3 * @file src/Content/Text/BBCode.php
6 namespace Friendica\Content\Text;
11 use Friendica\BaseObject;
12 use Friendica\Content\OEmbed;
13 use Friendica\Content\Smilies;
14 use Friendica\Core\Addon;
15 use Friendica\Core\Cache;
16 use Friendica\Core\Config;
17 use Friendica\Core\L10n;
18 use Friendica\Core\Logger;
19 use Friendica\Core\Protocol;
20 use Friendica\Core\Renderer;
21 use Friendica\Core\System;
22 use Friendica\Model\Contact;
23 use Friendica\Model\Event;
24 use Friendica\Network\Probe;
25 use Friendica\Object\Image;
26 use Friendica\Util\Map;
27 use Friendica\Util\Network;
28 use Friendica\Util\ParseUrl;
29 use Friendica\Util\Proxy as ProxyUtils;
31 class BBCode extends BaseObject
34 * @brief Fetches attachment data that were generated the old way
36 * @param string $body Message body
38 * 'type' -> Message type ("link", "video", "photo")
39 * 'text' -> Text before the shared message
40 * 'after' -> Text after the shared message
41 * 'image' -> Preview image of the message
42 * 'url' -> Url to the attached message
43 * 'title' -> Title of the attachment
44 * 'description' -> Description of the attachment
46 private static function getOldAttachmentData($body)
50 // Simplify image codes
51 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
53 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
54 foreach ($attached as $data) {
55 if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
59 $post["type"] = substr($data[1], 5);
61 $pos = strpos($body, $data[0]);
63 $post["text"] = trim(substr($body, 0, $pos));
64 $post["after"] = trim(substr($body, $pos + strlen($data[0])));
66 $post["text"] = trim(str_replace($data[0], "", $body));
69 $attacheddata = $data[2];
71 $URLSearchString = "^\[\]";
73 if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
75 $picturedata = Image::getInfoFromURL($matches[1]);
78 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
79 $post["image"] = $matches[1];
81 $post["preview"] = $matches[1];
86 if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
87 $post["url"] = $matches[1];
88 $post["title"] = $matches[2];
90 if (!empty($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 = "^\[\]";
245 $body = preg_replace("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
247 if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
248 if ((count($pictures) == 1) && !$has_title) {
249 // Checking, if the link goes to a picture
250 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
253 // Sometimes photo posts to the own album are not detected at the start.
254 // So we seem to cannot use the cache for these cases. That's strange.
255 if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
256 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
259 if ($data["type"] == "photo") {
260 $post["type"] = "photo";
261 if (isset($data["images"][0])) {
262 $post["image"] = $data["images"][0]["src"];
263 $post["url"] = $data["url"];
265 $post["image"] = $data["url"];
268 $post["preview"] = $pictures[0][2];
269 $post["text"] = str_replace($pictures[0][0], "", $body);
271 $imgdata = Image::getInfoFromURL($pictures[0][1]);
272 if ($imgdata && substr($imgdata["mime"], 0, 6) == "image/") {
273 $post["type"] = "photo";
274 $post["image"] = $pictures[0][1];
275 $post["preview"] = $pictures[0][2];
276 $post["text"] = str_replace($pictures[0][0], "", $body);
279 } elseif (count($pictures) > 0) {
280 $post["type"] = "link";
281 $post["url"] = $plink;
282 $post["image"] = $pictures[0][2];
283 $post["text"] = $body;
285 } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
286 if ((count($pictures) == 1) && !$has_title) {
287 $post["type"] = "photo";
288 $post["image"] = $pictures[0][1];
289 $post["text"] = str_replace($pictures[0][0], "", $body);
290 } elseif (count($pictures) > 0) {
291 $post["type"] = "link";
292 $post["url"] = $plink;
293 $post["image"] = $pictures[0][1];
294 $post["text"] = $body;
298 // Test for the external links
299 preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
300 preg_match_all("(\[url\=([$URLSearchString]*)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
302 $links = array_merge($links1, $links2);
304 // If there is only a single one, then use it.
305 // This should cover link posts via API.
306 if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
307 $post["type"] = "link";
308 $post["text"] = trim($body);
309 $post["url"] = $links[0][1];
312 // Now count the number of external media links
313 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
314 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
315 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
316 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
318 // Add them to the other external links
319 $links = array_merge($links, $links1, $links2, $links3, $links4);
321 // Are there more than one?
