 } elseif ($b["title"] != "") {
371 $post["text"] = trim($b["title"]);
376 // Fetch the abstract from the given target network
377 if ($target_network != "") {
378 $default_abstract = self::getAbstract($b["body"]);
379 $abstract = self::getAbstract($b["body"], $target_network);
381 // If we post to a network with no limit we only fetch
382 // an abstract exactly for this network
383 if (($limit == 0) && ($abstract == $default_abstract)) {
386 } else {// Try to guess the correct target network
389 $abstract = self::getAbstract($b["body"], NETWORK_TWITTER);
392 $abstract = self::getAbstract($b["body"], NETWORK_STATUSNET);
395 $abstract = self::getAbstract($b["body"], NETWORK_APPNET);
397 default: // We don't know the exact target.
398 // We fetch an abstract since there is a posting limit.
400 $abstract = self::getAbstract($b["body"]);
405 if ($abstract != "") {
406 $post["text"] = $abstract;
408 if ($post["type"] == "text") {
409 $post["type"] = "link";
410 $post["url"] = $b["plink"];
414 $html = self::convert($post["text"].$post["after"], false, $htmlmode);
415 $msg = HTML::toPlaintext($html, 0, true);
416 $msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8'));
419 if ($includedlinks) {
420 if ($post["type"] == "link") {
421 $link = $post["url"];
422 } elseif ($post["type"] == "text") {
423 $link = $post["url"];
424 } elseif ($post["type"] == "video") {
425 $link = $post["url"];
426 } elseif ($post["type"] == "photo") {
427 $link = $post["image"];
430 if (($msg == "") && isset($post["title"])) {
431 $msg = trim($post["title"]);
434 if (($msg == "") && isset($post["description"])) {
435 $msg = trim($post["description"]);
438 // If the link is already contained in the post, then it neeedn't to be added again
439 // But: if the link is beyond the limit, then it has to be added.
440 if (($link != "") && strstr($msg, $link)) {
441 $pos = strpos($msg, $link);
443 // Will the text be shortened in the link?
444 // Or is the link the last item in the post?
445 if (($limit > 0) && ($pos < $limit) && (($pos + 23 > $limit) || ($pos + strlen($link) == strlen($msg)))) {
446 $msg = trim(str_replace($link, "", $msg));
447 } elseif (($limit == 0) || ($pos < $limit)) {
448 // The limit has to be increased since it will be shortened - but not now
449 // Only do it with Twitter (htmlmode = 8)
450 if (($limit > 0) && (strlen($link) > 23) && ($htmlmode == 8)) {
451 $limit = $limit - 23 + strlen($link);
456 if ($post["type"] == "text") {
464 // Reduce multiple spaces
465 // When posted to a network with limited space, we try to gain space where possible
466 while (strpos($msg, " ") !== false) {
467 $msg = str_replace(" ", " ", $msg);
470 // Twitter is using its own limiter, so we always assume that shortened links will have this length
471 if (iconv_strlen($link, "UTF-8") > 0) {
472 $limit = $limit - 23;
475 if (iconv_strlen($msg, "UTF-8") > $limit) {
476 if (($post["type"] == "text") && isset($post["url"])) {
477 $post["url"] = $b["plink"];
478 } elseif (!isset($post["url"])) {
479 $limit = $limit - 23;
480 $post["url"] = $b["plink"];
481 // Which purpose has this line? It is now uncommented, but left as a reminder
482 //} elseif (strpos($b["body"], "[share") !== false) {
483 // $post["url"] = $b["plink"];
484 } elseif (PConfig::get($b["uid"], "system", "no_intelligent_shortening")) {
485 $post["url"] = $b["plink"];
487 $msg = Plaintext::shorten($msg, $limit);
491 $post["text"] = trim($msg);
496 public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
498 // Suppress "view full size"
499 if (intval(Config::get('system', 'no_view_full_size'))) {
500 $include_link = false;
503 // Picture addresses can contain special characters
504 $s = htmlspecialchars_decode($srctext);
507 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
509 foreach ($matches as $mtch) {
510 logger('scale_external_image: ' . $mtch[1]);
512 $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
513 if (stristr($mtch[1], $hostname)) {
517 // $scale_replace, if passed, is an array of two elements. The
518 // first is the name of the full-size image. The second is the
519 // name of a remote, scaled-down version of the full size image.
