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\Cache;
15 use Friendica\Core\Config;
16 use Friendica\Core\Hook;
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;
30 use Friendica\Util\Strings;
31 use Friendica\Util\XML;
33 class BBCode extends BaseObject
36 * @brief Fetches attachment data that were generated the old way
38 * @param string $body Message body
40 * 'type' -> Message type ("link", "video", "photo")
41 * 'text' -> Text before the shared message
42 * 'after' -> Text after the shared message
43 * 'image' -> Preview image of the message
44 * 'url' -> Url to the attached message
45 * 'title' -> Title of the attachment
46 * 'description' -> Description of the attachment
47 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
49 private static function getOldAttachmentData($body)
53 // Simplify image codes
54 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
56 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
57 foreach ($attached as $data) {
58 if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
62 $post["type"] = substr($data[1], 5);
64 $pos = strpos($body, $data[0]);
66 $post["text"] = trim(substr($body, 0, $pos));
67 $post["after"] = trim(substr($body, $pos + strlen($data[0])));
69 $post["text"] = trim(str_replace($data[0], "", $body));
73 $attacheddata = $data[2];
75 $URLSearchString = "^\[\]";
77 if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
79 $picturedata = Image::getInfoFromURL($matches[1]);
82 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
83 $post["image"] = $matches[1];
85 $post["preview"] = $matches[1];
90 if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
91 $post["url"] = $matches[1];
92 $post["title"] = $matches[2];
94 if (!empty($post["url"]) && (in_array($post["type"], ["link", "video"]))
95 && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
96 $post["url"] = $matches[1];
99 // Search for description
100 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
101 $post["description"] = $matches[1];
109 * @brief Fetches attachment data that were generated with the "attachment" element
111 * @param string $body Message body
113 * 'type' -> Message type ("link", "video", "photo")
114 * 'text' -> Text before the shared message
115 * 'after' -> Text after the shared message
116 * 'image' -> Preview image of the message
117 * 'url' -> Url to the attached message
118 * 'title' -> Title of the attachment
119 * 'description' -> Description of the attachment
120 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
122 public static function getAttachmentData($body)
126 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
127 return self::getOldAttachmentData($body);
130 $attributes = $match[2];
132 $data["text"] = trim($match[1]);
135 preg_match("/type='(.*?)'/ism", $attributes, $matches);
136 if (!empty($matches[1])) {
137 $type = strtolower($matches[1]);
140 preg_match('/type="(.*?)"/ism', $attributes, $matches);
141 if (!empty($matches[1])) {
142 $type = strtolower($matches[1]);
149 if (!in_array($type, ["link", "audio", "photo", "video"])) {
154 $data["type"] = $type;
158 preg_match("/url='(.*?)'/ism", $attributes, $matches);
159 if (!empty($matches[1])) {
163 preg_match('/url="(.*?)"/ism', $attributes, $matches);
164 if (!empty($matches[1])) {
169 $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
173 preg_match("/title='(.*?)'/ism", $attributes, $matches);
174 if (!empty($matches[1])) {
175 $title = $matches[1];
178 preg_match('/title="(.*?)"/ism', $attributes, $matches);
179 if (!empty($matches[1])) {
180 $title = $matches[1];
184 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
185 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
186 $title = str_replace(["[", "]"], ["[", "]"], $title);
187 $data["title"] = $title;
191 preg_match("/image='(.*?)'/ism", $attributes, $matches);
192 if (!empty($matches[1])) {
193 $image = $matches[1];
196 preg_match('/image="(.*?)"/ism', $attributes, $matches);
197 if (!empty($matches[1])) {
198 $image = $matches[1];
202 $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
206 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
207 if (!empty($matches[1])) {
208 $preview = $matches[1];
211 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
212 if (!empty($matches[1])) {
213 $preview = $matches[1];
216 if ($preview != "") {
217 $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
220 $data["description"] = trim($match[3]);
222 $data["after"] = trim($match[4]);
227 public static function getAttachedData($body, $item = [])
231 - type: link, video, photo
239 $has_title = !empty($item['title']);
240 $plink = defaults($item, 'plink', '');
241 $post = self::getAttachmentData($body);
243 // if nothing is found, it maybe having an image.
244 if (!isset($post["type"])) {
245 // Simplify image codes
246 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
248 $URLSearchString = "^\[\]";
250 $body = preg_replace("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
252 if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
253 if ((count($pictures) == 1) && !$has_title) {
254 // Checking, if the link goes to a picture
255 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
258 // Sometimes photo posts to the own album are not detected at the start.
259 // So we seem to cannot use the cache for these cases. That's strange.