322 if (count($links) > 1) {
323 // The post will be the type "text", which means a blog post
324 unset($post["type"]);
325 $post["url"] = $plink;
328 if (!isset($post["type"])) {
329 $post["type"] = "text";
330 $post["text"] = trim($body);
332 } elseif (isset($post["url"]) && ($post["type"] == "video")) {
333 $data = ParseUrl::getSiteinfoCached($post["url"], true);
335 if (isset($data["images"][0])) {
336 $post["image"] = $data["images"][0]["src"];
344 * @brief Converts a BBCode text into plaintext
346 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
350 public static function toPlaintext($text, $keep_urls = true)
352 $naked_text = preg_replace('/\[(.+?)\]\s*/','', $text);
354 $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text);
360 private static function proxyUrl($image, $simplehtml = false)
362 // Only send proxied pictures to API and for internal display
363 if (in_array($simplehtml, [false, 2])) {
364 return ProxyUtils::proxifyUrl($image);
370 public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
372 // Suppress "view full size"
373 if (intval(Config::get('system', 'no_view_full_size'))) {
374 $include_link = false;
377 // Picture addresses can contain special characters
378 $s = htmlspecialchars_decode($srctext);
381 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
383 foreach ($matches as $mtch) {
384 Logger::log('scale_external_image: ' . $mtch[1]);
386 $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
387 if (stristr($mtch[1], $hostname)) {
391 // $scale_replace, if passed, is an array of two elements. The
392 // first is the name of the full-size image. The second is the
393 // name of a remote, scaled-down version of the full size image.
394 // This allows Friendica to display the smaller remote image if
395 // one exists, while still linking to the full-size image
396 if ($scale_replace) {
397 $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
401 $i = Network::fetchUrl($scaled);
406 // guess mimetype from headers or filename
407 $type = Image::guessType($mtch[1], true);
410 $Image = new Image($i, $type);
411 if ($Image->isValid()) {
412 $orig_width = $Image->getWidth();
413 $orig_height = $Image->getHeight();
415 if ($orig_width > 640 || $orig_height > 640) {
416 $Image->scaleDown(640);
417 $new_width = $Image->getWidth();
418 $new_height = $Image->getHeight();
419 Logger::log('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], Logger::DEBUG);
422 '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
423 . "\n" . (($include_link)
424 ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
428 Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG);
435 // replace the special char encoding
436 $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
441 * The purpose of this function is to apply system message length limits to
442 * imported messages without including any embedded photos in the length
444 * @brief Truncates imported message body string length to max_import_size
445 * @param string $body
448 public static function limitBodySize($body)
450 $maxlen = Config::get('config', 'max_import_size', 0);
452 // If the length of the body, including the embedded images, is smaller
453 // than the maximum, then don't waste time looking for the images
454 if ($maxlen && (strlen($body) > $maxlen)) {
456 Logger::log('the total body length exceeds the limit', Logger::DEBUG);
462 $img_start = strpos($orig_body, '[img');
463 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
464 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
465 while (($img_st_close !== false) && ($img_end !== false)) {
467 $img_st_close++; // make it point to AFTER the closing bracket
468 $img_end += $img_start;
469 $img_end += strlen('[/img]');
471 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
472 // This is an embedded image
474 if (($textlen + $img_start) > $maxlen) {
475 if ($textlen < $maxlen) {
476 Logger::log('the limit happens before an embedded image', Logger::DEBUG);
477 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
481 $new_body = $new_body . substr($orig_body, 0, $img_start);
482 $textlen += $img_start;
485 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
488 if (($textlen + $img_end) > $maxlen) {
489 if ($textlen < $maxlen) {
490 Logger::log('the limit happens before the end of a non-embedded image', Logger::DEBUG);
491 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
495 $new_body = $new_body . substr($orig_body, 0, $img_end);
496 $textlen += $img_end;
499 $orig_body = substr($orig_body, $img_end);
501 if ($orig_body === false) {
502 // in case the body ends on a closing image tag
506 $img_start = strpos($orig_body, '[img');
507 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
508 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
511 if (($textlen + strlen($orig_body)) > $maxlen) {
512 if ($textlen < $maxlen) {
513 Logger::log('the limit happens after the end of the last image', Logger::DEBUG);
514 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
517 Logger::log('the text size with embedded images extracted did not violate the limit', Logger::DEBUG);
518 $new_body = $new_body . $orig_body;
528 * Processes [attachment] tags
530 * Note: Can produce a [bookmark] tag in the returned string
532 * @brief Processes [attachment] tags
533 * @param string $return
534 * @param bool|int $simplehtml
535 * @param bool $tryoembed
538 private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
540 $data = self::getAttachmentData($return);
541 if (empty($data) || empty($data["url"])) {
545 if (isset($data["title"])) {
546 $data["title"] = strip_tags($data["title"]);
547 $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
549 $data["title"] = null;