520 // This allows Friendica to display the smaller remote image if
521 // one exists, while still linking to the full-size image
522 if ($scale_replace) {
523 $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
527 $i = Network::fetchUrl($scaled);
532 // guess mimetype from headers or filename
533 $type = Image::guessType($mtch[1], true);
536 $Image = new Image($i, $type);
537 if ($Image->isValid()) {
538 $orig_width = $Image->getWidth();
539 $orig_height = $Image->getHeight();
541 if ($orig_width > 640 || $orig_height > 640) {
542 $Image->scaleDown(640);
543 $new_width = $Image->getWidth();
544 $new_height = $Image->getHeight();
545 logger('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], LOGGER_DEBUG);
548 '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
549 . "\n" . (($include_link)
550 ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
554 logger('scale_external_images: new string: ' . $s, LOGGER_DEBUG);
561 // replace the special char encoding
562 $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
567 * The purpose of this function is to apply system message length limits to
568 * imported messages without including any embedded photos in the length
570 * @brief Truncates imported message body string length to max_import_size
571 * @param string $body
574 public static function limitBodySize($body)
576 $maxlen = get_max_import_size();
578 // If the length of the body, including the embedded images, is smaller
579 // than the maximum, then don't waste time looking for the images
580 if ($maxlen && (strlen($body) > $maxlen)) {
582 logger('the total body length exceeds the limit', LOGGER_DEBUG);
588 $img_start = strpos($orig_body, '[img');
589 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
590 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
591 while (($img_st_close !== false) && ($img_end !== false)) {
593 $img_st_close++; // make it point to AFTER the closing bracket
594 $img_end += $img_start;
595 $img_end += strlen('[/img]');
597 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
598 // This is an embedded image
600 if (($textlen + $img_start) > $maxlen) {
601 if ($textlen < $maxlen) {
602 logger('the limit happens before an embedded image', LOGGER_DEBUG);
603 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
607 $new_body = $new_body . substr($orig_body, 0, $img_start);
608 $textlen += $img_start;
611 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
614 if (($textlen + $img_end) > $maxlen) {
615 if ($textlen < $maxlen) {
616 logger('the limit happens before the end of a non-embedded image', LOGGER_DEBUG);
617 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
621 $new_body = $new_body . substr($orig_body, 0, $img_end);
622 $textlen += $img_end;
625 $orig_body = substr($orig_body, $img_end);
627 if ($orig_body === false) {
628 // in case the body ends on a closing image tag
632 $img_start = strpos($orig_body, '[img');
633 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
634 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
637 if (($textlen + strlen($orig_body)) > $maxlen) {
638 if ($textlen < $maxlen) {
639 logger('the limit happens after the end of the last image', LOGGER_DEBUG);
640 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
643 logger('the text size with embedded images extracted did not violate the limit', LOGGER_DEBUG);
644 $new_body = $new_body . $orig_body;
654 * Processes [attachment] tags
656 * Note: Can produce a [bookmark] tag in the returned string
658 * @brief Processes [attachment] tags
659 * @param string $return
660 * @param bool|int $simplehtml
661 * @param bool $tryoembed
664 private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
666 $data = self::getAttachmentData($return);
671 if (isset($data["title"])) {
672 $data["title"] = strip_tags($data["title"]);
673 $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
676 if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) {
677 $data["preview"] = $data["image"];
682 if ($simplehtml == 7) {
683 $return = self::convertUrlForMastodon($data["url"]);
684 } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
685 $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data["url"], $data["title"]);
688 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
689 $return = OEmbed::getHTML($data['url'], $data['title']);
691 throw new Exception('OEmbed is disabled for this attachment.');
693 } catch (Exception $e) {
694 if ($simplehtml != 4) {
695 $return = sprintf('<div class="type-%s">', $data["type"]);
698 if ($data["image"] != "") {
699 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], proxy_url($data["image"]), $data["title"]);
700 } elseif ($data["preview"] != "") {
701 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], proxy_url($data["preview"]), $data["title"]);
704 if (($data["type"] == "photo") && ($data["url"] != "") && ($data["image"] != "")) {
705 $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"]);
707 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
710 if ($data["description"] != "" && $data["description"] != $data["title"]) {
711 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($data["description"])));
714 if ($data["type"] == "link") {
715 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
718 if ($simplehtml != 4) {
724 return trim($data["text"] . ' ' . $return . ' ' . $data["after"]);
727 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
729 $data = self::getAttachmentData($Text);
734 return $data["text"] . $data["after"];
737 $title = htmlentities($data["title"], ENT_QUOTES, 'UTF-8', false);
738 $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
739 if ($plaintext || (($title != "") && strstr($text, $title))) {
740 $data["title"] = $data["url"];
741 } elseif (($text != "") && strstr($title, $text)) {
742 $data["text"] = $data["title"];
743 $data["title"] = $data["url"];
746 if (($data["text"] == "") && ($data["title"] != "") && ($data["url"] == "")) {
747 return $data["title"] . $data["after"];
750 // If the link already is included in the post, don't add it again
751 if (($data["url"] != "") && strpos($data["text"], $data["url"])) {
752 return $data["text"] . $data["after"];
755 $text = $data["text"];
757 if (($data["url"] != "") && ($data["title"] != "")) {
758 $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
759 } elseif (($data["url"] != "")) {
760 $text .= "\n" . $data["url"];
763 return $text . "\n" . $data["after"];
767 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
769 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
770 * @param array $match Array with the matching values
771 * @return string reformatted link including HTML codes
773 private static function convertUrlForMastodonCallback($match)
777 if (isset($match[2]) && ($match[1] != $match[2])) {
781 $parts = parse_url($url);
782 if (!isset($parts['scheme'])) {
786 return self::convertUrlForMastodon($url);
790 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon and GNU Social.