260 if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
261 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
264 if ($data["type"] == "photo") {
265 $post["type"] = "photo";
266 if (isset($data["images"][0])) {
267 $post["image"] = $data["images"][0]["src"];
268 $post["url"] = $data["url"];
270 $post["image"] = $data["url"];
273 $post["preview"] = $pictures[0][2];
274 $post["text"] = str_replace($pictures[0][0], "", $body);
276 $imgdata = Image::getInfoFromURL($pictures[0][1]);
277 if ($imgdata && substr($imgdata["mime"], 0, 6) == "image/") {
278 $post["type"] = "photo";
279 $post["image"] = $pictures[0][1];
280 $post["preview"] = $pictures[0][2];
281 $post["text"] = str_replace($pictures[0][0], "", $body);
284 } elseif (count($pictures) > 0) {
285 $post["type"] = "link";
286 $post["url"] = $plink;
287 $post["image"] = $pictures[0][2];
288 $post["text"] = $body;
290 } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
291 if ((count($pictures) == 1) && !$has_title) {
292 $post["type"] = "photo";
293 $post["image"] = $pictures[0][1];
294 $post["text"] = str_replace($pictures[0][0], "", $body);
295 } elseif (count($pictures) > 0) {
296 $post["type"] = "link";
297 $post["url"] = $plink;
298 $post["image"] = $pictures[0][1];
299 $post["text"] = $body;
303 // Test for the external links
304 preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
305 preg_match_all("(\[url\=([$URLSearchString]*)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
307 $links = array_merge($links1, $links2);
309 // If there is only a single one, then use it.
310 // This should cover link posts via API.
311 if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
312 $post["type"] = "link";
313 $post["text"] = trim($body);
314 $post["url"] = $links[0][1];
317 // Now count the number of external media links
318 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
319 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
320 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
321 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
323 // Add them to the other external links
324 $links = array_merge($links, $links1, $links2, $links3, $links4);
326 // Are there more than one?
327 if (count($links) > 1) {
328 // The post will be the type "text", which means a blog post
329 unset($post["type"]);
330 $post["url"] = $plink;
333 if (!isset($post["type"])) {
334 $post["type"] = "text";
335 $post["text"] = trim($body);
337 } elseif (isset($post["url"]) && ($post["type"] == "video")) {
338 $data = ParseUrl::getSiteinfoCached($post["url"], true);
340 if (isset($data["images"][0])) {
341 $post["image"] = $data["images"][0]["src"];
349 * @brief Converts a BBCode text into plaintext
352 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
356 public static function toPlaintext($text, $keep_urls = true)
358 $naked_text = preg_replace('/\[(.+?)\]\s*/','', $text);
360 $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text);
366 private static function proxyUrl($image, $simplehtml = false)
368 // Only send proxied pictures to API and for internal display
369 if (in_array($simplehtml, [false, 2])) {
370 return ProxyUtils::proxifyUrl($image);
376 public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
378 // Suppress "view full size"
379 if (intval(Config::get('system', 'no_view_full_size'))) {
380 $include_link = false;
383 // Picture addresses can contain special characters
384 $s = htmlspecialchars_decode($srctext);
387 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
389 foreach ($matches as $mtch) {
390 Logger::log('scale_external_image: ' . $mtch[1]);
392 $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
393 if (stristr($mtch[1], $hostname)) {
397 // $scale_replace, if passed, is an array of two elements. The
398 // first is the name of the full-size image. The second is the
399 // name of a remote, scaled-down version of the full size image.
400 // This allows Friendica to display the smaller remote image if
401 // one exists, while still linking to the full-size image
402 if ($scale_replace) {
403 $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
407 $i = Network::fetchUrl($scaled);
412 // guess mimetype from headers or filename
413 $type = Image::guessType($mtch[1], true);
416 $Image = new Image($i, $type);
417 if ($Image->isValid()) {
418 $orig_width = $Image->getWidth();
419 $orig_height = $Image->getHeight();
421 if ($orig_width > 640 || $orig_height > 640) {
422 $Image->scaleDown(640);
423 $new_width = $Image->getWidth();
424 $new_height = $Image->getHeight();
425 Logger::log('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], Logger::DEBUG);
428 '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
429 . "\n" . (($include_link)
430 ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
434 Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG);
441 // replace the special char encoding
442 $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
447 * The purpose of this function is to apply system message length limits to
448 * imported messages without including any embedded photos in the length
450 * @brief Truncates imported message body string length to max_import_size
451 * @param string $body
453 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
455 public static function limitBodySize($body)
457 $maxlen = Config::get('config', 'max_import_size', 0);
459 // If the length of the body, including the embedded images, is smaller
460 // than the maximum, then don't waste time looking for the images
461 if ($maxlen && (strlen($body) > $maxlen)) {
463 Logger::log('the total body length exceeds the limit', Logger::DEBUG);
469 $img_start = strpos($orig_body, '[img');
470 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
471 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
472 while (($img_st_close !== false) && ($img_end !== false)) {
474 $img_st_close++; // make it point to AFTER the closing bracket
475 $img_end += $img_start;
476 $img_end += strlen('[/img]');