552 if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data["image"])) {
553 $data["preview"] = $data["image"];
558 if ($simplehtml == 7) {
559 $return = self::convertUrlForOStatus($data["url"]);
560 } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
561 $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data["url"], $data["title"]);
564 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
565 $return = OEmbed::getHTML($data['url'], $data['title']);
567 throw new Exception('OEmbed is disabled for this attachment.');
569 } catch (Exception $e) {
570 $data["title"] = defaults($data, 'title', $data['url']);
572 if ($simplehtml != 4) {
573 $return = sprintf('<div class="type-%s">', $data["type"]);
576 if (!empty($data['title']) && !empty($data['url'])) {
577 if (!empty($data["image"]) && empty($data["text"]) && ($data["type"] == "photo")) {
578 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
580 if (!empty($data["image"])) {
581 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
582 } elseif (!empty($data["preview"])) {
583 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], self::proxyUrl($data["preview"], $simplehtml), $data["title"]);
585 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
589 if (!empty($data["description"]) && $data["description"] != $data["title"]) {
590 // Sanitize the HTML by converting it to BBCode
591 $bbcode = HTML::toBBCode($data["description"]);
592 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
595 if (!empty($data['url'])) {
596 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
599 if ($simplehtml != 4) {
605 return trim(defaults($data, 'text', '') . ' ' . $return . ' ' . defaults($data, 'after', ''));
608 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
610 $data = self::getAttachmentData($Text);
615 return $data["text"] . defaults($data, 'after', '');
618 $title = htmlentities(defaults($data, 'title', ''), ENT_QUOTES, 'UTF-8', false);
619 $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
620 if ($plaintext || (($title != "") && strstr($text, $title))) {
621 $data["title"] = $data["url"];
622 } elseif (($text != "") && strstr($title, $text)) {
623 $data["text"] = $data["title"];
624 $data["title"] = $data["url"];
627 if (($data["text"] == "") && ($data["title"] != "") && ($data["url"] == "")) {
628 return $data["title"] . $data["after"];
631 // If the link already is included in the post, don't add it again
632 if (!empty($data["url"]) && strpos($data["text"], $data["url"])) {
633 return $data["text"] . $data["after"];
636 $text = $data["text"];
638 if (!empty($data["url"]) && !empty($data["title"])) {
639 $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
640 } elseif (!empty($data["url"])) {
641 $text .= "\n[url]" . $data["url"] . "[/url]";
644 return $text . "\n" . $data["after"];
648 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
650 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
651 * @param array $match Array with the matching values
652 * @return string reformatted link including HTML codes
654 private static function convertUrlForOStatusCallback($match)
658 if (isset($match[2]) && ($match[1] != $match[2])) {
662 $parts = parse_url($url);
663 if (!isset($parts['scheme'])) {
667 return self::convertUrlForOStatus($url);
671 * @brief Converts [url] BBCodes in a format that looks fine on OStatus systems.
672 * @param string $url URL that is about to be reformatted
673 * @return string reformatted link including HTML codes
675 private static function convertUrlForOStatus($url)
677 $parts = parse_url($url);
678 $scheme = $parts['scheme'] . '://';
679 $styled_url = str_replace($scheme, '', $url);
681 if (strlen($styled_url) > 30) {
682 $styled_url = substr($styled_url, 0, 30) . "…";
685 $html = '<a href="%s" target="_blank">%s</a>';
687 return sprintf($html, $url, $styled_url);
691 * [noparse][i]italic[/i][/noparse] turns into
692 * [noparse][ i ]italic[ /i ][/noparse],
693 * to hide them from parser.
695 private static function escapeNoparseCallback($match)
697 $whole_match = $match[0];
698 $captured = $match[1];
699 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
700 $new_str = str_replace($captured, $spacefied, $whole_match);
705 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
706 * now turns back and the [noparse] tags are trimed
707 * returning [i]italic[/i]
709 private static function unescapeNoparseCallback($match)
711 $captured = $match[1];
712 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
717 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
720 * @param string $text Text to search
721 * @param string $name Tag name
722 * @param int $occurrences Number of first occurrences to skip
723 * @return boolean|array
725 public static function getTagPosition($text, $name, $occurrences = 0)
727 if ($occurrences < 0) {
732 for ($i = 0; $i <= $occurrences; $i++) {
733 if ($start_open !== false) {
734 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
738 if ($start_open === false) {
742 $start_equal = strpos($text, '=', $start_open);
743 $start_close = strpos($text, ']', $start_open);
745 if ($start_close === false) {
751 $end_open = strpos($text, '[/' . $name . ']', $start_close);
753 if ($end_open === false) {
759 'open' => $start_open,
760 'close' => $start_close
764 'close' => $end_open + strlen('[/' . $name . ']')
768 if ($start_equal !== false) {
769 $res['start']['equal'] = $start_equal + 1;
776 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
778 * @param type $pattern Preg pattern string
779 * @param type $replace Preg replace string
780 * @param type $name BBCode tag name
781 * @param type $text Text to search
784 public static function pregReplaceInTag($pattern, $replace, $name, $text)