791 * @param string $url URL that is about to be reformatted
792 * @return string reformatted link including HTML codes
794 private static function convertUrlForMastodon($url)
796 $parts = parse_url($url);
797 $scheme = $parts['scheme'] . '://';
798 $styled_url = str_replace($scheme, '', $url);
800 $html = '<a href="%s" class="attachment" rel="nofollow noopener" target="_blank">' .
801 '<span class="invisible">%s</span>';
803 if (strlen($styled_url) > 30) {
804 $html .= '<span class="ellipsis">%s</span>' .
805 '<span class="invisible">%s</span></a>';
807 $ellipsis = substr($styled_url, 0, 30);
808 $rest = substr($styled_url, 30);
809 return sprintf($html, $url, $scheme, $ellipsis, $rest);
812 return sprintf($html, $url, $scheme, $styled_url);
817 * [noparse][i]italic[/i][/noparse] turns into
818 * [noparse][ i ]italic[ /i ][/noparse],
819 * to hide them from parser.
821 private static function escapeNoparseCallback($match)
823 $whole_match = $match[0];
824 $captured = $match[1];
825 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
826 $new_str = str_replace($captured, $spacefied, $whole_match);
831 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
832 * now turns back and the [noparse] tags are trimed
833 * returning [i]italic[/i]
835 private static function unescapeNoparseCallback($match)
837 $captured = $match[1];
838 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
843 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
846 * @param string $text Text to search
847 * @param string $name Tag name
848 * @param int $occurrences Number of first occurrences to skip
849 * @return boolean|array
851 public static function getTagPosition($text, $name, $occurrences = 0)
853 if ($occurrences < 0) {
858 for ($i = 0; $i <= $occurrences; $i++) {
859 if ($start_open !== false) {
860 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
864 if ($start_open === false) {
868 $start_equal = strpos($text, '=', $start_open);
869 $start_close = strpos($text, ']', $start_open);
871 if ($start_close === false) {
877 $end_open = strpos($text, '[/' . $name . ']', $start_close);
879 if ($end_open === false) {
885 'open' => $start_open,
886 'close' => $start_close
890 'close' => $end_open + strlen('[/' . $name . ']')
894 if ($start_equal !== false) {
895 $res['start']['equal'] = $start_equal + 1;
902 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
904 * @param type $pattern Preg pattern string
905 * @param type $replace Preg replace string
906 * @param type $name BBCode tag name
907 * @param type $text Text to search
910 public static function pregReplaceInTag($pattern, $replace, $name, $text)
913 $pos = self::getTagPosition($text, $name, $occurrences);
914 while ($pos !== false && $occurrences++ < 1000) {
915 $start = substr($text, 0, $pos['start']['open']);
916 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
917 $end = substr($text, $pos['end']['close']);
918 if ($end === false) {
922 $subject = preg_replace($pattern, $replace, $subject);
923 $text = $start . $subject . $end;
925 $pos = self::getTagPosition($text, $name, $occurrences);
931 private static function extractImagesFromItemBody($body)
938 $img_start = strpos($orig_body, '[img');
939 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
940 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
941 while (($img_st_close !== false) && ($img_end !== false)) {
942 $img_st_close++; // make it point to AFTER the closing bracket
943 $img_end += $img_start;
945 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
946 // This is an embedded image
947 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
948 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
952 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
955 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
957 if ($orig_body === false) {
958 // in case the body ends on a closing image tag
962 $img_start = strpos($orig_body, '[img');
963 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
964 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
967 $new_body = $new_body . $orig_body;
969 return ['body' => $new_body, 'images' => $saved_image];
972 private static function interpolateSavedImagesIntoItemBody($body, array $images)
977 foreach ($images as $image) {
978 // We're depending on the property of 'foreach' (specified on the PHP website) that
979 // it loops over the array starting from the first element and going sequentially
980 // to the last element
981 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
982 '<img src="' . proxy_url($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
990 * Processes [share] tags
992 * Note: Can produce a [bookmark] tag in the output
994 * @brief Processes [share] tags
995 * @param array $share preg_match_callback result array
996 * @param bool|int $simplehtml
999 private static function convertShare($share, $simplehtml)
1001 $attributes = $share[2];
1004 preg_match("/author='(.*?)'/ism", $attributes, $matches);
1005 if (x($matches, 1)) {
1006 $author = html_entity_decode($matches[1], ENT_QUOTES, 'UTF-8');
1009 preg_match('/author="(.*?)"/ism', $attributes, $matches);
1010 if (x($matches, 1)) {
1011 $author = $matches[1];
1015 preg_match("/profile='(.*?)'/ism", $attributes, $matches);
1016 if (x($matches, 1)) {
1017 $profile = $matches[1];
1020 preg_match('/profile="(.*?)"/ism', $attributes, $matches);
1021 if (x($matches, 1)) {
1022 $profile = $matches[1];
1026 preg_match("/avatar='(.*?)'/ism", $attributes, $matches);
1027 if (x($matches, 1)) {
1028 $avatar = $matches[1];
1031 preg_match('/avatar="(.*?)"/ism', $attributes, $matches);
1032 if (x($matches, 1)) {
1033 $avatar = $matches[1];
1037 preg_match("/link='(.*?)'/ism", $attributes, $matches);
1038 if (x($matches, 1)) {
1039 $link = $matches[1];
1042 preg_match('/link="(.*?)"/ism', $attributes, $matches);
1043 if (x($matches, 1)) {
1044 $link = $matches[1];
1049 preg_match("/posted='(.*?)'/ism", $attributes, $matches);
1050 if (x($matches, 1)) {
1051 $posted = $matches[1];
1054 preg_match('/posted="(.*?)"/ism', $attributes, $matches);
1055 if (x($matches, 1)) {
1056 $posted = $matches[1];
1059 // We only call this so that a previously unknown contact can be added.