478 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
479 // This is an embedded image
481 if (($textlen + $img_start) > $maxlen) {
482 if ($textlen < $maxlen) {
483 Logger::log('the limit happens before an embedded image', Logger::DEBUG);
484 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
488 $new_body = $new_body . substr($orig_body, 0, $img_start);
489 $textlen += $img_start;
492 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
495 if (($textlen + $img_end) > $maxlen) {
496 if ($textlen < $maxlen) {
497 Logger::log('the limit happens before the end of a non-embedded image', Logger::DEBUG);
498 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
502 $new_body = $new_body . substr($orig_body, 0, $img_end);
503 $textlen += $img_end;
506 $orig_body = substr($orig_body, $img_end);
508 if ($orig_body === false) {
509 // in case the body ends on a closing image tag
513 $img_start = strpos($orig_body, '[img');
514 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
515 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
518 if (($textlen + strlen($orig_body)) > $maxlen) {
519 if ($textlen < $maxlen) {
520 Logger::log('the limit happens after the end of the last image', Logger::DEBUG);
521 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
524 Logger::log('the text size with embedded images extracted did not violate the limit', Logger::DEBUG);
525 $new_body = $new_body . $orig_body;
535 * Processes [attachment] tags
537 * Note: Can produce a [bookmark] tag in the returned string
539 * @brief Processes [attachment] tags
540 * @param string $return
541 * @param bool|int $simplehtml
542 * @param bool $tryoembed
544 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
546 private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
548 $data = self::getAttachmentData($return);
549 if (empty($data) || empty($data["url"])) {
553 if (isset($data["title"])) {
554 $data["title"] = strip_tags($data["title"]);
555 $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
557 $data["title"] = null;
560 if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data["image"])) {
561 $data["preview"] = $data["image"];
566 if ($simplehtml == 7) {
567 $return = self::convertUrlForOStatus($data["url"]);
568 } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
569 $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data["url"], $data["title"]);
572 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
573 $return = OEmbed::getHTML($data['url'], $data['title']);
575 throw new Exception('OEmbed is disabled for this attachment.');
577 } catch (Exception $e) {
578 $data["title"] = defaults($data, 'title', $data['url']);
580 if ($simplehtml != 4) {
581 $return = sprintf('<div class="type-%s">', $data["type"]);
584 if (!empty($data['title']) && !empty($data['url'])) {
585 if (!empty($data["image"]) && empty($data["text"]) && ($data["type"] == "photo")) {
586 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
588 if (!empty($data["image"])) {
589 $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), $data["title"]);
590 } elseif (!empty($data["preview"])) {
591 $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"]);
593 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
597 if (!empty($data["description"]) && $data["description"] != $data["title"]) {
598 // Sanitize the HTML by converting it to BBCode
599 $bbcode = HTML::toBBCode($data["description"]);
600 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
603 if (!empty($data['url'])) {
604 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
607 if ($simplehtml != 4) {
613 return trim(defaults($data, 'text', '') . ' ' . $return . ' ' . defaults($data, 'after', ''));
616 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
618 $data = self::getAttachmentData($Text);
623 return $data["text"] . defaults($data, 'after', '');
626 $title = htmlentities(defaults($data, 'title', ''), ENT_QUOTES, 'UTF-8', false);
627 $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
628 if ($plaintext || (($title != "") && strstr($text, $title))) {
629 $data["title"] = $data["url"];
630 } elseif (($text != "") && strstr($title, $text)) {
631 $data["text"] = $data["title"];
632 $data["title"] = $data["url"];
635 if (empty($data["text"]) && !empty($data["title"]) && empty($data["url"])) {
636 return $data["title"] . $data["after"];
639 // If the link already is included in the post, don't add it again
640 if (!empty($data["url"]) && strpos($data["text"], $data["url"])) {
641 return $data["text"] . $data["after"];
644 $text = $data["text"];
646 if (!empty($data["url"]) && !empty($data["title"])) {
647 $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
648 } elseif (!empty($data["url"])) {
649 $text .= "\n[url]" . $data["url"] . "[/url]";
652 return $text . "\n" . $data["after"];
656 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
658 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
659 * @param array $match Array with the matching values
660 * @return string reformatted link including HTML codes
662 private static function convertUrlForOStatusCallback($match)
666 if (isset($match[2]) && ($match[1] != $match[2])) {
670 $parts = parse_url($url);
671 if (!isset($parts['scheme'])) {
675 return self::convertUrlForOStatus($url);
679 * @brief Converts [url] BBCodes in a format that looks fine on OStatus systems.
680 * @param string $url URL that is about to be reformatted
681 * @return string reformatted link including HTML codes
683 private static function convertUrlForOStatus($url)
685 $parts = parse_url($url);
686 $scheme = $parts['scheme'] . '://';
687 $styled_url = str_replace($scheme, '', $url);
689 if (strlen($styled_url) > 30) {
690 $styled_url = substr($styled_url, 0, 30) . "…";
693 $html = '<a href="%s" target="_blank">%s</a>';
695 return sprintf($html, $url, $styled_url);
699 * [noparse][i]italic[/i][/noparse] turns into
700 * [noparse][ i ]italic[ /i ][/noparse],
701 * to hide them from parser.