787 $pos = self::getTagPosition($text, $name, $occurrences);
788 while ($pos !== false && $occurrences++ < 1000) {
789 $start = substr($text, 0, $pos['start']['open']);
790 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
791 $end = substr($text, $pos['end']['close']);
792 if ($end === false) {
796 $subject = preg_replace($pattern, $replace, $subject);
797 $text = $start . $subject . $end;
799 $pos = self::getTagPosition($text, $name, $occurrences);
805 private static function extractImagesFromItemBody($body)
812 $img_start = strpos($orig_body, '[img');
813 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
814 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
815 while (($img_st_close !== false) && ($img_end !== false)) {
816 $img_st_close++; // make it point to AFTER the closing bracket
817 $img_end += $img_start;
819 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
820 // This is an embedded image
821 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
822 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
826 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
829 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
831 if ($orig_body === false) {
832 // in case the body ends on a closing image tag
836 $img_start = strpos($orig_body, '[img');
837 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
838 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
841 $new_body = $new_body . $orig_body;
843 return ['body' => $new_body, 'images' => $saved_image];
846 private static function interpolateSavedImagesIntoItemBody($body, array $images)
851 foreach ($images as $image) {
852 // We're depending on the property of 'foreach' (specified on the PHP website) that
853 // it loops over the array starting from the first element and going sequentially
854 // to the last element
855 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
856 '<img src="' . self::proxyUrl($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
864 * This function converts a [share] block to text according to a provided callback function whose signature is:
866 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
869 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
871 * - $author_contact is a contact record array
872 * - $content is the inner content of the [share] block
873 * - $is_quote_share indicates whether there's any content before the [share] block
874 * - Return value is the string that should replace the [share] block in the provided text
876 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
878 * @param string $text A BBCode string
879 * @param callable $callback
880 * @return string The BBCode string with all [share] blocks replaced
882 public static function convertShare($text, callable $callback)
884 $return = preg_replace_callback(
885 "/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
886 function ($match) use ($callback) {
887 $attribute_string = $match[2];
890 foreach(['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
891 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
892 $attributes[$field] = html_entity_decode(defaults($matches, 2, ''), ENT_QUOTES, 'UTF-8');
895 // We only call this so that a previously unknown contact can be added.
896 // This is important for the function "Model\Contact::getDetailsByURL()".
897 // This function then can fetch an entry from the contact table.
898 Contact::getIdForURL($attributes['profile'], 0, true);
900 $author_contact = Contact::getDetailsByURL($attributes['profile']);
901 $author_contact['addr'] = defaults($author_contact, 'addr' , Protocol::getAddrFromProfileUrl($attributes['profile']));
903 $attributes['author'] = defaults($author_contact, 'name' , $attributes['author']);
904 $attributes['avatar'] = defaults($author_contact, 'micro', $attributes['avatar']);
905 $attributes['profile'] = defaults($author_contact, 'url' , $attributes['profile']);
907 if ($attributes['avatar']) {
908 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
911 return $match[1] . $callback($attributes, $author_contact, $match[3], trim($match[1]) != '');
920 * Default [share] tag conversion callback
922 * Note: Can produce a [bookmark] tag in the output
924 * @see BBCode::convertShare()
925 * @param array $attributes [share] block attribute values
926 * @param array $author_contact Contact row of the shared author
927 * @param string $content Inner content of the [share] block
928 * @param boolean $is_quote_share Whether there is content before the [share] block
929 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
932 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
934 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
936 switch ($simplehtml) {
938 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' <a href="' . $attributes['profile'] . '">' . $mention . '</a>: </p>' . "\n" . '«' . $content . '»';
941 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
944 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . $mention . ':</b></p>' . "\n";
946 if (stripos(normalise_link($attributes['link']), 'http://twitter.com/') === 0) {
947 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
949 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
951 if ($attributes['link'] != '') {
952 $text .= '<p><a href="' . $attributes['link'] . '">[l]</a></p>' . "\n";
958 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
959 $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
960 $headline .= ':</b></p>' . "\n";
962 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
966 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
968 case 7: // statusnet/GNU Social
969 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