1060 // This is important for the function "Model\Contact::getDetailsByURL()".
1061 // This function then can fetch an entry from the contact table.
1062 Contact::getIdForURL($profile, 0, true);
1064 $data = Contact::getDetailsByURL($profile);
1066 if (x($data, "name") && x($data, "addr")) {
1067 $userid_compact = $data["name"] . " (" . $data["addr"] . ")";
1069 $userid_compact = Protocol::getAddrFromProfileUrl($profile, $author);
1072 if (x($data, "addr")) {
1073 $userid = $data["addr"];
1075 $userid = Protocol::formatMention($profile, $author);
1078 if (x($data, "name")) {
1079 $author = $data["name"];
1082 if (x($data, "micro")) {
1083 $avatar = $data["micro"];
1086 $preshare = trim($share[1]);
1087 if ($preshare != "") {
1088 $preshare .= "<br />";
1091 switch ($simplehtml) {
1093 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' <a href="' . $profile . '">' . $userid . "</a>: <br />»" . $share[3] . "«";
1096 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1099 $headline = '<b>' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . $userid . ':</b><br />';
1101 $text = trim($share[1]);
1107 if (stripos(normalise_link($link), 'http://twitter.com/') === 0) {
1108 $text .= $headline . '<blockquote>' . trim($share[3]) . "</blockquote><br />";
1111 $text .= '<br /><a href="' . $link . '">[l]</a>';
1114 $text .= '<br /><a href="' . $link . '">' . $link . '</a>';
1119 $headline = '<br /><b>' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8');
1120 $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $link, $userid, $posted);
1121 $headline .= ":</b><br />";
1123 $text = trim($share[1]);
1129 $text .= $headline . '<blockquote class="shared_content">' . trim($share[3]) . "</blockquote><br />";
1133 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1136 $text = $preshare . ">> @" . $userid_compact . ": <br />" . $share[3];
1138 case 7: // statusnet/GNU Social
1139 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3];
1142 $text = $preshare . "RT @" . $userid_compact . ": " . $share[3];
1144 case 9: // Google+/Facebook
1145 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1148 $text .= "<br /><br />" . $link;
1152 // Transforms quoted tweets in rich attachments to avoid nested tweets
1153 if (stripos(normalise_link($link), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($link)) {
1155 $oembed = OEmbed::getHTML($link, $preshare);
1156 } catch (Exception $e) {
1157 $oembed = sprintf('[bookmark=%s]%s[/bookmark]', $link, $preshare);
1160 $text = $preshare . $oembed;
1162 $text = trim($share[1]) . "\n";
1164 $avatar = proxy_url($avatar, false, PROXY_SIZE_THUMB);
1166 $tpl = get_markup_template('shared_content.tpl');
1167 $text .= replace_macros($tpl, [
1168 '$profile' => $profile,
1169 '$avatar' => $avatar,
1170 '$author' => $author,
1172 '$posted' => $posted,
1173 '$content' => trim($share[3])
1182 private static function removePictureLinksCallback($match)
1184 $text = Cache::get($match[1]);
1186 if (is_null($text)) {
1187 $a = self::getApp();
1189 $stamp1 = microtime(true);
1191 $ch = @curl_init($match[1]);
1192 @curl_setopt($ch, CURLOPT_NOBODY, true);
1193 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1194 @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
1196 $curl_info = @curl_getinfo($ch);
1198 $a->save_timestamp($stamp1, "network");
1200 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1201 $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
1203 $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
1205 // if its not a picture then look if its a page that contains a picture link
1206 $body = Network::fetchUrl($match[1]);
1208 $doc = new DOMDocument();
1209 @$doc->loadHTML($body);
1210 $xpath = new DomXPath($doc);
1211 $list = $xpath->query("//meta[@name]");
1212 foreach ($list as $node) {
1215 if ($node->attributes->length) {
1216 foreach ($node->attributes as $attribute) {
1217 $attr[$attribute->name] = $attribute->value;
1221 if (strtolower($attr["name"]) == "twitter:image") {
1222 $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
1226 Cache::set($match[1], $text);
1232 private static function expandLinksCallback($match)
1234 if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1235 return ($match[1] . "[url]" . $match[2] . "[/url]");
1237 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1241 private static function cleanPictureLinksCallback($match)
1243 $text = Cache::get($match[1]);
1245 if (is_null($text)) {