703 private static function escapeNoparseCallback($match)
705 $whole_match = $match[0];
706 $captured = $match[1];
707 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
708 $new_str = str_replace($captured, $spacefied, $whole_match);
713 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
714 * now turns back and the [noparse] tags are trimed
715 * returning [i]italic[/i]
717 private static function unescapeNoparseCallback($match)
719 $captured = $match[1];
720 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
725 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
728 * @param string $text Text to search
729 * @param string $name Tag name
730 * @param int $occurrences Number of first occurrences to skip
731 * @return boolean|array
733 public static function getTagPosition($text, $name, $occurrences = 0)
735 if ($occurrences < 0) {
740 for ($i = 0; $i <= $occurrences; $i++) {
741 if ($start_open !== false) {
742 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
746 if ($start_open === false) {
750 $start_equal = strpos($text, '=', $start_open);
751 $start_close = strpos($text, ']', $start_open);
753 if ($start_close === false) {
759 $end_open = strpos($text, '[/' . $name . ']', $start_close);
761 if ($end_open === false) {
767 'open' => $start_open,
768 'close' => $start_close
772 'close' => $end_open + strlen('[/' . $name . ']')
776 if ($start_equal !== false) {
777 $res['start']['equal'] = $start_equal + 1;
784 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
786 * @param string $pattern Preg pattern string
787 * @param string $replace Preg replace string
788 * @param string $name BBCode tag name
789 * @param string $text Text to search
792 public static function pregReplaceInTag($pattern, $replace, $name, $text)
795 $pos = self::getTagPosition($text, $name, $occurrences);
796 while ($pos !== false && $occurrences++ < 1000) {
797 $start = substr($text, 0, $pos['start']['open']);
798 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
799 $end = substr($text, $pos['end']['close']);
800 if ($end === false) {
804 $subject = preg_replace($pattern, $replace, $subject);
805 $text = $start . $subject . $end;
807 $pos = self::getTagPosition($text, $name, $occurrences);
813 private static function extractImagesFromItemBody($body)
820 $img_start = strpos($orig_body, '[img');
821 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
822 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
823 while (($img_st_close !== false) && ($img_end !== false)) {
824 $img_st_close++; // make it point to AFTER the closing bracket
825 $img_end += $img_start;
827 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
828 // This is an embedded image
829 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
830 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
834 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
837 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
839 if ($orig_body === false) {
840 // in case the body ends on a closing image tag
844 $img_start = strpos($orig_body, '[img');
845 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
846 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
849 $new_body = $new_body . $orig_body;
851 return ['body' => $new_body, 'images' => $saved_image];
854 private static function interpolateSavedImagesIntoItemBody($body, array $images)
859 foreach ($images as $image) {
860 // We're depending on the property of 'foreach' (specified on the PHP website) that
861 // it loops over the array starting from the first element and going sequentially
862 // to the last element
863 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
864 '<img src="' . self::proxyUrl($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
872 * This function converts a [share] block to text according to a provided callback function whose signature is:
874 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
877 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
879 * - $author_contact is a contact record array
880 * - $content is the inner content of the [share] block
881 * - $is_quote_share indicates whether there's any content before the [share] block
882 * - Return value is the string that should replace the [share] block in the provided text
884 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
886 * @param string $text A BBCode string
887 * @param callable $callback
888 * @return string The BBCode string with all [share] blocks replaced
890 public static function convertShare($text, callable $callback)
892 $return = preg_replace_callback(
893 "/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
894 function ($match) use ($callback) {
895 $attribute_string = $match[2];
898 foreach(['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
899 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
900 $attributes[$field] = html_entity_decode(defaults($matches, 2, ''), ENT_QUOTES, 'UTF-8');
903 // We only call this so that a previously unknown contact can be added.
904 // This is important for the function "Model\Contact::getDetailsByURL()".
905 // This function then can fetch an entry from the contact table.
906 Contact::getIdForURL($attributes['profile'], 0, true);
908 $author_contact = Contact::getDetailsByURL($attributes['profile']);
909 $author_contact['addr'] = defaults($author_contact, 'addr' , Protocol::getAddrFromProfileUrl($attributes['profile']));
911 $attributes['author'] = defaults($author_contact, 'name' , $attributes['author']);
912 $attributes['avatar'] = defaults($author_contact, 'micro', $attributes['avatar']);
913 $attributes['profile'] = defaults($author_contact, 'url' , $attributes['profile']);
915 if ($attributes['avatar']) {
916 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
919 return $match[1] . $callback($attributes, $author_contact, $match[3], trim($match[1]) != '');
928 * Default [share] tag conversion callback
930 * Note: Can produce a [bookmark] tag in the output
932 * @see BBCode::convertShare()
933 * @param array $attributes [share] block attribute values
934 * @param array $author_contact Contact row of the shared author
935 * @param string $content Inner content of the [share] block
936 * @param boolean $is_quote_share Whether there is content before the [share] block
937 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
939 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
941 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
943 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
945 switch ($simplehtml) {
947 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' <a href="' . $attributes['profile'] . '">' . $mention . '</a>: </p>' . "\n" . '«' . $content . '»';
950 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
953 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
954 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
956 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
958 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
959 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