972 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n";
973 $text .= '<p>' . $content . '</p>' . "\n";
975 if ($attributes['link'] != '') {
976 $text .= '<p>' . $attributes['link'] . '</p>';
980 // Transforms quoted tweets in rich attachments to avoid nested tweets
981 if (stripos(normalise_link($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) {
983 $text = ($is_quote_share? '<br />' : '') . OEmbed::getHTML($attributes['link']);
984 } catch (Exception $e) {
985 $text = ($is_quote_share? '<br />' : '') . sprintf('[bookmark=%s]%s[/bookmark]', $attributes['link'], $content);
988 $text = ($is_quote_share? "\n" : '');
990 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
991 $text .= Renderer::replaceMacros($tpl, [
992 '$profile' => $attributes['profile'],
993 '$avatar' => $attributes['avatar'],
994 '$author' => $attributes['author'],
995 '$link' => $attributes['link'],
996 '$posted' => $attributes['posted'],
997 '$content' => trim($content)
1006 private static function removePictureLinksCallback($match)
1008 $text = Cache::get($match[1]);
1010 if (is_null($text)) {
1011 $a = self::getApp();
1013 $stamp1 = microtime(true);
1015 $ch = @curl_init($match[1]);
1016 @curl_setopt($ch, CURLOPT_NOBODY, true);
1017 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1018 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1020 $curl_info = @curl_getinfo($ch);
1022 $a->saveTimestamp($stamp1, "network");
1024 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1025 $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
1027 $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
1029 // if its not a picture then look if its a page that contains a picture link
1030 $body = Network::fetchUrl($match[1]);
1032 $doc = new DOMDocument();
1033 @$doc->loadHTML($body);
1034 $xpath = new DOMXPath($doc);
1035 $list = $xpath->query("//meta[@name]");
1036 foreach ($list as $node) {
1039 if ($node->attributes->length) {
1040 foreach ($node->attributes as $attribute) {
1041 $attr[$attribute->name] = $attribute->value;
1045 if (strtolower($attr["name"]) == "twitter:image") {
1046 $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
1050 Cache::set($match[1], $text);
1056 private static function expandLinksCallback($match)
1058 if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1059 return ($match[1] . "[url]" . $match[2] . "[/url]");
1061 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1065 private static function cleanPictureLinksCallback($match)
1067 $text = Cache::get($match[1]);
1069 if (is_null($text)) {
1070 $a = self::getApp();
1072 $stamp1 = microtime(true);
1074 $ch = @curl_init($match[1]);
1075 @curl_setopt($ch, CURLOPT_NOBODY, true);
1076 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1077 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1079 $curl_info = @curl_getinfo($ch);
1081 $a->saveTimestamp($stamp1, "network");
1083 // if its a link to a picture then embed this picture
1084 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1085 $text = "[img]" . $match[1] . "[/img]";
1087 $text = "[img]" . $match[2] . "[/img]";
1089 // if its not a picture then look if its a page that contains a picture link
1090 $body = Network::fetchUrl($match[1]);
1092 $doc = new DOMDocument();
1093 @$doc->loadHTML($body);
1094 $xpath = new DOMXPath($doc);
1095 $list = $xpath->query("//meta[@name]");
1096 foreach ($list as $node) {
1098 if ($node->attributes->length) {
1099 foreach ($node->attributes as $attribute) {
1100 $attr[$attribute->name] = $attribute->value;
1104 if (strtolower($attr["name"]) == "twitter:image") {
1105 $text = "[img]" . $attr["content"] . "[/img]";
1109 Cache::set($match[1], $text);
1115 public static function cleanPictureLinks($text)
1117 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1122 * @brief Converts a BBCode message to HTML message
1124 * BBcode 2 HTML was written by WAY2WEB.net
1125 * extended to work with Mistpark/Friendica - Mike Macgirvin
1127 * Simple HTML values meaning:
1128 * - 0: Friendica display
1130 * - 2: Used for Google+, Windows Phone push, Friendica API
1131 * - 3: Used before converting to Markdown in bb2diaspora.php
1132 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1135 * - 7: Used for dfrn, OStatus
1136 * - 8: Used for WP backlink text setting
1138 * @param string $text
1139 * @param bool $try_oembed
1140 * @param int $simple_html
1141 * @param bool $for_plaintext
1144 public static function convert($text, $try_oembed = true, $simple_html = false, $for_plaintext = false)
1146 $a = self::getApp();
1149 * preg_match_callback function to replace potential Oembed tags with Oembed content
1151 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1153 * $match[2] = $title or absent
1155 $try_oembed_callback = function ($match)
1158 $title = defaults($match, 2, null);
1161 $return = OEmbed::getHTML($url, $title);
1162 } catch (Exception $ex) {
1163 $return = $match[0];
1169 // Extracting multi-line code blocks before the whitespace processing
1172 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#is",
1173 function ($matches) use (&$codeblocks) {
1174 $return = $matches[0];
1175 if (strpos($matches[2], "\n") !== false) {
1176 $return = '#codeblock-' . count($codeblocks) . '#';
1178 $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . trim($matches[2], "\n\r") . '</code></pre>';
1185 // Hide all [noparse] contained bbtags by spacefying them
1186 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1188 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1189 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1190 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1192 // Remove the abstract element. It is a non visible element.
1193 $text = self::stripAbstract($text);
1195 // Move all spaces out of the tags
1196 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1197 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1199 // Extract the private images which use data urls since preg has issues with
1200 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1201 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1203 $extracted = self::extractImagesFromItemBody($text);
1204 $text = $extracted['body'];
1205 $saved_image = $extracted['images'];
1207 // If we find any event code, turn it into an event.
1208 // After we're finished processing the bbcode we'll
1209 // replace all of the event code with a reformatted version.
1211 $ev = Event::fromBBCode($text);
1213 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1214 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1216 $text = str_replace("<", "<", $text);
1217 $text = str_replace(">", ">", $text);
1219 // remove some newlines before the general conversion
1220 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1221 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1223 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1225 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1228 // Convert new line chars to html <br /> tags
1230 // nlbr seems to be hopelessly messed up
1231 // $Text = nl2br($Text);
1233 // We'll emulate it.
1235 $text = trim($text);
1236 $text = str_replace("\r\n", "\n", $text);
1238 // removing multiplicated newlines
1239 if (Config::get("system", "remove_multiplicated_lines")) {
1240 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1241 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1242 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1243 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1246 $text = str_replace($search, $replace, $text);
1247 } while ($oldtext != $text);
1250 // Set up the parameters for a URL search string
1251 $URLSearchString = "^\[\]";
1252 // Set up the parameters for a MAIL search string
1253 $MAILSearchString = $URLSearchString;
1255 // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text
1256 if (!$for_plaintext) {
1257 // Autolink feature (thanks to http://code.seebz.net/p/autolink-php/)
1258 // Currently disabled, since the function is too greedy
1259 // $autolink_regex = "`([^\]\=\"']|^)(https?\://[^\s<]+[^\s<\.\)])`ism";
1260 $autolink_regex = "/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism";
1261 $text = preg_replace($autolink_regex, '$1[url]$2[/url]', $text);
1262 if ($simple_html == 7) {
1263 $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1264 $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1267 $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text);
1268 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1272 // Handle attached links or videos
1273 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1275 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1277 // Remove all hashtag addresses
1278 if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7])) {
1279 $text = preg_replace("/([#@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $text);
1280 } elseif ($simple_html == 3) {
1281 // The ! is converted to @ since Diaspora only understands the @
1282 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1283 '@<a href="$2">$3</a>',
1285 } elseif ($simple_html == 7) {
1286 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1287 '$1<span class="vcard"><a href="$2" class="url" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1289 } elseif (!$simple_html) {
1290 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1291 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1295 // Bookmarks in red - will be converted to bookmarks in friendica
1296 $text = preg_replace("/#\^\[url\]([$URLSearchString]*)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1297 $text = preg_replace("/#\^\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1298 $text = preg_replace("/#\[url\=[$URLSearchString]*\]\^\[\/url\]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i",
1299 "[bookmark=$1]$2[/bookmark]", $text);
1301 if (in_array($simple_html, [2, 6, 7, 8, 9])) {
1302 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1303 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1304 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1307 if ($simple_html == 5) {
1308 $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1311 // Perform URL Search
1313 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1316 if ($simple_html == 5) {
1317 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1319 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1322 // Handle Diaspora posts
1323 $text = preg_replace_callback(
1324 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1326 return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1330 $text = preg_replace_callback(
1331 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1333 return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1337 // Server independent link to posts and comments
1338 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1339 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1340 $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
1342 $text = preg_replace("/([#])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1343 '$1<a href="' . System::baseUrl() . '/search?tag=$3" class="tag" title="$3">$3</a>', $text);
1345 $text = preg_replace("/\[url\=([$URLSearchString]*)\]#(.*?)\[\/url\]/ism",
1346 '#<a href="' . System::baseUrl() . '/search?tag=$2" class="tag" title="$2">$2</a>', $text);
1348 $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1349 $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1350 //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $Text);
1352 // Red compatibility, though the link can't be authenticated on Friendica
1353 $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1356 // we may need to restrict this further if it picks up too many strays
1357 // link acct:user@host to a webfinger profile redirector
1359 $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);
1361 // Perform MAIL Search
1362 $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1363 $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1365 // leave open the posibility of [map=something]
1366 // this is replaced in prepare_body() which has knowledge of the item location
1368 if (strpos($text, '[/map]') !== false) {
1369 $text = preg_replace_callback(
1370 "/\[map\](.*?)\[\/map\]/ism",
1371 function ($match) use ($simple_html) {
1372 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1377 if (strpos($text, '[map=') !== false) {
1378 $text = preg_replace_callback(
1379 "/\[map=(.*?)\]/ism",
1380 function ($match) use ($simple_html) {
1381 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1386 if (strpos($text, '[map]') !== false) {
1387 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1390 // Check for headers
1391 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1392 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1393 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1394 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1395 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1396 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1398 // Check for paragraph
1399 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1401 // Check for bold text
1402 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1404 // Check for Italics text
1405 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1407 // Check for Underline text
1408 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1410 // Check for strike-through text
1411 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1413 // Check for over-line text
1414 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1416 // Check for colored text
1417 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1419 // Check for sized text
1420 // [size=50] --> font-size: 50px (with the unit).
1421 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1422 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1424 // Check for centered text
1425 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1427 // Check for list text
1428 $text = str_replace("[*]", "<li>", $text);
1430 // Check for style sheet commands
1431 $text = preg_replace_callback(
1432 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1434 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1439 // Check for CSS classes
1440 $text = preg_replace_callback(
1441 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1443 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1448 // handle nested lists
1451 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1452 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1453 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1454 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1455 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1456 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1457 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1458 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1459 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1460 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1461 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1462 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1463 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1464 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1467 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1468 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1469 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1470 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1472 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1473 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1475 $text = str_replace('[hr]', '<hr />', $text);
1477 // This is actually executed in prepare_body()
1479 $text = str_replace('[nosmile]', '', $text);
1481 // Check for font change text
1482 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1484 // Declare the format for [code] layout
1486 $CodeLayout = '<code>$1</code>';
1487 // Check for [code] text
1488 $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text);
1490 // Declare the format for [spoiler] layout
1491 $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
1493 // Check for [spoiler] text
1494 // handle nested quotes
1496 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1497 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
1500 // Check for [spoiler=Author] text
1502 $t_wrote = L10n::t('$1 wrote:');
1504 // handle nested quotes
1506 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1507 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1508 "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
1512 // Declare the format for [quote] layout
1513 $QuoteLayout = '<blockquote>$1</blockquote>';
1515 // Check for [quote] text
1516 // handle nested quotes
1518 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1519 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1522 // Check for [quote=Author] text
1524 $t_wrote = L10n::t('$1 wrote:');
1526 // handle nested quotes
1528 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1529 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1530 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1535 // [img=widthxheight]image source[/img]
1536 $text = preg_replace_callback(
1537 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1538 function ($matches) use ($simple_html) {
1539 if (strpos($matches[3], "data:image/") === 0) {
1543 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1544 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1549 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1550 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1552 $text = preg_replace_callback("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism",
1553 function ($matches) use ($simple_html) {
1554 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1555 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1556 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '">';
1561 // [img]pathtoimage[/img]
1562 $text = preg_replace_callback(
1563 "/\[img\](.*?)\[\/img\]/ism",
1564 function ($matches) use ($simple_html) {
1565 if (strpos($matches[1], "data:image/") === 0) {
1569 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1570 return "[img]" . $matches[1] . "[/img]";
1575 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1576 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1579 $text = self::convertShare(
1581 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1582 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1586 $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);
1587 $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 />', $text);
1588 //$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);
1592 $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);
1593 $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);
1595 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1596 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1598 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1599 '<a href="$1" target="_blank">$1</a>', $text);
1600 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1601 '<a href="$1" target="_blank">$1</a>', $text);
1604 // html5 video and audio
1608 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1610 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1613 // Youtube extensions
1615 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1616 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1617 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1620 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1621 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1622 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1625 $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);
1627 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1628 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1632 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1633 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1636 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1637 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1640 $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);
1642 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1643 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1647 $text = OEmbed::BBCode2HTML($text);
1649 // Avoid triple linefeeds through oembed
1650 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1652 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1653 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1654 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1655 // start which is always required). Allow desc with a missing summary for compatibility.
1657 if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) {
1658 $sub = Event::getHTML($ev, $simple_html);
1660 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1661 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1662 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1663 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1664 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1665 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1666 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1669 // Replace non graphical smilies for external posts
1671 $text = Smilies::replace($text, false, true);
1674 // Unhide all [noparse] contained bbtags unspacefying them
1675 // and triming the [noparse] tag.
1677 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1678 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1679 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1681 /// @todo What is the meaning of these lines?
1682 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1683 $text = preg_replace('/\&\#039\;/', '\'', $text);
1685 // Currently deactivated, it made problems with " inside of alt texts.
1686 //$text = preg_replace('/\"\;/', '"', $text);
1688 // fix any escaped ampersands that may have been converted into links
1689 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1691 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1692 $allowed_src_protocols = ['http', 'redir', 'cid'];
1693 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1694 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
1696 // sanitize href attributes (only whitelisted protocols URLs)
1697 // default value for backward compatibility
1698 $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']);
1700 // Always allowed protocol even if config isn't set or not including it
1701 $allowed_link_protocols[] = 'http';
1702 $allowed_link_protocols[] = 'redir/';
1704 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1705 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
1708 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1711 // Restore code blocks
1712 $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
1713 function ($matches) use ($codeblocks) {
1714 $return = $matches[0];
1715 if (isset($codeblocks[intval($matches[1])])) {
1716 $return = $codeblocks[$matches[1]];
1723 // Clean up the HTML by loading and saving the HTML with the DOM.
1724 // Bad structured html can break a whole page.
1725 // For performance reasons do it only with ativated item cache or at export.
1726 if (!$try_oembed || (get_itemcachepath() != "")) {
1727 $doc = new DOMDocument();
1728 $doc->preserveWhiteSpace = false;
1730 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1732 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1733 $encoding = '<?xml encoding="UTF-8">';
1734 @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
1735 $doc->encoding = 'UTF-8';
1736 $text = $doc->saveHTML();
1737 $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
1739 $text = str_replace('<br></li>', '</li>', $text);
1741 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1744 // Clean up some useless linebreaks in lists
1745 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1746 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1747 //$Text = str_replace('</li><br />', '</li>', $Text);
1748 //$Text = str_replace('<br /><li>', '<li>', $Text);
1749 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1751 Addon::callHooks('bbcode', $text);
1757 * @brief Strips the "abstract" tag from the provided text
1759 * @param string $text The text with BBCode
1760 * @return string The same text - but without "abstract" element
1762 public static function stripAbstract($text)
1764 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1765 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1771 * @brief Returns the value of the "abstract" element
1773 * @param string $text The text that maybe contains the element
1774 * @param string $addon The addon for which the abstract is meant for
1775 * @return string The abstract
1777 public static function getAbstract($text, $addon = "")
1781 $addon = strtolower($addon);
1783 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1784 foreach ($results AS $result) {
1785 $abstracts[strtolower($result[1])] = $result[2];
1789 if (isset($abstracts[$addon])) {
1790 $abstract = $abstracts[$addon];
1793 if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1794 $abstract = $result[1];
1801 * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
1803 * @param array $match Matching values for the callback
1804 * @return string Replaced mention
1806 private static function bbCodeMention2DiasporaCallback($match)
1808 $contact = Contact::getDetailsByURL($match[3]);
1810 if (empty($contact['addr'])) {
1811 $contact = Probe::uri($match[3]);
1814 if (empty($contact['addr'])) {
1818 $mention = '@{' . $match[2] . '; ' . $contact['addr'] . '}';
1823 * @brief Converts a BBCode text into Markdown
1825 * This function converts a BBCode item body to be sent to Markdown-enabled
1826 * systems like Diaspora and Libertree
1828 * @param string $text
1829 * @param bool $for_diaspora Diaspora requires more changes than Libertree
1832 public static function toMarkdown($text, $for_diaspora = true)
1834 $a = self::getApp();
1836 $original_text = $text;
1838 // Since Diaspora is creating a summary for links, this function removes them before posting
1839 if ($for_diaspora) {
1840 $text = self::removeShareInformation($text);
1844 * Transform #tags, strip off the [url] and replace spaces with underscore
1846 $url_search_string = "^\[\]";
1847 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
1848 function ($matches) {
1849 return '#' . str_replace(' ', '_', $matches[2]);
1854 // Converting images with size parameters to simple images. Markdown doesn't know it.
1855 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
1857 // Convert it to HTML - don't try oembed
1858 if ($for_diaspora) {
1859 $text = self::convert($text, false, 3);
1861 // Add all tags that maybe were removed
1862 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
1864 foreach ($tags[2] as $tag) {
1865 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
1866 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
1867 $tagline .= '#' . $tag . ' ';
1870 $text = $text . " " . $tagline;
1873 $text = self::convert($text, false, 4);
1876 // mask some special HTML chars from conversation to markdown
1877 $text = str_replace(['<', '>', '&'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text);
1879 // If a link is followed by a quote then there should be a newline before it
1880 // Maybe we should make this newline at every time before a quote.
1881 $text = str_replace(["</a><blockquote>"], ["</a><br><blockquote>"], $text);
1883 $stamp1 = microtime(true);
1885 // Now convert HTML to Markdown
1886 $text = HTML::toMarkdown($text);
1888 // unmask the special chars back to HTML
1889 $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
1891 $a->saveTimestamp($stamp1, "parser");
1893 // Libertree has a problem with escaped hashtags.
1894 $text = str_replace(['\#'], ['#'], $text);
1896 // Remove any leading or trailing whitespace, as this will mess up
1897 // the Diaspora signature verification and cause the item to disappear
1898 $text = trim($text);
1900 if ($for_diaspora) {
1901 $url_search_string = "^\[\]";
1902 $text = preg_replace_callback(
1903 "/([@]\[(.*?)\])\(([$url_search_string]*?)\)/ism",
1904 ['self', 'bbCodeMention2DiasporaCallback'],
1909 Addon::callHooks('bb2diaspora', $text);