1246 $a = self::getApp();
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 its a link to a picture then embed this picture
1260 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1261 $text = "[img]" . $match[1] . "[/img]";
1263 $text = "[img]" . $match[2] . "[/img]";
1265 // if its not a picture then look if its a page that contains a picture link
1266 $body = Network::fetchUrl($match[1]);
1268 $doc = new DOMDocument();
1269 @$doc->loadHTML($body);
1270 $xpath = new DOMXPath($doc);
1271 $list = $xpath->query("//meta[@name]");
1272 foreach ($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 = "[img]" . $attr["content"] . "[/img]";
1285 Cache::set($match[1], $text);
1291 public static function cleanPictureLinks($text)
1293 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1297 private static function textHighlightCallback($match)
1299 if (in_array(strtolower($match[1]),
1300 ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby',
1301 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh'])
1303 return text_highlight($match[2], strtolower($match[1]));
1309 * @brief Converts a BBCode message to HTML message
1311 * BBcode 2 HTML was written by WAY2WEB.net
1312 * extended to work with Mistpark/Friendica - Mike Macgirvin
1314 * Simple HTML values meaning:
1315 * - 0: Friendica display
1317 * - 2: Used for Facebook, Google+, Windows Phone push, Friendica API
1318 * - 3: Used before converting to Markdown in bb2diaspora.php
1319 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1321 * - 6: Used for Appnet
1322 * - 7: Used for dfrn, OStatus
1323 * - 8: Used for WP backlink text setting
1325 * @param string $text
1326 * @param bool $try_oembed
1327 * @param int $simple_html
1328 * @param bool $for_plaintext
1331 public static function convert($text, $try_oembed = true, $simple_html = false, $for_plaintext = false)
1333 $a = self::getApp();
1336 * preg_match_callback function to replace potential Oembed tags with Oembed content
1338 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1340 * $match[2] = $title or absent
1342 $try_oembed_callback = function ($match)
1345 $title = defaults($match, 2, null);
1348 $return = OEmbed::getHTML($url, $title);
1349 } catch (Exception $ex) {
1350 $return = $match[0];
1356 // Hide all [noparse] contained bbtags by spacefying them
1357 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1359 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1360 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1361 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1363 // Remove the abstract element. It is a non visible element.
1364 $text = self::stripAbstract($text);
1366 // Move all spaces out of the tags
1367 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1368 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1370 // Extract the private images which use data urls since preg has issues with
1371 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1372 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1374 $extracted = self::extractImagesFromItemBody($text);
1375 $text = $extracted['body'];
1376 $saved_image = $extracted['images'];
1378 // If we find any event code, turn it into an event.
1379 // After we're finished processing the bbcode we'll
1380 // replace all of the event code with a reformatted version.
1382 $ev = bbtoevent($text);
1384 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1385 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1387 $text = str_replace("<", "<", $text);
1388 $text = str_replace(">", ">", $text);
1390 // remove some newlines before the general conversion
1391 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1392 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1394 $text = preg_replace("/\n\[code\]/ism", "[code]", $text);
1395 $text = preg_replace("/\[\/code\]\n/ism", "[/code]", $text);
1397 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1399 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1402 // Check for [code] text here, before the linefeeds are messed with.
1403 // The highlighter will unescape and re-escape the content.
1404 if (strpos($text, '[code=') !== false) {
1405 $text = preg_replace_callback("/\[code=(.*?)\](.*?)\[\/code\]/ism", 'self::textHighlightCallback', $text);
1407 // Convert new line chars to html <br /> tags
1409 // nlbr seems to be hopelessly messed up
1410 // $Text = nl2br($Text);
1412 // We'll emulate it.
1414 $text = trim($text);
1415 $text = str_replace("\r\n", "\n", $text);
1417 // removing multiplicated newlines
1418 if (Config::get("system", "remove_multiplicated_lines")) {
1419 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1420 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1421 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1422 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1425 $text = str_replace($search, $replace, $text);
1426 } while ($oldtext != $text);
1429 // Set up the parameters for a URL search string
1430 $URLSearchString = "^\[\]";
1431 // Set up the parameters for a MAIL search string
1432 $MAILSearchString = $URLSearchString;
1434 // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text
1435 if (!$for_plaintext) {
1436 // Autolink feature (thanks to http://code.seebz.net/p/autolink-php/)
1437 // Currently disabled, since the function is too greedy
1438 // $autolink_regex = "`([^\]\=\"']|^)(https?\://[^\s<]+[^\s<\.\)])`ism";
1439 $autolink_regex = "/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism";
1440 $text = preg_replace($autolink_regex, '$1[url]$2[/url]', $text);
1441 if ($simple_html == 7) {
1442 $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForMastodonCallback', $text);
1443 $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForMastodonCallback', $text);
1446 $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text);
1447 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1451 // Handle attached links or videos
1452 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1454 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1456 // Remove all hashtag addresses
1457 if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7])) {
1458 $text = preg_replace("/([#@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $text);
1459 } elseif ($simple_html == 3) {
1460 // The ! is converted to @ since Diaspora only understands the @
1461 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1462 '@<a href="$2">$3</a>',
1464 } elseif ($simple_html == 7) {
1465 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1466 '$1<span class="vcard"><a href="$2" class="url" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1468 } elseif (!$simple_html) {
1469 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1470 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1474 // Bookmarks in red - will be converted to bookmarks in friendica
1475 $text = preg_replace("/#\^\[url\]([$URLSearchString]*)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1476 $text = preg_replace("/#\^\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1477 $text = preg_replace("/#\[url\=[$URLSearchString]*\]\^\[\/url\]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i",
1478 "[bookmark=$1]$2[/bookmark]", $text);
1480 if (in_array($simple_html, [2, 6, 7, 8, 9])) {
1481 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1482 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1483 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1486 if ($simple_html == 5) {
1487 $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1490 // Perform URL Search
1492 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1495 if ($simple_html == 5) {
1496 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1498 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1501 // Handle Diaspora posts
1502 $text = preg_replace_callback(
1503 "&\[url=/posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1505 return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1509 // Server independent link to posts and comments
1510 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1511 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1512 $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
1514 $text = preg_replace("/([#])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1515 '$1<a href="' . System::baseUrl() . '/search?tag=$3" class="tag" title="$3">$3</a>', $text);
1517 $text = preg_replace("/\[url\=([$URLSearchString]*)\]#(.*?)\[\/url\]/ism",
1518 '#<a href="' . System::baseUrl() . '/search?tag=$2" class="tag" title="$2">$2</a>', $text);
1520 $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1521 $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1522 //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $Text);
1524 // Red compatibility, though the link can't be authenticated on Friendica
1525 $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1528 // we may need to restrict this further if it picks up too many strays
1529 // link acct:user@host to a webfinger profile redirector
1531 $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);
1533 // Perform MAIL Search
1534 $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1535 $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1537 // leave open the posibility of [map=something]
1538 // this is replaced in prepare_body() which has knowledge of the item location
1540 if (strpos($text, '[/map]') !== false) {
1541 $text = preg_replace_callback(
1542 "/\[map\](.*?)\[\/map\]/ism",
1544 // the extra space in the following line is intentional
1545 // Whyyy? - @MrPetovan
1546 return str_replace($match[0], '<div class="map" >' . Map::byLocation($match[1]) . '</div>', $match[0]);
1551 if (strpos($text, '[map=') !== false) {
1552 $text = preg_replace_callback(
1553 "/\[map=(.*?)\]/ism",
1555 // the extra space in the following line is intentional
1556 // Whyyy? - @MrPetovan
1557 return str_replace($match[0], '<div class="map" >' . Map::byCoordinates(str_replace('/', ' ', $match[1])) . '</div>', $match[0]);
1562 if (strpos($text, '[map]') !== false) {
1563 $text = preg_replace("/\[map\]/", '<div class="map"></div>', $text);
1566 // Check for headers
1567 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1568 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1569 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1570 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1571 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1572 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1574 // Check for paragraph
1575 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1577 // Check for bold text
1578 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1580 // Check for Italics text
1581 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1583 // Check for Underline text
1584 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1586 // Check for strike-through text
1587 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<strike>$1</strike>', $text);
1589 // Check for over-line text
1590 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1592 // Check for colored text
1593 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1595 // Check for sized text
1596 // [size=50] --> font-size: 50px (with the unit).
1597 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1598 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1600 // Check for centered text
1601 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1603 // Check for list text
1604 $text = str_replace("[*]", "<li>", $text);
1606 // Check for style sheet commands
1607 $text = preg_replace_callback(
1608 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1610 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1615 // Check for CSS classes
1616 $text = preg_replace_callback(
1617 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1619 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1624 // handle nested lists
1627 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1628 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1629 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1630 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1631 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1632 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1633 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1634 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1635 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1636 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1637 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1638 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1639 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1640 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1643 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1644 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1645 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1646 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1648 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1649 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1651 $text = str_replace('[hr]', '<hr />', $text);
1653 // This is actually executed in prepare_body()
1655 $text = str_replace('[nosmile]', '', $text);
1657 // Check for font change text
1658 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1660 // Declare the format for [code] layout
1662 $CodeLayout = '<code>$1</code>';
1663 // Check for [code] text
1664 $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text);
1666 // Declare the format for [spoiler] layout
1667 $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
1669 // Check for [spoiler] text
1670 // handle nested quotes
1672 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1673 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
1676 // Check for [spoiler=Author] text
1678 $t_wrote = L10n::t('$1 wrote:');
1680 // handle nested quotes
1682 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1683 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1684 "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
1688 // Declare the format for [quote] layout
1689 $QuoteLayout = '<blockquote>$1</blockquote>';
1691 // Check for [quote] text
1692 // handle nested quotes
1694 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1695 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1698 // Check for [quote=Author] text
1700 $t_wrote = L10n::t('$1 wrote:');
1702 // handle nested quotes
1704 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1705 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1706 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1711 // [img=widthxheight]image source[/img]
1712 $text = preg_replace_callback(
1713 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1714 function ($matches) {
1715 if (strpos($matches[3], "data:image/") === 0) {
1719 $matches[3] = proxy_url($matches[3]);
1720 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1725 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1726 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1729 // [img]pathtoimage[/img]
1730 $text = preg_replace_callback(
1731 "/\[img\](.*?)\[\/img\]/ism",
1732 function ($matches) {
1733 if (strpos($matches[1], "data:image/") === 0) {
1737 $matches[1] = proxy_url($matches[1]);
1738 return "[img]" . $matches[1] . "[/img]";
1743 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1744 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1747 $text = preg_replace_callback("/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
1748 function ($match) use ($simple_html) {
1749 return self::convertShare($match, $simple_html);
1752 $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);
1753 $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 />', $text);
1754 //$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);
1758 $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);
1759 $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);
1761 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1762 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1764 $text = preg_replace("/\[video\](.*?)\[\/video\]/",
1765 '<a href="$1" target="_blank">$1</a>', $text);
1766 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/",
1767 '<a href="$1" target="_blank">$1</a>', $text);
1770 // html5 video and audio
1774 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1776 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1779 // Youtube extensions
1781 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1782 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1783 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1786 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1787 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1788 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1791 $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);
1793 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1794 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1798 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1799 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1802 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1803 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1806 $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);
1808 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1809 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1813 $text = OEmbed::BBCode2HTML($text);
1815 // Avoid triple linefeeds through oembed
1816 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1818 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1819 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1820 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1821 // start which is always required). Allow desc with a missing summary for compatibility.
1823 if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) {
1824 $sub = format_event_html($ev, $simple_html);
1826 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1827 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1828 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1829 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1830 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1831 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1832 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1835 // Replace non graphical smilies for external posts
1837 $text = Smilies::replace($text, false, true);
1840 // Replace inline code blocks
1841 $text = preg_replace_callback("|(?!<br[^>]*>)<code>([^<]*)</code>(?!<br[^>]*>)|ism",
1842 function ($match) use ($simple_html) {
1843 $return = '<key>' . $match[1] . '</key>';
1844 // Use <code> for Diaspora inline code blocks
1845 if ($simple_html === 3) {
1846 $return = '<code>' . $match[1] . '</code>';
1852 // Unhide all [noparse] contained bbtags unspacefying them
1853 // and triming the [noparse] tag.
1855 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1856 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1857 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1860 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1861 $text = preg_replace('/\&\#039\;/', '\'', $text);
1862 $text = preg_replace('/\"\;/', '"', $text);
1864 // fix any escaped ampersands that may have been converted into links
1865 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1867 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1868 $allowed_src_protocols = ['http', 'redir', 'cid'];
1869 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1870 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
1872 // sanitize href attributes (only whitelisted protocols URLs)
1873 // default value for backward compatibility
1874 $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']);
1876 // Always allowed protocol even if config isn't set or not including it
1877 $allowed_link_protocols[] = 'http';
1878 $allowed_link_protocols[] = 'redir/';
1880 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1881 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
1884 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1887 // Clean up the HTML by loading and saving the HTML with the DOM.
1888 // Bad structured html can break a whole page.
1889 // For performance reasons do it only with ativated item cache or at export.
1890 if (!$try_oembed || (get_itemcachepath() != "")) {
1891 $doc = new DOMDocument();
1892 $doc->preserveWhiteSpace = false;
1894 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1896 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1897 $encoding = '<?xml encoding="UTF-8">';
1898 @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
1899 $doc->encoding = 'UTF-8';
1900 $text = $doc->saveHTML();
1901 $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
1903 $text = str_replace('<br></li>', '</li>', $text);
1905 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1908 // Clean up some useless linebreaks in lists
1909 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1910 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1911 //$Text = str_replace('</li><br />', '</li>', $Text);
1912 //$Text = str_replace('<br /><li>', '<li>', $Text);
1913 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1915 Addon::callHooks('bbcode', $text);
1921 * @brief Strips the "abstract" tag from the provided text
1923 * @param string $text The text with BBCode
1924 * @return string The same text - but without "abstract" element
1926 public static function stripAbstract($text)
1928 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1929 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1935 * @brief Returns the value of the "abstract" element
1937 * @param string $text The text that maybe contains the element
1938 * @param string $addon The addon for which the abstract is meant for
1939 * @return string The abstract
1941 private static function getAbstract($text, $addon = "")
1945 $addon = strtolower($addon);
1947 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1948 foreach ($results AS $result) {
1949 $abstracts[strtolower($result[1])] = $result[2];
1953 if (isset($abstracts[$addon])) {
1954 $abstract = $abstracts[$addon];
1957 if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1958 $abstract = $result[1];
1965 * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
1967 * @param array $match Matching values for the callback
1968 * @return string Replaced mention
1970 private static function bbCodeMention2DiasporaCallback($match)
1972 $contact = Contact::getDetailsByURL($match[3]);
1974 if (empty($contact['addr'])) {
1975 $contact = Probe::uri($match[3]);
1978 if (empty($contact['addr'])) {
1982 $mention = '@{' . $match[2] . '; ' . $contact['addr'] . '}';
1987 * @brief Converts a BBCode text into Markdown
1989 * This function converts a BBCode item body to be sent to Markdown-enabled
1990 * systems like Diaspora and Libertree
1992 * @param string $text
1993 * @param bool $for_diaspora Diaspora requires more changes than Libertree
1996 public static function toMarkdown($text, $for_diaspora = true)
1998 $a = self::getApp();
2000 $original_text = $text;
2002 // Since Diaspora is creating a summary for links, this function removes them before posting
2003 if ($for_diaspora) {
2004 $text = self::removeShareInformation($text);
2008 * Transform #tags, strip off the [url] and replace spaces with underscore
2010 $url_search_string = "^\[\]";
2011 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
2012 function ($matches) {
2013 return '#' . str_replace(' ', '_', $matches[2]);
2018 // Converting images with size parameters to simple images. Markdown doesn't know it.
2019 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2021 // Extracting multi-line code blocks before the whitespace processing/code highlighter in self::convert()
2024 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#is",
2025 function ($matches) use (&$codeblocks) {
2026 $return = $matches[0];
2027 if (strpos($matches[2], "\n") !== false) {
2028 $return = '#codeblock-' . count($codeblocks) . '#';
2030 $prefix = '````' . $matches[1] . PHP_EOL;
2031 $codeblocks[] = $prefix . trim($matches[2]) . PHP_EOL . '````';
2038 // Convert it to HTML - don't try oembed
2039 if ($for_diaspora) {
2040 $text = self::convert($text, false, 3);
2042 // Add all tags that maybe were removed
2043 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2045 foreach ($tags[2] as $tag) {
2046 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2047 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2048 $tagline .= '#' . $tag . ' ';
2051 $text = $text . " " . $tagline;
2054 $text = self::convert($text, false, 4);
2057 // mask some special HTML chars from conversation to markdown
2058 $text = str_replace(['<', '>', '&'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text);
2060 // If a link is followed by a quote then there should be a newline before it
2061 // Maybe we should make this newline at every time before a quote.
2062 $text = str_replace(["</a><blockquote>"], ["</a><br><blockquote>"], $text);
2064 $stamp1 = microtime(true);
2066 // Now convert HTML to Markdown
2067 $converter = new HtmlConverter();
2068 $text = $converter->convert($text);
2070 // unmask the special chars back to HTML
2071 $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
2073 $a->save_timestamp($stamp1, "parser");
2075 // Libertree has a problem with escaped hashtags.
2076 $text = str_replace(['\#'], ['#'], $text);
2078 // Remove any leading or trailing whitespace, as this will mess up
2079 // the Diaspora signature verification and cause the item to disappear
2080 $text = trim($text);
2082 if ($for_diaspora) {
2083 $url_search_string = "^\[\]";
2084 $text = preg_replace_callback(
2085 "/([@]\[(.*?)\])\(([$url_search_string]*?)\)/ism",
2086 ['self', 'bbCodeMention2DiasporaCallback'],
2091 // Restore code blocks
2092 $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
2093 function ($matches) use ($codeblocks) {
2095 if (isset($codeblocks[intval($matches[1])])) {
2096 $return = $codeblocks[$matches[1]];
2103 Addon::callHooks('bb2diaspora', $text);