962 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
964 if (empty($attributes['posted']) && !empty($attributes['link'])) {
965 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
971 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
972 $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
973 $headline .= ':</b></p>' . "\n";
975 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
979 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
981 case 7: // statusnet/GNU Social
982 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
985 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n";
986 $text .= '<p>' . $content . '</p>' . "\n";
988 if ($attributes['link'] != '') {
989 $text .= '<p>' . $attributes['link'] . '</p>';
993 // Transforms quoted tweets in rich attachments to avoid nested tweets
994 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) {
996 $text = ($is_quote_share? '<br />' : '') . OEmbed::getHTML($attributes['link']);
997 } catch (Exception $e) {
998 $text = ($is_quote_share? '<br />' : '') . sprintf('[bookmark=%s]%s[/bookmark]', $attributes['link'], $content);
1001 $text = ($is_quote_share? "\n" : '');
1003 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1004 $text .= Renderer::replaceMacros($tpl, [
1005 '$profile' => $attributes['profile'],
1006 '$avatar' => $attributes['avatar'],
1007 '$author' => $attributes['author'],
1008 '$link' => $attributes['link'],
1009 '$posted' => $attributes['posted'],
1010 '$content' => trim($content)
1019 private static function removePictureLinksCallback($match)
1021 $text = Cache::get($match[1]);
1023 if (is_null($text)) {
1024 $a = self::getApp();
1026 $stamp1 = microtime(true);
1028 $ch = @curl_init($match[1]);
1029 @curl_setopt($ch, CURLOPT_NOBODY, true);
1030 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1031 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1033 $curl_info = @curl_getinfo($ch);
1035 $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack());
1037 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1038 $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
1040 $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
1042 // if its not a picture then look if its a page that contains a picture link
1043 $body = Network::fetchUrl($match[1]);
1045 $doc = new DOMDocument();
1046 @$doc->loadHTML($body);
1047 $xpath = new DOMXPath($doc);
1048 $list = $xpath->query("//meta[@name]");
1049 foreach ($list as $node) {
1052 if ($node->attributes->length) {
1053 foreach ($node->attributes as $attribute) {
1054 $attr[$attribute->name] = $attribute->value;
1058 if (strtolower($attr["name"]) == "twitter:image") {
1059 $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
1063 Cache::set($match[1], $text);
1069 private static function expandLinksCallback($match)
1071 if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1072 return ($match[1] . "[url]" . $match[2] . "[/url]");
1074 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1078 private static function cleanPictureLinksCallback($match)
1080 $text = Cache::get($match[1]);
1082 if (is_null($text)) {
1083 $a = self::getApp();
1085 $stamp1 = microtime(true);
1087 $ch = @curl_init($match[1]);
1088 @curl_setopt($ch, CURLOPT_NOBODY, true);
1089 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1090 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1092 $curl_info = @curl_getinfo($ch);
1094 $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack());
1096 // if its a link to a picture then embed this picture
1097 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1098 $text = "[img]" . $match[1] . "[/img]";
1100 $text = "[img]" . $match[2] . "[/img]";
1102 // if its not a picture then look if its a page that contains a picture link
1103 $body = Network::fetchUrl($match[1]);
1105 $doc = new DOMDocument();
1106 @$doc->loadHTML($body);
1107 $xpath = new DOMXPath($doc);
1108 $list = $xpath->query("//meta[@name]");
1109 foreach ($list as $node) {
1111 if ($node->attributes->length) {
1112 foreach ($node->attributes as $attribute) {
1113 $attr[$attribute->name] = $attribute->value;
1117 if (strtolower($attr["name"]) == "twitter:image") {
1118 $text = "[img]" . $attr["content"] . "[/img]";
1122 Cache::set($match[1], $text);
1128 public static function cleanPictureLinks($text)
1130 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1135 * @brief Converts a BBCode message to HTML message
1137 * BBcode 2 HTML was written by WAY2WEB.net
1138 * extended to work with Mistpark/Friendica - Mike Macgirvin
1140 * Simple HTML values meaning:
1141 * - 0: Friendica display
1143 * - 2: Used for Google+, Windows Phone push, Friendica API
1144 * - 3: Used before converting to Markdown in bb2diaspora.php
1145 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1148 * - 7: Used for dfrn, OStatus
1149 * - 8: Used for WP backlink text setting
1151 * @param string $text
1152 * @param bool $try_oembed
1153 * @param int $simple_html
1154 * @param bool $for_plaintext
1156 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1158 public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false)
1160 $a = self::getApp();
1163 * preg_match_callback function to replace potential Oembed tags with Oembed content
1165 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1167 * $match[2] = $title or absent
1169 $try_oembed_callback = function ($match)
1172 $title = defaults($match, 2, null);
1175 $return = OEmbed::getHTML($url, $title);
1176 } catch (Exception $ex) {
1177 $return = $match[0];
1183 // Extracting multi-line code blocks before the whitespace processing
1186 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1187 function ($matches) use (&$codeblocks) {
1188 $return = $matches[0];
1189 if (strpos($matches[2], "\n") !== false) {
1190 $return = '#codeblock-' . count($codeblocks) . '#';
1192 $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . trim($matches[2], "\n\r") . '</code></pre>';
1199 // Hide all [noparse] contained bbtags by spacefying them
1200 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1202 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1203 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1204 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1206 // Remove the abstract element. It is a non visible element.
1207 $text = self::stripAbstract($text);
1209 // Move all spaces out of the tags
1210 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1211 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1213 // Extract the private images which use data urls since preg has issues with
1214 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1215 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1217 $extracted = self::extractImagesFromItemBody($text);
1218 $text = $extracted['body'];
1219 $saved_image = $extracted['images'];
1221 // If we find any event code, turn it into an event.
1222 // After we're finished processing the bbcode we'll
1223 // replace all of the event code with a reformatted version.
1225 $ev = Event::fromBBCode($text);
1227 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1228 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1230 $text = str_replace("<", "<", $text);
1231 $text = str_replace(">", ">", $text);
1233 // remove some newlines before the general conversion
1234 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1235 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1237 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1239 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1242 // Convert new line chars to html <br /> tags
1244 // nlbr seems to be hopelessly messed up
1245 // $Text = nl2br($Text);
1247 // We'll emulate it.
1249 $text = trim($text);
1250 $text = str_replace("\r\n", "\n", $text);
1252 // removing multiplicated newlines
1253 if (Config::get("system", "remove_multiplicated_lines")) {
1254 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1255 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1256 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1257 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1260 $text = str_replace($search, $replace, $text);
1261 } while ($oldtext != $text);
1264 // Set up the parameters for a URL search string
1265 $URLSearchString = "^\[\]";
1266 // Set up the parameters for a MAIL search string
1267 $MAILSearchString = $URLSearchString;
1269 // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text
1270 if (!$for_plaintext) {
1271 // Autolink feature (thanks to http://code.seebz.net/p/autolink-php/)
1272 // Currently disabled, since the function is too greedy
1273 // $autolink_regex = "`([^\]\=\"']|^)(https?\://[^\s<]+[^\s<\.\)])`ism";
1274 $autolink_regex = "/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism";
1275 $text = preg_replace($autolink_regex, '$1[url]$2[/url]', $text);
1276 if ($simple_html == 7) {
1277 $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1278 $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1281 $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text);
1282 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1286 // Handle attached links or videos
1287 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1289 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1291 // Remove all hashtag addresses
1292 if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7])) {
1293 $text = preg_replace("/([#@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $text);
1294 } elseif ($simple_html == 3) {
1295 // The ! is converted to @ since Diaspora only understands the @
1296 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1297 '@<a href="$2">$3</a>',
1299 } elseif ($simple_html == 7) {
1300 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1301 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1303 } elseif (!$simple_html) {
1304 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1305 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1309 // Bookmarks in red - will be converted to bookmarks in friendica
1310 $text = preg_replace("/#\^\[url\]([$URLSearchString]*)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1311 $text = preg_replace("/#\^\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1312 $text = preg_replace("/#\[url\=[$URLSearchString]*\]\^\[\/url\]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i",
1313 "[bookmark=$1]$2[/bookmark]", $text);
1315 if (in_array($simple_html, [2, 6, 7, 8, 9])) {
1316 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1317 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1318 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1321 if ($simple_html == 5) {
1322 $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1325 // Perform URL Search
1327 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1330 if ($simple_html == 5) {
1331 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1333 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1336 // Handle Diaspora posts
1337 $text = preg_replace_callback(
1338 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1340 return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1344 $text = preg_replace_callback(
1345 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1347 return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1351 // Server independent link to posts and comments
1352 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1353 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1354 $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
1358 * - #[url=<anything>]<term>[/url]
1359 * - [url=<anything>]#<term>[/url]
1361 $text = preg_replace_callback("/(?:#\[url\=[$URLSearchString]*\]|\[url\=[$URLSearchString]*\]#)(.*?)\[\/url\]/ism", function($matches) {
1363 . System::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1364 . '" class="tag" title="' . XML::escape($matches[1]) . '">'
1365 . XML::escape($matches[1])
1369 // We need no target="_blank" for local links
1370 // convert links start with System::baseUrl() as local link without the target="_blank" attribute
1371 $escapedBaseUrl = preg_quote(System::baseUrl(), '/');
1372 $text = preg_replace("/\[url\](".$escapedBaseUrl."[$URLSearchString]*)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1373 $text = preg_replace("/\[url\=(".$escapedBaseUrl."[$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1375 $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1376 $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1378 // Red compatibility, though the link can't be authenticated on Friendica
1379 $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1382 // we may need to restrict this further if it picks up too many strays
1383 // link acct:user@host to a webfinger profile redirector
1385 $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);
1387 // Perform MAIL Search
1388 $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1389 $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1391 // leave open the posibility of [map=something]
1392 // this is replaced in Item::prepareBody() which has knowledge of the item location
1394 if (strpos($text, '[/map]') !== false) {
1395 $text = preg_replace_callback(
1396 "/\[map\](.*?)\[\/map\]/ism",
1397 function ($match) use ($simple_html) {
1398 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1403 if (strpos($text, '[map=') !== false) {
1404 $text = preg_replace_callback(
1405 "/\[map=(.*?)\]/ism",
1406 function ($match) use ($simple_html) {
1407 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1412 if (strpos($text, '[map]') !== false) {
1413 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1416 // Check for headers
1417 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1418 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1419 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1420 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1421 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1422 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1424 // Check for paragraph
1425 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1427 // Check for bold text
1428 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1430 // Check for Italics text
1431 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1433 // Check for Underline text
1434 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1436 // Check for strike-through text
1437 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1439 // Check for over-line text
1440 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1442 // Check for colored text
1443 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1445 // Check for sized text
1446 // [size=50] --> font-size: 50px (with the unit).
1447 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1448 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1450 // Check for centered text
1451 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1453 // Check for list text
1454 $text = str_replace("[*]", "<li>", $text);
1456 // Check for style sheet commands
1457 $text = preg_replace_callback(
1458 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1460 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1465 // Check for CSS classes
1466 $text = preg_replace_callback(
1467 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1469 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1474 // handle nested lists
1477 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1478 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1479 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1480 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1481 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1482 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1483 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1484 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1485 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1486 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1487 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1488 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1489 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1490 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1493 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1494 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1495 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1496 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1498 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1499 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1501 $text = str_replace('[hr]', '<hr />', $text);
1503 // This is actually executed in Item::prepareBody()
1505 $text = str_replace('[nosmile]', '', $text);
1507 // Check for font change text
1508 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1510 // Declare the format for [code] layout
1512 $CodeLayout = '<code>$1</code>';
1513 // Check for [code] text
1514 $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text);
1516 // Declare the format for [spoiler] layout
1517 $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
1519 // Check for [spoiler] text
1520 // handle nested quotes
1522 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1523 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
1526 // Check for [spoiler=Author] text
1528 $t_wrote = L10n::t('$1 wrote:');
1530 // handle nested quotes
1532 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1533 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1534 "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
1538 // Declare the format for [quote] layout
1539 $QuoteLayout = '<blockquote>$1</blockquote>';
1541 // Check for [quote] text
1542 // handle nested quotes
1544 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1545 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1548 // Check for [quote=Author] text
1550 $t_wrote = L10n::t('$1 wrote:');
1552 // handle nested quotes
1554 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1555 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1556 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1561 // [img=widthxheight]image source[/img]
1562 $text = preg_replace_callback(
1563 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1564 function ($matches) use ($simple_html) {
1565 if (strpos($matches[3], "data:image/") === 0) {
1569 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1570 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1575 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1576 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1578 $text = preg_replace_callback("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism",
1579 function ($matches) use ($simple_html) {
1580 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1581 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1582 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '">';
1587 // [img]pathtoimage[/img]
1588 $text = preg_replace_callback(
1589 "/\[img\](.*?)\[\/img\]/ism",
1590 function ($matches) use ($simple_html) {
1591 if (strpos($matches[1], "data:image/") === 0) {
1595 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1596 return "[img]" . $matches[1] . "[/img]";
1601 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1602 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1605 $text = self::convertShare(
1607 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1608 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1612 $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);
1613 $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 />', $text);
1614 //$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);
1618 $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);
1619 $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);
1621 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1622 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1624 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1625 '<a href="$1" target="_blank">$1</a>', $text);
1626 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1627 '<a href="$1" target="_blank">$1</a>', $text);
1630 // html5 video and audio
1634 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1636 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1639 // Youtube extensions
1641 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1642 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1643 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1646 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1647 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1648 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1651 $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);
1653 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1654 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1658 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1659 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1662 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1663 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1666 $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);
1668 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1669 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1673 $text = OEmbed::BBCode2HTML($text);
1675 // Avoid triple linefeeds through oembed
1676 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1678 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1679 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1680 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1681 // start which is always required). Allow desc with a missing summary for compatibility.
1683 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1684 $sub = Event::getHTML($ev, $simple_html);
1686 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1687 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1688 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1689 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1690 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1691 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1692 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1695 // Replace non graphical smilies for external posts
1697 $text = Smilies::replace($text);
1700 // Unhide all [noparse] contained bbtags unspacefying them
1701 // and triming the [noparse] tag.
1703 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1704 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1705 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1707 /// @todo What is the meaning of these lines?
1708 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1709 $text = preg_replace('/\&\#039\;/', '\'', $text);
1711 // Currently deactivated, it made problems with " inside of alt texts.
1712 //$text = preg_replace('/\"\;/', '"', $text);
1714 // fix any escaped ampersands that may have been converted into links
1715 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1717 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1718 $allowed_src_protocols = ['http', 'redir', 'cid'];
1719 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1720 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
1722 // sanitize href attributes (only whitelisted protocols URLs)
1723 // default value for backward compatibility
1724 $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']);
1726 // Always allowed protocol even if config isn't set or not including it
1727 $allowed_link_protocols[] = 'http';
1728 $allowed_link_protocols[] = 'redir/';
1730 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1731 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
1734 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1737 // Restore code blocks
1738 $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
1739 function ($matches) use ($codeblocks) {
1740 $return = $matches[0];
1741 if (isset($codeblocks[intval($matches[1])])) {
1742 $return = $codeblocks[$matches[1]];
1749 // Clean up the HTML by loading and saving the HTML with the DOM.
1750 // Bad structured html can break a whole page.
1751 // For performance reasons do it only with activated item cache or at export.
1752 if (!$try_oembed || (get_itemcachepath() != "")) {
1753 $doc = new DOMDocument();
1754 $doc->preserveWhiteSpace = false;
1756 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1758 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1759 $encoding = '<?xml encoding="UTF-8">';
1760 @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
1761 $doc->encoding = 'UTF-8';
1762 $text = $doc->saveHTML();
1763 $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
1765 $text = str_replace('<br></li>', '</li>', $text);
1767 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1770 // Clean up some useless linebreaks in lists
1771 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1772 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1773 //$Text = str_replace('</li><br />', '</li>', $Text);
1774 //$Text = str_replace('<br /><li>', '<li>', $Text);
1775 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1777 Hook::callAll('bbcode', $text);
1783 * @brief Strips the "abstract" tag from the provided text
1785 * @param string $text The text with BBCode
1786 * @return string The same text - but without "abstract" element
1788 public static function stripAbstract($text)
1790 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1791 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1797 * @brief Returns the value of the "abstract" element
1799 * @param string $text The text that maybe contains the element
1800 * @param string $addon The addon for which the abstract is meant for
1801 * @return string The abstract
1803 public static function getAbstract($text, $addon = "")
1807 $addon = strtolower($addon);
1809 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1810 foreach ($results AS $result) {
1811 $abstracts[strtolower($result[1])] = $result[2];
1815 if (isset($abstracts[$addon])) {
1816 $abstract = $abstracts[$addon];
1819 if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1820 $abstract = $result[1];
1827 * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
1829 * @param array $match Matching values for the callback
1830 * [1] = Mention type (! or @)
1833 * @return string Replaced mention
1834 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1835 * @throws \ImagickException
1837 private static function bbCodeMention2DiasporaCallback($match)
1839 $contact = Contact::getDetailsByURL($match[3]);
1841 if (empty($contact['addr'])) {
1842 $contact = Probe::uri($match[3]);
1845 if (empty($contact['addr'])) {
1849 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
1854 * @brief Converts a BBCode text into Markdown
1856 * This function converts a BBCode item body to be sent to Markdown-enabled
1857 * systems like Diaspora and Libertree
1859 * @param string $text
1860 * @param bool $for_diaspora Diaspora requires more changes than Libertree
1862 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1864 public static function toMarkdown($text, $for_diaspora = true)
1866 $a = self::getApp();
1868 $original_text = $text;
1870 // Since Diaspora is creating a summary for links, this function removes them before posting
1871 if ($for_diaspora) {
1872 $text = self::removeShareInformation($text);
1876 * Transform #tags, strip off the [url] and replace spaces with underscore
1878 $url_search_string = "^\[\]";
1879 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
1880 function ($matches) {
1881 return '#' . str_replace(' ', '_', $matches[2]);
1886 // Converting images with size parameters to simple images. Markdown doesn't know it.
1887 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
1889 // Convert it to HTML - don't try oembed
1890 if ($for_diaspora) {
1891 $text = self::convert($text, false, 3);
1893 // Add all tags that maybe were removed
1894 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
1896 foreach ($tags[2] as $tag) {
1897 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
1898 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
1899 $tagline .= '#' . $tag . ' ';
1902 $text = $text . " " . $tagline;
1905 $text = self::convert($text, false, 4);
1908 // mask some special HTML chars from conversation to markdown
1909 $text = str_replace(['<', '>', '&'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text);
1911 // If a link is followed by a quote then there should be a newline before it
1912 // Maybe we should make this newline at every time before a quote.
1913 $text = str_replace(["</a><blockquote>"], ["</a><br><blockquote>"], $text);
1915 $stamp1 = microtime(true);
1917 // Now convert HTML to Markdown
1918 $text = HTML::toMarkdown($text);
1920 // unmask the special chars back to HTML
1921 $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
1923 $a->getProfiler()->saveTimestamp($stamp1, "parser", System::callstack());
1925 // Libertree has a problem with escaped hashtags.
1926 $text = str_replace(['\#'], ['#'], $text);
1928 // Remove any leading or trailing whitespace, as this will mess up
1929 // the Diaspora signature verification and cause the item to disappear
1930 $text = trim($text);
1932 if ($for_diaspora) {
1933 $url_search_string = "^\[\]";
1934 $text = preg_replace_callback(
1935 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
1936 ['self', 'bbCodeMention2DiasporaCallback'],
1941 Hook::callAll('bb2diaspora', $text);
1947 * @brief Pull out all #hashtags and @person tags from $string.
1949 * We also get @person@domain.com - which would make
1950 * the regex quite complicated as tags can also
1951 * end a sentence. So we'll run through our results
1952 * and strip the period from any tags which end with one.
1953 * Returns array of tags found, or empty array.
1955 * @param string $string Post content
1957 * @return array List of tag and person names
1959 public static function getTags($string)
1963 // Convert hashtag links to hashtags
1964 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2', $string);
1966 // ignore anything in a code block
1967 $string = preg_replace('/\[code.*?\].*?\[\/code\]/sm', '', $string);
1969 // Force line feeds at bbtags
1970 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
1972 // ignore anything in a bbtag
1973 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
1975 // Match full names against @tags including the space between first and last
1976 // We will look these up afterward to see if they are full names or not recognisable.
1978 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
1979 foreach ($matches[1] as $match) {
1980 if (strstr($match, ']')) {
1981 // we might be inside a bbcode color tag - leave it alone
1985 if (substr($match, -1, 1) === '.') {
1986 $ret[] = substr($match, 0, -1);
1993 // Otherwise pull out single word tags. These can be @nickname, @first_last
1996 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
1997 foreach ($matches[1] as $match) {
1998 if (strstr($match, ']')) {
1999 // we might be inside a bbcode color tag - leave it alone
2002 if (substr($match, -1, 1) === '.') {
2003 $match = substr($match,0,-1);
2005 // ignore strictly numeric tags like #1
2006 if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
2009 // try not to catch url fragments
2010 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {