3 * @copyright Copyright (C) 2010-2023, the Friendica project
5 * @license GNU AGPL version 3 or any later version
7 * This program is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU Affero General Public License as
9 * published by the Free Software Foundation, either version 3 of the
10 * License, or (at your option) any later version.
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU Affero General Public License for more details.
17 * You should have received a copy of the GNU Affero General Public License
18 * along with this program. If not, see <https://www.gnu.org/licenses/>.
22 namespace Friendica\Content\Text;
27 use Friendica\Content\ContactSelector;
28 use Friendica\Content\Item;
29 use Friendica\Content\OEmbed;
30 use Friendica\Content\PageInfo;
31 use Friendica\Content\Smilies;
32 use Friendica\Core\Hook;
33 use Friendica\Core\Logger;
34 use Friendica\Core\Protocol;
35 use Friendica\Core\Renderer;
37 use Friendica\Model\Contact;
38 use Friendica\Model\Event;
39 use Friendica\Model\Post;
40 use Friendica\Model\Tag;
41 use Friendica\Network\HTTPClient\Client\HttpClientAccept;
42 use Friendica\Network\HTTPClient\Client\HttpClientOptions;
43 use Friendica\Util\Map;
44 use Friendica\Util\ParseUrl;
45 use Friendica\Util\Proxy;
46 use Friendica\Util\Strings;
47 use Friendica\Util\XML;
51 // Update this value to the current date whenever changes are made to BBCode::convert
52 const VERSION = '2021-07-28';
56 const MASTODON_API = 2;
59 const TWITTER_API = 5;
64 const ACTIVITYPUB = 9;
67 const SHARED_ANCHOR = '<hr class="shared-anchor">';
68 const TOP_ANCHOR = '<br class="top-anchor">';
69 const BOTTOM_ANCHOR = '<br class="button-anchor">';
71 const PREVIEW_NONE = 0;
72 const PREVIEW_NO_IMAGE = 1;
73 const PREVIEW_LARGE = 2;
74 const PREVIEW_SMALL = 3;
77 * Fetches attachment data that were generated with the "attachment" element
79 * @param string $body Message body
81 * 'type' -> Message type ('link', 'video', 'photo')
82 * 'text' -> Text before the shared message
83 * 'after' -> Text after the shared message
84 * 'image' -> Preview image of the message
85 * 'url' -> Url to the attached message
86 * 'title' -> Title of the attachment
87 * 'description' -> Description of the attachment
88 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
90 public static function getAttachmentData(string $body): array
92 DI::profiler()->startRecording('rendering');
101 'provider_name' => '',
102 'provider_url' => '',
107 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
108 DI::profiler()->stopRecording();
112 $attributes = $match[2];
114 $data['text'] = trim($match[1]);
116 foreach (['type', 'url', 'title', 'image', 'preview', 'publisher_name', 'publisher_url', 'author_name', 'author_url'] as $field) {
117 preg_match('/' . preg_quote($field, '/') . '=("|\')(.*?)\1/ism', $attributes, $matches);
118 $value = $matches[2] ?? '';
122 case 'publisher_name':
123 $data['provider_name'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
126 case 'publisher_url':
127 $data['provider_url'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
131 $data['author_name'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
132 if ($data['provider_name'] == $data['author_name']) {
133 $data['author_name'] = '';
138 $data['author_url'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
139 if ($data['provider_url'] == $data['author_url']) {
140 $data['author_url'] = '';
145 $value = self::toPlaintext(html_entity_decode($value, ENT_QUOTES, 'UTF-8'));
146 $value = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
147 $value = str_replace(['[', ']'], ['[', ']'], $value);
148 $data['title'] = $value;
151 $data[$field] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
157 if (!in_array($data['type'], ['link', 'audio', 'photo', 'video'])) {
158 DI::profiler()->stopRecording();
162 $data['description'] = trim($match[3]);
164 $data['after'] = trim($match[4]);
166 $parts = parse_url($data['url']);
167 if (!empty($parts['scheme']) && !empty($parts['host'])) {
168 if (empty($data['provider_name'])) {
169 $data['provider_name'] = $parts['host'];
171 if (empty($data['provider_url']) || empty(parse_url($data['provider_url'], PHP_URL_SCHEME))) {
172 $data['provider_url'] = $parts['scheme'] . '://' . $parts['host'];
174 if (!empty($parts['port'])) {
175 $data['provider_url'] .= ':' . $parts['port'];
180 DI::profiler()->stopRecording();
185 * Remove [attachment] BBCode and replaces it with a regular [url]
187 * @param string $body
188 * @param boolean $no_link_desc No link description
189 * @return string with replaced body
191 public static function replaceAttachment(string $body, bool $no_link_desc = false): string
193 return preg_replace_callback(
194 "/\s*\[attachment (.*?)\](.*?)\[\/attachment\]\s*/ism",
195 function ($match) use ($body, $no_link_desc) {
196 $attach_data = self::getAttachmentData($match[0]);
197 if (empty($attach_data['url'])) {
199 } elseif (strpos(str_replace($match[0], '', $body), $attach_data['url']) !== false) {
201 } elseif (empty($attach_data['title']) || $no_link_desc) {
202 return " \n[url]" . $attach_data['url'] . "[/url]\n";
204 return " \n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
212 * Remove [attachment] BBCode
214 * @param string $body
215 * @return string with removed attachment
217 public static function removeAttachment(string $body): string
219 return trim(preg_replace("/\s*\[attachment .*?\].*?\[\/attachment\]\s*/ism", '', $body));
223 * Converts a BBCode text into plaintext
225 * @param string $text
226 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
229 public static function toPlaintext(string $text, bool $keep_urls = true): string
231 DI::profiler()->startRecording('rendering');
232 // Remove pictures in advance to avoid unneeded proxy calls
233 $text = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $2 ', $text);
234 $text = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $text);
237 $text = self::replaceAttachment($text);
239 $naked_text = HTML::toPlaintext(self::convert($text, false, BBCode::EXTERNAL, true), 0, !$keep_urls);
241 DI::profiler()->stopRecording();
245 private static function proxyUrl(string $image, int $simplehtml = self::INTERNAL, int $uriid = 0, string $size = ''): string
247 // Only send proxied pictures to API and for internal display
248 if (!in_array($simplehtml, [self::INTERNAL, self::MASTODON_API, self::TWITTER_API])) {
250 } elseif ($uriid > 0) {
251 return Post\Link::getByLink($uriid, $image, $size);
253 return Proxy::proxifyUrl($image, $size);
258 * Truncates imported message body string length to max_import_size
260 * The purpose of this function is to apply system message length limits to
261 * imported messages without including any embedded photos in the length
263 * @param string $body
265 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
267 public static function limitBodySize(string $body): string
269 DI::profiler()->startRecording('rendering');
270 $maxlen = DI::config()->get('config', 'max_import_size', 0);
272 // If the length of the body, including the embedded images, is smaller
273 // than the maximum, then don't waste time looking for the images
274 if ($maxlen && (strlen($body) > $maxlen)) {
276 Logger::info('the total body length exceeds the limit', ['maxlen' => $maxlen, 'body_len' => strlen($body)]);
282 $img_start = strpos($orig_body, '[img');
283 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
284 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
285 while (($img_st_close !== false) && ($img_end !== false)) {
287 $img_st_close++; // make it point to AFTER the closing bracket
288 $img_end += $img_start;
289 $img_end += strlen('[/img]');
291 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
292 // This is an embedded image
294 if (($textlen + $img_start) > $maxlen) {
295 if ($textlen < $maxlen) {
296 Logger::debug('the limit happens before an embedded image');
297 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
301 $new_body = $new_body . substr($orig_body, 0, $img_start);
302 $textlen += $img_start;
305 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
308 if (($textlen + $img_end) > $maxlen) {
309 if ($textlen < $maxlen) {
310 Logger::debug('the limit happens before the end of a non-embedded image');
311 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
315 $new_body = $new_body . substr($orig_body, 0, $img_end);
316 $textlen += $img_end;
319 $orig_body = substr($orig_body, $img_end);
321 if ($orig_body === false) {
322 // in case the body ends on a closing image tag
326 $img_start = strpos($orig_body, '[img');
327 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
328 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
331 if (($textlen + strlen($orig_body)) > $maxlen) {
332 if ($textlen < $maxlen) {
333 Logger::debug('the limit happens after the end of the last image');
334 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
337 Logger::debug('the text size with embedded images extracted did not violate the limit');
338 $new_body = $new_body . $orig_body;
341 DI::profiler()->stopRecording();
344 DI::profiler()->stopRecording();
350 * Processes [attachment] tags
352 * Note: Can produce a [bookmark] tag in the returned string
354 * @param string $text
355 * @param integer $simplehtml
356 * @param bool $tryoembed
360 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
362 public static function convertAttachment(string $text, int $simplehtml = self::INTERNAL, bool $tryoembed = true, array $data = [], int $uriid = 0, int $preview_mode = self::PREVIEW_LARGE): string
364 DI::profiler()->startRecording('rendering');
365 $data = $data ?: self::getAttachmentData($text);
366 if (empty($data) || empty($data['url'])) {
367 DI::profiler()->stopRecording();
371 if (isset($data['title'])) {
372 $data['title'] = strip_tags($data['title']);
373 $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
378 if (((strpos($data['text'], '[img=') !== false) || (strpos($data['text'], '[img]') !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
379 $data['preview'] = $data['image'];
385 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
386 $return = OEmbed::getHTML($data['url'], $data['title']);
388 throw new Exception('OEmbed is disabled for this attachment.');
390 } catch (Exception $e) {
391 $data['title'] = ($data['title'] ?? '') ?: $data['url'];
393 if ($simplehtml != self::CONNECTORS) {
394 $return = sprintf('<div class="type-%s">', $data['type']);
397 if ($preview_mode == self::PREVIEW_NO_IMAGE) {
398 unset($data['image']);
399 unset($data['preview']);
402 if (!empty($data['title']) && !empty($data['url'])) {
403 $preview_class = $preview_mode == self::PREVIEW_LARGE ? 'attachment-image' : 'attachment-preview';
404 if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
405 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="' . $preview_class . '" /></a>', $data['url'], self::proxyUrl($data['image'], $simplehtml, $uriid), $data['title']);
407 if (!empty($data['image'])) {
408 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="' . $preview_class . '" /></a><br>', $data['url'], self::proxyUrl($data['image'], $simplehtml, $uriid), $data['title']);
409 } elseif (!empty($data['preview'])) {
410 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br>', $data['url'], self::proxyUrl($data['preview'], $simplehtml, $uriid), $data['title']);
412 $return .= sprintf('<h4><a href="%s" target="_blank" rel="noopener noreferrer">%s</a></h4>', $data['url'], $data['title']);
416 if (!empty($data['description']) && $data['description'] != $data['title']) {
418 $return .= sprintf('<blockquote>%s</blockquote>', trim(HTML::purify($data['description'])));
421 if (!empty($data['provider_url']) && !empty($data['provider_name'])) {
422 if (!empty($data['author_name'])) {
423 $return .= sprintf('<sup><a href="%s" target="_blank" rel="noopener noreferrer">%s (%s)</a></sup>', $data['provider_url'], $data['author_name'], $data['provider_name']);
425 $return .= sprintf('<sup><a href="%s" target="_blank" rel="noopener noreferrer">%s</a></sup>', $data['provider_url'], $data['provider_name']);
429 if ($simplehtml != self::CONNECTORS) {
434 DI::profiler()->stopRecording();
435 return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
438 public static function removeShareInformation(string $text, bool $plaintext = false, bool $nolink = false): string
440 DI::profiler()->startRecording('rendering');
441 $data = self::getAttachmentData($text);
444 DI::profiler()->stopRecording();
447 DI::profiler()->stopRecording();
448 return $data['text'] . ($data['after'] ?? '');
451 $title = htmlentities($data['title'] ?? '', ENT_QUOTES, 'UTF-8', false);
452 $text = htmlentities($data['text'], ENT_QUOTES, 'UTF-8', false);
453 if ($plaintext || (($title != '') && strstr($text, $title))) {
454 $data['title'] = $data['url'];
455 } elseif (($text != '') && strstr($title, $text)) {
456 $data['text'] = $data['title'];
457 $data['title'] = $data['url'];
460 if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
461 DI::profiler()->stopRecording();
462 return $data['title'] . $data['after'];
465 // If the link already is included in the post, don't add it again
466 if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
467 DI::profiler()->stopRecording();
468 return $data['text'] . $data['after'];
471 $text = $data['text'];
473 if (!empty($data['url']) && !empty($data['title'])) {
474 $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
475 } elseif (!empty($data['url'])) {
476 $text .= "\n[url]" . $data['url'] . '[/url]';
479 DI::profiler()->stopRecording();
480 return $text . "\n" . $data['after'];
484 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
486 * @param array $match Array with the matching values
487 * @return string reformatted link including HTML codes
489 private static function convertUrlForActivityPubCallback(array $match): string
493 if (isset($match[2]) && ($match[1] != $match[2])) {
497 $parts = parse_url($url);
498 if (!isset($parts['scheme'])) {
502 return self::convertUrlForActivityPub($url);
506 * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
508 * @param string $url URL that is about to be reformatted
509 * @return string reformatted link including HTML codes
511 private static function convertUrlForActivityPub(string $url): string
513 return sprintf('<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>', $url, Strings::getStyledURL($url));
517 * [noparse][i]italic[/i][/noparse] turns into
518 * [noparse][ i ]italic[ /i ][/noparse],
519 * to hide them from parser.
521 * @param array $match
524 private static function escapeNoparseCallback(array $match): string
526 $whole_match = $match[0];
527 $captured = $match[1];
528 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
529 $new_str = str_replace($captured, $spacefied, $whole_match);
534 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
535 * now turns back and the [noparse] tags are trimmed
536 * returning [i]italic[/i]
538 * @param array $match
541 private static function unescapeNoparseCallback(array $match): string
543 $captured = $match[1];
544 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
549 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
552 * @param string $text Text to search
553 * @param string $name Tag name
554 * @param int $occurrences Number of first occurrences to skip
555 * @return boolean|array
557 public static function getTagPosition(string $text, string $name, int $occurrences = 0)
559 DI::profiler()->startRecording('rendering');
560 if ($occurrences < 0) {
565 for ($i = 0; $i <= $occurrences; $i++) {
566 if ($start_open !== false) {
567 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
571 if ($start_open === false) {
572 DI::profiler()->stopRecording();
576 $start_equal = strpos($text, '=', $start_open);
577 $start_close = strpos($text, ']', $start_open);
579 if ($start_close === false) {
580 DI::profiler()->stopRecording();
586 $end_open = strpos($text, '[/' . $name . ']', $start_close);
588 if ($end_open === false) {
589 DI::profiler()->stopRecording();
595 'open' => $start_open,
596 'close' => $start_close
600 'close' => $end_open + strlen('[/' . $name . ']')
604 if ($start_equal !== false) {
605 $res['start']['equal'] = $start_equal + 1;
608 DI::profiler()->stopRecording();
613 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
615 * @param string $pattern Preg pattern string
616 * @param string $replace Preg replace string
617 * @param string $name BBCode tag name
618 * @param string $text Text to search
621 public static function pregReplaceInTag(string $pattern, string $replace, string $name, string $text): string
623 DI::profiler()->startRecording('rendering');
625 $pos = self::getTagPosition($text, $name, $occurrences);
626 while ($pos !== false && $occurrences++ < 1000) {
627 $start = substr($text, 0, $pos['start']['open']);
628 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
629 $end = substr($text, $pos['end']['close']);
630 if ($end === false) {
634 $subject = preg_replace($pattern, $replace, $subject);
635 $text = $start . $subject . $end;
637 $pos = self::getTagPosition($text, $name, $occurrences);
640 DI::profiler()->stopRecording();
644 private static function extractImagesFromItemBody(string $body): array
651 $img_start = strpos($orig_body, '[img');
652 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
653 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
654 while (($img_st_close !== false) && ($img_end !== false)) {
655 $img_st_close++; // make it point to AFTER the closing bracket
656 $img_end += $img_start;
658 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
659 // This is an embedded image
660 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
661 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
665 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
668 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
670 if ($orig_body === false) {
671 // in case the body ends on a closing image tag
675 $img_start = strpos($orig_body, '[img');
676 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
677 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
680 $new_body = $new_body . $orig_body;
682 return ['body' => $new_body, 'images' => $saved_image];
685 private static function interpolateSavedImagesIntoItemBody(int $uriid, string $body, array $images): string
690 foreach ($images as $image) {
691 // We're depending on the property of 'foreach' (specified on the PHP website) that
692 // it loops over the array starting from the first element and going sequentially
693 // to the last element
694 $newbody = str_replace(
695 '[$#saved_image' . $cnt . '#$]',
696 '<img src="' . self::proxyUrl($image, self::INTERNAL, $uriid) . '" alt="' . DI::l10n()->t('Image/photo') . '" />',
706 * @param string $text A BBCode string
707 * @return array Empty array if no share tag is present or the following array, missing attributes end up empty strings:
708 * - comment : Text before the opening share tag
709 * - shared : Text inside the share tags
710 * - author : (Optional) Display name of the shared author
711 * - profile : (Optional) Profile page URL of the shared author
712 * - avatar : (Optional) Profile picture URL of the shared author
713 * - link : (Optional) Canonical URL of the shared post
714 * - posted : (Optional) Date the shared post was initially posted ("Y-m-d H:i:s" in GMT)
715 * - message_id: (Optional) Shared post URI if any
716 * - guid : (Optional) Shared post GUID if any
718 public static function fetchShareAttributes(string $text): array
720 DI::profiler()->startRecording('rendering');
721 if (preg_match('~(.*?)\[share](.*)\[/share]~ism', $text, $matches)) {
722 DI::profiler()->stopRecording();
730 'message_id' => trim($matches[2]),
731 'comment' => trim($matches[1]),
735 // See Issue https://github.com/friendica/friendica/issues/10454
736 // Hashtags in usernames are expanded to links. This here is a quick fix.
737 $text = preg_replace('~([@!#])\[url=.*?](.*?)\[/url]~ism', '$1$2', $text);
739 if (!preg_match('~(.*?)\[share(.*?)](.*)\[/share]~ism', $text, $matches)) {
740 DI::profiler()->stopRecording();
744 $attributes = self::extractShareAttributes($matches[2]);
746 $attributes['comment'] = trim($matches[1]);
747 $attributes['shared'] = trim($matches[3]);
749 DI::profiler()->stopRecording();
754 * @see BBCode::fetchShareAttributes()
755 * @param string $shareString Internal opening share tag string matched by the regular expression
756 * @return array A fixed attribute array where missing attribute are represented by empty strings
758 private static function extractShareAttributes(string $shareString): array
761 foreach (['author', 'profile', 'avatar', 'link', 'posted', 'guid', 'message_id'] as $field) {
762 preg_match("/$field=(['\"])(.+?)\\1/ism", $shareString, $matches);
763 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
770 * Remove the share block
772 * @param string $body
775 public static function removeSharedData(string $body): string
777 return trim(preg_replace("/\s*\[share.*?\].*?\[\/share\]\s*/ism", '', $body));
781 * This function converts a [share] block to text according to a provided callback function whose signature is:
783 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
786 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
788 * - $author_contact is a contact record array
789 * - $content is the inner content of the [share] block
790 * - $is_quote_share indicates whether there's any content before the [share] block
791 * - Return value is the string that should replace the [share] block in the provided text
793 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
795 * @param string $text A BBCode string
796 * @param callable $callback
797 * @return string The BBCode string with all [share] blocks replaced
799 public static function convertShare(string $text, callable $callback, int $uriid = 0): string
801 DI::profiler()->startRecording('rendering');
802 $return = preg_replace_callback(
803 '~(.*?)\[share(.*?)](.*)\[/share]~ism',
804 function ($match) use ($callback, $uriid) {
805 $attributes = self::extractShareAttributes($match[2]);
807 $author_contact = Contact::getByURL($attributes['profile'], false, ['id', 'url', 'addr', 'name', 'micro']);
808 $author_contact['url'] = ($author_contact['url'] ?? $attributes['profile']);
809 $author_contact['addr'] = ($author_contact['addr'] ?? '');
811 $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author'];
812 $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
813 $attributes['profile'] = ($author_contact['url'] ?? '') ?: $attributes['profile'];
815 if (!empty($author_contact['id'])) {
816 $attributes['avatar'] = Contact::getAvatarUrlForId($author_contact['id'], Proxy::SIZE_THUMB);
817 } elseif ($attributes['avatar']) {
818 $attributes['avatar'] = self::proxyUrl($attributes['avatar'], self::INTERNAL, $uriid, Proxy::SIZE_THUMB);
821 $content = preg_replace(Strings::autoLinkRegEx(), '<a href="$1">$1</a>', $match[3]);
823 return $match[1] . $callback($attributes, $author_contact, $content, trim($match[1]) != '');
828 DI::profiler()->stopRecording();
829 return trim($return);
833 * Convert complex IMG and ZMG elements
835 * @param [type] $text
836 * @param integer $simplehtml
837 * @param integer $uriid
840 private static function convertImages(string $text, int $simplehtml, int $uriid = 0): string
842 DI::profiler()->startRecording('rendering');
843 $return = preg_replace_callback(
844 "/\[[zi]mg(.*?)\]([^\[\]]*)\[\/[zi]mg\]/ism",
845 function ($match) use ($simplehtml, $uriid) {
846 $attribute_string = $match[1];
848 foreach (['alt', 'width', 'height'] as $field) {
849 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
850 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
853 $img_str = '<img src="' . self::proxyUrl($match[2], $simplehtml, $uriid) . '"';
854 foreach ($attributes as $key => $value) {
855 if (!empty($value)) {
856 $img_str .= ' ' . $key . '="' . htmlspecialchars($value, ENT_COMPAT) . '"';
859 return $img_str . '>';
864 DI::profiler()->stopRecording();
869 * Default [share] tag conversion callback
871 * Note: Can produce a [bookmark] tag in the output
873 * @see BBCode::convertShare()
874 * @param array $attributes [share] block attribute values
875 * @param array $author_contact Contact row of the shared author
876 * @param string $content Inner content of the [share] block
877 * @param boolean $is_quote_share Whether there is content before the [share] block
878 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
880 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
882 private static function convertShareCallback(array $attributes, array $author_contact, string $content, bool $is_quote_share, int $simplehtml): string
884 DI::profiler()->startRecording('rendering');
885 $mention = $attributes['author'] . ' (' . ($author_contact['addr'] ?? '') . ')';
887 switch ($simplehtml) {
888 case self::MASTODON_API:
889 case self::TWITTER_API:
890 $text = ($is_quote_share ? '<br>' : '') .
891 '<b><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . "</a>:</b><br>\n" .
892 '<blockquote class="shared_content" dir="auto">' . $content . '</blockquote>';
895 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
896 $text = ($is_quote_share ? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
898 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
900 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
901 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
904 $text = ($is_quote_share ? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
906 if (empty($attributes['posted']) && !empty($attributes['link'])) {
907 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
912 case self::CONNECTORS:
913 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
914 $headline .= DI::l10n()->t('<a href="%1$s" target="_blank" rel="noopener noreferrer">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
915 $headline .= ':</b></p>' . "\n";
917 $text = ($is_quote_share ? '<hr />' : '') . $headline . '<blockquote class="shared_content" dir="auto">' . trim($content) . '</blockquote>' . "\n";
921 $text = ($is_quote_share ? '<br>' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
923 case self::ACTIVITYPUB:
924 $author = '@<span class="vcard"><a href="' . $author_contact['url'] . '" class="url u-url mention" title="' . $author_contact['addr'] . '"><span class="fn nickname mention">' . $author_contact['addr'] . '</span></a>:</span>';
925 $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
928 $text = ($is_quote_share ? "\n" : '');
930 $contact = Contact::getByURL($attributes['profile'], false, ['network']);
931 $network = $contact['network'] ?? Protocol::PHANTOM;
933 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
934 $text .= BBCode::SHARED_ANCHOR . Renderer::replaceMacros($tpl, [
935 '$profile' => $attributes['profile'],
936 '$avatar' => $attributes['avatar'],
937 '$author' => $attributes['author'],
938 '$link' => $attributes['link'],
939 '$link_title' => DI::l10n()->t('Link to source'),
940 '$posted' => $attributes['posted'],
941 '$guid' => $attributes['guid'],
942 '$network_name' => ContactSelector::networkToName($network, $attributes['profile']),
943 '$network_icon' => ContactSelector::networkToIcon($network, $attributes['profile']),
944 '$content' => self::TOP_ANCHOR . self::setMentions(trim($content), 0, $network) . self::BOTTOM_ANCHOR,
952 private static function removePictureLinksCallback(array $match): string
954 $cache_key = 'remove:' . $match[1];
955 $text = DI::cache()->get($cache_key);
957 if (is_null($text)) {
958 $curlResult = DI::httpClient()->head($match[1], [HttpClientOptions::TIMEOUT => DI::config()->get('system', 'xrd_timeout')]);
959 if ($curlResult->isSuccess()) {
960 $mimetype = $curlResult->getHeader('Content-Type')[0] ?? '';
965 if (substr($mimetype, 0, 6) == 'image/') {
966 $text = '[url=' . $match[1] . ']' . $match[1] . '[/url]';
968 $text = '[url=' . $match[2] . ']' . $match[2] . '[/url]';
970 // if its not a picture then look if its a page that contains a picture link
971 $body = DI::httpClient()->fetch($match[1], HttpClientAccept::HTML, 0);
973 DI::cache()->set($cache_key, $text);
977 $doc = new DOMDocument();
978 @$doc->loadHTML($body);
979 $xpath = new DOMXPath($doc);
980 $list = $xpath->query('//meta[@name]');
981 foreach ($list as $node) {
984 if ($node->attributes->length) {
985 foreach ($node->attributes as $attribute) {
986 $attr[$attribute->name] = $attribute->value;
990 if (strtolower($attr['name']) == 'twitter:image') {
991 $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
995 DI::cache()->set($cache_key, $text);
1002 * Callback: Expands links from given $match array
1004 * @param array $match Array with link match
1005 * @return string BBCode
1007 private static function expandLinksCallback(array $match): string
1009 if (($match[3] == '') || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1010 return ($match[1] . '[url]' . $match[2] . '[/url]');
1012 return ($match[1] . $match[3] . ' [url]' . $match[2] . '[/url]');
1017 * Callback: Cleans picture links
1019 * @param array $match Array with link match
1020 * @return string BBCode
1022 private static function cleanPictureLinksCallback(array $match): string
1024 // When the picture link is the own photo path then we can avoid fetching the link
1025 $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()) . '/photos/');
1026 if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
1027 if (!empty($match[3])) {
1028 $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
1030 $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
1035 $cache_key = 'clean:' . $match[1];
1036 $text = DI::cache()->get($cache_key);
1037 if (!is_null($text)) {
1041 $curlResult = DI::httpClient()->head($match[1], [HttpClientOptions::TIMEOUT => DI::config()->get('system', 'xrd_timeout')]);
1042 if ($curlResult->isSuccess()) {
1043 $mimetype = $curlResult->getHeader('Content-Type')[0] ?? '';
1048 // if its a link to a picture then embed this picture
1049 if (substr($mimetype, 0, 6) == 'image/') {
1050 $text = '[img]' . $match[1] . '[/img]';
1052 if (!empty($match[3])) {
1053 $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
1055 $text = '[img]' . $match[2] . '[/img]';
1058 // if its not a picture then look if its a page that contains a picture link
1059 $body = DI::httpClient()->fetch($match[1], HttpClientAccept::HTML, 0);
1061 DI::cache()->set($cache_key, $text);
1065 $doc = new DOMDocument();
1066 @$doc->loadHTML($body);
1067 $xpath = new DOMXPath($doc);
1068 $list = $xpath->query('//meta[@name]');
1069 foreach ($list as $node) {
1071 if ($node->attributes->length) {
1072 foreach ($node->attributes as $attribute) {
1073 $attr[$attribute->name] = $attribute->value;
1077 if (strtolower($attr['name']) == "twitter:image") {
1078 if (!empty($match[3])) {
1079 $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
1081 $text = "[img]" . $attr['content'] . "[/img]";
1086 DI::cache()->set($cache_key, $text);
1092 * Cleans picture links
1094 * @param string $text HTML/BBCode string
1095 * @return string Cleaned HTML/BBCode
1097 public static function cleanPictureLinks(string $text): string
1099 DI::profiler()->startRecording('rendering');
1100 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", [self::class, 'cleanPictureLinksCallback'], $text);
1101 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", [self::class, 'cleanPictureLinksCallback'], $return);
1102 DI::profiler()->stopRecording();
1109 * @param string $text HTML/BBCode string
1110 * @return string Cleaned HTML/BBCode
1112 public static function removeLinks(string $bbcode): string
1114 DI::profiler()->startRecording('rendering');
1115 $bbcode = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $1 ', $bbcode);
1116 $bbcode = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $bbcode);
1118 $bbcode = preg_replace('/[@!#]\[url\=.*?\].*?\[\/url\]/ism', '', $bbcode);
1119 $bbcode = preg_replace("/\[url=[^\[\]]*\](.*)\[\/url\]/Usi", ' $1 ', $bbcode);
1120 $bbcode = preg_replace('/[@!#]?\[url.*?\[\/url\]/ism', '', $bbcode);
1121 DI::profiler()->stopRecording();
1126 * Replace names in mentions with nicknames
1128 * @param string $body HTML/BBCode
1129 * @return string Body with replaced mentions
1131 public static function setMentionsToNicknames(string $body): string
1133 DI::profiler()->startRecording('rendering');
1134 $regexp = "/([@!])\[url\=([^\[\]]*)\].*?\[\/url\]/ism";
1135 $body = preg_replace_callback($regexp, [self::class, 'mentionCallback'], $body);
1136 DI::profiler()->stopRecording();
1141 * Callback function to replace a Friendica style mention in a mention with the nickname
1143 * @param array $match Matching values for the callback
1144 * @return string Replaced mention or empty string
1145 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1147 private static function mentionCallback(array $match): string
1149 if (empty($match[2])) {
1153 $data = Contact::getByURL($match[2], false, ['url', 'nick']);
1154 if (empty($data['nick'])) {
1158 return $match[1] . '[url=' . $data['url'] . ']' . $data['nick'] . '[/url]';
1162 * Normalize links to Youtube and Vimeo to a unified format.
1164 * @param string $text
1167 private static function normalizeVideoLinks(string $text): string
1169 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1170 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1171 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/shorts\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1172 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1174 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1175 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1181 * Expand Youtube and Vimeo links to
1183 * @param string $text
1186 public static function expandVideoLinks(string $text): string
1188 $text = self::normalizeVideoLinks($text);
1189 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '[url=https://www.youtube.com/watch?v=$1]https://www.youtube.com/watch?v=$1[/url]', $text);
1190 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '[url=https://vimeo.com/$1]https://vimeo.com/$1[/url]', $text);
1196 * Converts a BBCode message for a given URI-ID to a HTML message
1198 * BBcode 2 HTML was written by WAY2WEB.net
1199 * extended to work with Mistpark/Friendica - Mike Macgirvin
1201 * Simple HTML values meaning:
1202 * - 0: Friendica display
1204 * - 2: Used for Windows Phone push, Friendica API
1205 * - 3: Used before converting to Markdown in bb2diaspora.php
1206 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1209 * - 7: Used for dfrn, OStatus
1210 * - 8: Used for Twitter, WP backlink text setting
1214 * @param string $text
1215 * @param int $simple_html
1217 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1219 public static function convertForUriId(int $uriid = null, string $text = null, int $simple_html = self::INTERNAL): string
1221 $try_oembed = ($simple_html == self::INTERNAL);
1223 return self::convert($text ?? '', $try_oembed, $simple_html, false, $uriid ?? 0);
1227 * Converts a BBCode message to HTML message
1229 * BBcode 2 HTML was written by WAY2WEB.net
1230 * extended to work with Mistpark/Friendica - Mike Macgirvin
1232 * Simple HTML values meaning:
1233 * - 0: Friendica display
1235 * - 2: Used for Windows Phone push, Friendica API
1236 * - 3: Used before converting to Markdown in bb2diaspora.php
1237 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1240 * - 7: Used for dfrn, OStatus
1241 * - 8: Used for Twitter, WP backlink text setting
1244 * @param string $text
1245 * @param bool $try_oembed
1246 * @param int $simple_html
1247 * @param bool $for_plaintext
1249 * @return string Converted code or empty string
1250 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1252 public static function convert(string $text = null, bool $try_oembed = true, int $simple_html = self::INTERNAL, bool $for_plaintext = false, int $uriid = 0): string
1254 // Accounting for null default column values
1255 if (is_null($text) || $text === '') {
1259 DI::profiler()->startRecording('rendering');
1261 Hook::callAll('bbcode', $text);
1265 $text = self::performWithEscapedTags($text, ['code'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a, $uriid) {
1266 $text = self::performWithEscapedTags($text, ['noparse', 'nobb', 'pre'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a, $uriid) {
1268 * preg_match_callback function to replace potential Oembed tags with Oembed content
1270 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1272 * $match[2] = $title or absent
1274 $try_oembed_callback = function (array $match) {
1276 $title = $match[2] ?? '';
1279 $return = OEmbed::getHTML($url, $title);
1280 } catch (Exception $ex) {
1281 $return = $match[0];
1287 // Remove the abstract element. It is a non visible element.
1288 $text = self::stripAbstract($text);
1290 // Line ending normalisation
1291 $text = str_replace("\r\n", "\n", $text);
1293 // Move new lines outside of tags
1294 $text = preg_replace("#\[(\w*)](\n*)#ism", '$2[$1]', $text);
1295 $text = preg_replace("#(\n*)\[/(\w*)]#ism", '[/$2]$1', $text);
1297 // Extract the private images which use data urls since preg has issues with
1298 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1299 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1301 $extracted = self::extractImagesFromItemBody($text);
1302 $text = $extracted['body'];
1303 $saved_image = $extracted['images'];
1305 // If we find any event code, turn it into an event.
1306 // After we're finished processing the bbcode we'll
1307 // replace all of the event code with a reformatted version.
1309 $ev = Event::fromBBCode($text);
1311 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1312 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1314 $text = str_replace("<", "<", $text);
1315 $text = str_replace(">", ">", $text);
1317 // remove some newlines before the general conversion
1318 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1]$2[/share]\n", $text);
1319 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "\n[quote$1]$2[/quote]\n", $text);
1321 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1323 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1326 // Remove linefeeds inside of the table elements. See issue #6799
1328 "\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
1329 "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
1330 "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
1331 "\n[hr]", "[hr]\n", " [hr]", "[hr] ",
1332 "\n[attachment ", " [attachment ", "\n[/attachment]", "[/attachment]\n", " [/attachment]", "[/attachment] ",
1333 "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] ",
1334 " \n", "\t\n", "[/li]\n", "\n[li]", "\n[*]",
1337 "[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
1338 "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
1339 "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
1340 "[hr]", "[hr]", "[hr]", "[hr]",
1341 "[attachment ", "[attachment ", "[/attachment]", "[/attachment]", "[/attachment]", "[/attachment]",
1342 "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]",
1343 "\n", "\n", "[/li]", "[li]", "[*]",
1347 $text = str_replace($search, $replace, $text);
1348 } while ($oldtext != $text);
1350 // Replace these here only once
1351 $search = ["\n[table]", "[/table]\n"];
1352 $replace = ["[table]", "[/table]"];
1353 $text = str_replace($search, $replace, $text);
1355 // Trim new lines regardless of the system.remove_multiplicated_lines config value
1356 $text = trim($text, "\n");
1358 // removing multiplicated newlines
1359 if (DI::config()->get('system', 'remove_multiplicated_lines')) {
1361 "\n\n\n", "[/quote]\n\n", "\n[/quote]", "\n[ul]", "[/ul]\n", "\n[ol]", "[/ol]\n", "\n\n[share ", "[/attachment]\n",
1362 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"
1365 "\n\n", "[/quote]\n", "[/quote]", "[ul]", "[/ul]", "[ol]", "[/ol]", "\n[share ", "[/attachment]",
1366 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"
1370 $text = str_replace($search, $replace, $text);
1371 } while ($oldtext != $text);
1374 /// @todo Have a closer look at the different html modes
1375 // Handle attached links or videos
1376 if ($simple_html == self::NPF) {
1377 $text = self::removeAttachment($text);
1378 } elseif (in_array($simple_html, [self::MASTODON_API, self::TWITTER_API, self::ACTIVITYPUB])) {
1379 $text = self::replaceAttachment($text);
1380 } elseif (!in_array($simple_html, [self::INTERNAL, self::EXTERNAL, self::CONNECTORS])) {
1381 $text = self::replaceAttachment($text, true);
1383 $text = self::convertAttachment($text, $simple_html, $try_oembed, [], $uriid);
1386 $nosmile = strpos($text, '[nosmile]') !== false;
1387 $text = str_replace('[nosmile]', '', $text);
1389 // Replace non graphical smilies for external posts
1391 $text = self::performWithEscapedTags($text, ['img'], function ($text) use ($simple_html, $for_plaintext) {
1392 return Smilies::replace($text, ($simple_html != self::INTERNAL) || $for_plaintext);
1396 // leave open the possibility of [map=something]
1397 // this is replaced in Item::prepareBody() which has knowledge of the item location
1398 if (strpos($text, '[/map]') !== false) {
1399 $text = preg_replace_callback(
1400 "/\[map\](.*?)\[\/map\]/ism",
1401 function ($match) use ($simple_html) {
1402 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1408 if (strpos($text, '[map=') !== false) {
1409 $text = preg_replace_callback(
1410 "/\[map=(.*?)\]/ism",
1411 function ($match) use ($simple_html) {
1412 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1418 if (strpos($text, '[map]') !== false) {
1419 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1422 // Check for headers
1424 if ($simple_html == self::INTERNAL) {
1425 //Ensure to always start with <h4> if possible
1427 for ($level = 6; $level > 0; $level--) {
1428 if (preg_match("(\[h$level\].*?\[\/h$level\])ism", $text)) {
1432 if ($heading_count > 0) {
1433 $heading = min($heading_count + 3, 6);
1434 for ($level = 6; $level > 0; $level--) {
1435 if (preg_match("(\[h$level\].*?\[\/h$level\])ism", $text)) {
1436 $text = preg_replace("(\[h$level\](.*?)\[\/h$level\])ism", "</p><h$heading>$1</h$heading><p>", $text);
1442 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '</p><h1>$1</h1><p>', $text);
1443 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '</p><h2>$1</h2><p>', $text);
1444 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '</p><h3>$1</h3><p>', $text);
1445 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '</p><h4>$1</h4><p>', $text);
1446 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '</p><h5>$1</h5><p>', $text);
1447 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '</p><h6>$1</h6><p>', $text);
1450 // Check for paragraph
1451 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1453 // Check for bold text
1454 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1456 // Check for Italics text
1457 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1459 // Check for Underline text
1460 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1462 // Check for strike-through text
1463 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1465 // Check for over-line text
1466 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1468 // Check for colored text
1469 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1471 // Check for sized text
1472 // [size=50] --> font-size: 50px (with the unit).
1473 if ($simple_html != self::DIASPORA) {
1474 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", '<span style="font-size:$1px;line-height:normal;">$2</span>', $text);
1475 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", '<span style="font-size:$1;line-height:normal;">$2</span>', $text);
1477 // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
1478 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text);
1482 // Check for centered text
1483 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", '<div style="text-align:center;">$1</div>', $text);
1485 // Check for list text
1486 $text = str_replace("[*]", "<li>", $text);
1488 // Check for block-level custom CSS
1489 $text = preg_replace('#(?<=^|\n)\[style=(.*?)](.*?)\[/style](?:\n|$)#ism', '<div style="$1">$2</div>', $text);
1491 // Check for inline custom CSS
1492 $text = preg_replace("(\[style=(.*?)\](.*?)\[\/style\])ism", '<span style="$1">$2</span>', $text);
1494 // Mastodon Emoji (internal tag, do not document for users)
1495 $text = preg_replace("(\[emoji=(.*?)](.*?)\[/emoji])ism", '<span class="mastodon emoji"><img src="$1" alt="$2" title="$2"/></span>', $text);
1497 // Check for CSS classes
1498 // @deprecated since 2021.12, left for backward-compatibility reasons
1499 $text = preg_replace("(\[class=(.*?)\](.*?)\[\/class\])ism", '<span class="$1">$2</span>', $text);
1500 // Add HTML new lines
1501 $text = str_replace("\n\n", '</p><p>', $text);
1502 $text = str_replace("\n", '<br>', $text);
1504 // handle nested lists
1507 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1508 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1509 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1510 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1511 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '</p><ul class="listbullet" style="list-style-type: circle;">$1</ul><p>', $text);
1512 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '</p><ul class="listnone" style="list-style-type: none;">$1</ul><p>', $text);
1513 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '</p><ul class="listdecimal" style="list-style-type: decimal;">$1</ul><p>', $text);
1514 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '</p><ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul><p>', $text);
1515 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '</p><ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul><p>', $text);
1516 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '</p><ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul><p>', $text);
1517 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '</p><ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul><p>', $text);
1518 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '</p><ul>$1</ul><p>', $text);
1519 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '</p><ol>$1</ol><p>', $text);
1520 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1523 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1524 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1525 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1526 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '</p><table>$1</table><p>', $text);
1528 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '</p><table border="1" >$1</table><p>', $text);
1529 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '</p><table border="0" >$1</table><p>', $text);
1531 $text = str_replace('[hr]', '</p><hr /><p>', $text);
1533 if (!$for_plaintext) {
1534 $text = self::performWithEscapedTags($text, ['url', 'img', 'audio', 'video', 'youtube', 'vimeo', 'share', 'attachment', 'iframe', 'bookmark'], function ($text) {
1535 return preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1539 // Check for font change text
1540 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1542 // Declare the format for [spoiler] layout
1543 $SpoilerLayout = '<details class="spoiler"><summary>' . DI::l10n()->t('Click to open/close') . '</summary>$1</details>';
1545 // Check for [spoiler] text
1546 // handle nested quotes
1548 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1549 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", $SpoilerLayout, $text);
1552 // Check for [spoiler=Title] text
1554 // handle nested quotes
1556 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1557 $text = preg_replace(
1558 "/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1559 '<details class="spoiler"><summary>$1</summary>$2</details>',
1564 // Declare the format for [quote] layout
1565 $QuoteLayout = '</p><blockquote>$1</blockquote><p>';
1567 // Check for [quote] text
1568 // handle nested quotes
1570 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1571 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1574 // Check for [quote=Author] text
1576 $t_wrote = DI::l10n()->t('$1 wrote:');
1578 // handle nested quotes
1580 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1581 $text = preg_replace(
1582 "/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1583 "<p><strong class=" . '"author"' . ">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1589 // [img=widthxheight]image source[/img]
1590 $text = preg_replace_callback(
1591 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1592 function ($matches) use ($simple_html, $uriid) {
1593 if (strpos($matches[3], "data:image/") === 0) {
1597 $matches[3] = self::proxyUrl($matches[3], $simple_html, $uriid);
1598 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1603 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1604 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1606 $text = preg_replace_callback(
1607 "/\[[iz]mg\=(.*?)\](.*?)\[\/[iz]mg\]/ism",
1608 function ($matches) use ($simple_html, $uriid) {
1609 $matches[1] = self::proxyUrl($matches[1], $simple_html, $uriid);
1610 $alt = htmlspecialchars($matches[2], ENT_COMPAT);
1611 // Fix for Markdown problems with Diaspora, see issue #12701
1612 if (($simple_html != self::DIASPORA) || strpos($matches[2], '"') === false) {
1613 return '<img src="' . $matches[1] . '" alt="' . $alt . '" title="' . $alt . '">';
1615 return '<img src="' . $matches[1] . '" alt="' . $alt . '">';
1622 // [img]pathtoimage[/img]
1623 $text = preg_replace_callback(
1624 "/\[[iz]mg\](.*?)\[\/[iz]mg\]/ism",
1625 function ($matches) use ($simple_html, $uriid) {
1626 if (strpos($matches[1], "data:image/") === 0) {
1630 $matches[1] = self::proxyUrl($matches[1], $simple_html, $uriid);
1631 return "[img]" . $matches[1] . "[/img]";
1636 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1637 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1639 $text = self::convertImages($text, $simple_html, $uriid);
1641 $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br><img src="' . DI::baseUrl() . '/images/lock_icon.gif" alt="' . DI::l10n()->t('Encrypted content') . '" title="' . DI::l10n()->t('Encrypted content') . '" /><br>', $text);
1642 $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br><img src="' . DI::baseUrl() . '/images/lock_icon.gif" alt="' . DI::l10n()->t('Encrypted content') . '" title="' . '$1' . ' ' . DI::l10n()->t('Encrypted content') . '" /><br>', $text);
1643 //$text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . DI::l10n()->t('Encrypted content') . '" title="' . '$1' . ' ' . DI::l10n()->t('Encrypted content') . '" /><br>', $text);
1645 // Simplify "video" element
1646 $text = preg_replace('(\[video[^\]]*?\ssrc\s?=\s?([^\s\]]+)[^\]]*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
1648 if ($simple_html == self::NPF) {
1649 $text = preg_replace(
1650 "/\[video\](.*?)\[\/video\]/ism",
1651 '</p><video src="$1" controls width="100%" height="auto">$1</video><p>',
1654 $text = preg_replace(
1655 "/\[audio\](.*?)\[\/audio\]/ism",
1656 '</p><audio src="$1" controls>$1">$1</audio><p>',
1659 } elseif ($try_oembed) {
1660 // html5 video and audio
1661 $text = preg_replace(
1662 "/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism",
1663 '<video src="$1" controls width="100%" height="auto"><a href="$1">$1</a></video>',
1667 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1668 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1670 $text = preg_replace(
1671 "/\[video\](.*?)\[\/video\]/ism",
1672 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>',
1675 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", '<audio src="$1" controls><a href="$1">$1</a></audio>', $text);
1677 $text = preg_replace(
1678 "/\[video\](.*?)\[\/video\]/ism",
1679 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>',
1682 $text = preg_replace(
1683 "/\[audio\](.*?)\[\/audio\]/ism",
1684 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>',
1689 // Backward compatibility, [iframe] support has been removed in version 2020.12
1690 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1692 $text = self::normalizeVideoLinks($text);
1694 // Youtube extensions
1696 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '<iframe width="' . $a->getThemeInfoValue('videowidth') . '" height="' . $a->getThemeInfoValue('videoheight') . '" src="https://www.youtube.com/embed/$1" frameborder="0" ></iframe>', $text);
1698 $text = preg_replace(
1699 "/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1700 '<a href="https://www.youtube.com/watch?v=$1" target="_blank" rel="noopener noreferrer">https://www.youtube.com/watch?v=$1</a>',
1707 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '<iframe width="' . $a->getThemeInfoValue('videowidth') . '" height="' . $a->getThemeInfoValue('videoheight') . '" src="https://player.vimeo.com/video/$1" frameborder="0" ></iframe>', $text);
1709 $text = preg_replace(
1710 "/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1711 '<a href="https://vimeo.com/$1" target="_blank" rel="noopener noreferrer">https://vimeo.com/$1</a>',
1717 $text = OEmbed::BBCode2HTML($text);
1719 // Avoid triple linefeeds through oembed
1720 $text = str_replace("<br style='clear:left'></span><br><br>", "<br style='clear:left'></span><br>", $text);
1722 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1723 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1724 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1725 // start which is always required). Allow desc with a missing summary for compatibility.
1727 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1728 $sub = Event::getHTML($ev, $simple_html, $uriid);
1730 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1731 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1732 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1733 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1734 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1735 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1738 if (!$for_plaintext && DI::config()->get('system', 'big_emojis') && ($simple_html != self::DIASPORA)) {
1739 $conv = html_entity_decode(str_replace([' ', "\n", "\r"], '', $text));
1740 // Emojis are always 4 byte Unicode characters
1741 if (!empty($conv) && (strlen($conv) / mb_strlen($conv) == 4)) {
1742 $text = '<span style="font-size: xx-large; line-height: normal;">' . $text . '</span>';
1746 // Handle mentions and hashtag links
1747 if ($simple_html == self::DIASPORA) {
1748 // The ! is converted to @ since Diaspora only understands the @
1749 $text = preg_replace(
1750 "/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1751 '@<a href="$2">$3</a>',
1754 } elseif (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1755 $text = preg_replace(
1756 "/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1757 '<span class="h-card"><a href="$2" class="u-url mention">$1<span>$3</span></a></span>',
1760 $text = preg_replace(
1761 "/([#])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1762 '<a href="$2" class="mention hashtag" rel="tag">$1<span>$3</span></a>',
1765 } elseif (in_array($simple_html, [self::INTERNAL, self::EXTERNAL, self::TWITTER_API])) {
1766 $text = preg_replace(
1767 "/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1768 '<bdi>$1<a href="$2" class="userinfo mention" title="$3">$3</a></bdi>',
1771 } elseif ($simple_html == self::MASTODON_API) {
1772 $text = preg_replace(
1773 "/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1774 '<a class="u-url mention status-link" href="$2" rel="nofollow noopener noreferrer" target="_blank" title="$3">$1<span>$3</span></a>',
1777 $text = preg_replace(
1778 "/([#])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1779 '<a class="mention hashtag status-link" href="$2" rel="tag">$1<span>$3</span></a>',
1783 $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1786 if (!$for_plaintext) {
1787 if (in_array($simple_html, [self::OSTATUS, self::MASTODON_API, self::TWITTER_API, self::ACTIVITYPUB])) {
1788 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", [self::class, 'convertUrlForActivityPubCallback'], $text);
1789 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", [self::class, 'convertUrlForActivityPubCallback'], $text);
1792 $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1793 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", [self::class, 'removePictureLinksCallback'], $text);
1796 // Bookmarks in red - will be converted to bookmarks in friendica
1797 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1798 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1799 $text = preg_replace(
1800 "/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1801 "[bookmark=$1]$2[/bookmark]",
1805 if (in_array($simple_html, [self::OSTATUS, self::TWITTER, self::BLUESKY])) {
1806 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", [self::class, 'expandLinksCallback'], $text);
1807 //$text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $text);
1808 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]', $text);
1811 // Perform URL Search
1813 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1816 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1818 // Handle Diaspora posts
1819 $text = preg_replace_callback(
1820 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1822 return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1827 $text = preg_replace_callback(
1828 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1830 return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1835 // Server independent link to posts and comments
1836 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1837 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1838 $text = preg_replace($expression, DI::baseUrl() . "/display/$1", $text);
1842 * - #[url=<anything>]<term>[/url]
1843 * - [url=<anything>]#<term>[/url]
1845 self::performWithEscapedTags($text, ['url', 'share'], function ($text) use ($simple_html) {
1846 $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function ($matches) use ($simple_html) {
1847 if ($simple_html == self::ACTIVITYPUB) {
1848 return '<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1849 . '" data-tag="' . XML::escape($matches[1]) . '" rel="tag ugc">#'
1850 . XML::escape($matches[1]) . '</a>';
1852 return '#<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1853 . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
1854 . XML::escape($matches[1]) . '</a>';
1860 // We need no target="_blank" rel="noopener noreferrer" for local links
1861 // convert links start with DI::baseUrl() as local link without the target="_blank" rel="noopener noreferrer" attribute
1862 $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
1863 $text = preg_replace("/\[url\](" . $escapedBaseUrl . ".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1864 $text = preg_replace("/\[url\=(" . $escapedBaseUrl . ".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1866 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1867 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1869 // Red compatibility, though the link can't be authenticated on Friendica
1870 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1873 // we may need to restrict this further if it picks up too many strays
1874 // link acct:user@host to a webfinger profile redirector
1876 $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="' . DI::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
1878 // Perform MAIL Search
1879 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1880 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1882 /// @todo What is the meaning of these lines?
1883 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1884 $text = preg_replace('/\&\#039\;/', '\'', $text);
1886 // Currently deactivated, it made problems with " inside of alt texts.
1887 //$text = preg_replace('/\"\;/', '"', $text);
1889 // fix any escaped ampersands that may have been converted into links
1890 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1892 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1893 $allowed_src_protocols = ['//', 'http://', 'https://', 'contact/redir/', 'cid:'];
1895 array_walk($allowed_src_protocols, function (&$value) {
1896 $value = preg_quote($value, '#');
1899 $text = preg_replace(
1900 '#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1901 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">',
1905 // sanitize href attributes (only allowlisted protocols URLs)
1906 // default value for backward compatibility
1907 $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
1909 // Always allowed protocol even if config isn't set or not including it
1910 $allowed_link_protocols[] = '//';
1911 $allowed_link_protocols[] = 'http://';
1912 $allowed_link_protocols[] = 'https://';
1913 $allowed_link_protocols[] = 'contact/redir/';
1915 array_walk($allowed_link_protocols, function (&$value) {
1916 $value = preg_quote($value, '#');
1919 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1920 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
1923 $text = self::convertShare(
1925 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1926 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1931 $text = self::interpolateSavedImagesIntoItemBody($uriid, $text, $saved_image);
1934 }); // Escaped noparse, nobb, pre
1936 // Remove escaping tags and replace new lines that remain
1937 $text = preg_replace_callback('/\[(noparse|nobb)](.*?)\[\/\1]/ism', function ($match) {
1938 return str_replace("\n", "<br>", $match[2]);
1941 // Additionally, [pre] tags preserve spaces
1942 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", function ($match) {
1943 return str_replace([' ', "\n"], [' ', "<br>"], htmlentities($match[1], ENT_NOQUOTES, 'UTF-8'));
1949 $text = preg_replace_callback(
1950 "#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1951 function ($matches) {
1952 if (strpos($matches[2], "\n") !== false) {
1953 $return = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlentities(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
1955 $return = '<code>' . htmlentities($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
1963 // Default iframe allowed domains/path
1964 $allowedIframeDomains = [
1965 DI::baseUrl()->getHost()
1966 . (DI::baseUrl()->getPath() ? '/' . DI::baseUrl()->getPath() : '')
1967 . '/oembed/', # The path part has to change with the source in Content\Oembed::iframe
1968 'www.youtube.com/embed/',
1969 'player.vimeo.com/video/',
1972 $allowedIframeDomains = array_merge(
1973 $allowedIframeDomains,
1974 DI::config()->get('system', 'allowed_oembed') ?
1975 explode(',', DI::config()->get('system', 'allowed_oembed'))
1979 if (strpos($text, '<p>') !== false || strpos($text, '</p>') !== false) {
1980 $text = '<p>' . $text . '</p>';
1983 $text = HTML::purify($text, $allowedIframeDomains);
1984 DI::profiler()->stopRecording();
1990 * Strips the "abstract" tag from the provided text
1992 * @param string $text The text with BBCode
1993 * @return string The same text - but without "abstract" element
1995 public static function stripAbstract(string $text): string
1997 DI::profiler()->startRecording('rendering');
1999 $text = BBCode::performWithEscapedTags($text, ['code', 'noparse', 'nobb', 'pre'], function ($text) {
2000 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", ' ', $text);
2001 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", ' ', $text);
2005 DI::profiler()->stopRecording();
2010 * Returns the value of the "abstract" element
2012 * @param string $text The text that maybe contains the element
2013 * @param string $addon The addon for which the abstract is meant for
2014 * @return string The abstract
2016 public static function getAbstract(string $text, string $addon = ''): string
2018 DI::profiler()->startRecording('rendering');
2019 $addon = strtolower($addon);
2021 $abstract = BBCode::performWithEscapedTags($text, ['code', 'noparse', 'nobb', 'pre'], function ($text) use ($addon) {
2022 if ($addon && preg_match('#\[abstract=' . preg_quote($addon, '#') . '](.*?)\[/abstract]#ism', $text, $matches)) {
2026 if (preg_match("#\[abstract](.*?)\[/abstract]#ism", $text, $matches)) {
2033 DI::profiler()->stopRecording();
2038 * Callback function to replace a Friendica style mention in a mention for Diaspora
2040 * @param array $match Matching values for the callback
2041 * [1] = Mention type (! or @)
2044 * @return string Replaced mention
2045 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2046 * @throws \ImagickException
2048 private static function bbCodeMention2DiasporaCallback(array $match): string
2050 $contact = Contact::getByURL($match[3], false, ['addr']);
2051 if (empty($contact['addr'])) {
2055 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
2060 * Converts a BBCode text into Markdown
2062 * This function converts a BBCode item body to be sent to Markdown-enabled
2063 * systems like Diaspora and Libertree
2065 * @param string $text
2066 * @param bool $for_diaspora Diaspora requires more changes than Libertree
2068 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2070 public static function toMarkdown(string $text, bool $for_diaspora = true): string
2072 DI::profiler()->startRecording('rendering');
2073 $original_text = $text;
2075 // Since Diaspora is creating a summary for links, this function removes them before posting
2076 if ($for_diaspora) {
2077 $text = self::removeShareInformation($text);
2081 * Transform #tags, strip off the [url] and replace spaces with underscore
2083 $url_search_string = "^\[\]";
2084 $text = preg_replace_callback(
2085 "/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
2086 function ($matches) {
2087 return '#' . str_replace(' ', '_', $matches[2]);
2092 // Converting images with size parameters to simple images. Markdown doesn't know it.
2093 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2095 // Convert it to HTML - don't try oembed
2096 if ($for_diaspora) {
2097 $text = self::convertForUriId(0, $text, self::DIASPORA);
2099 // Add all tags that maybe were removed
2100 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2102 foreach ($tags[2] as $tag) {
2103 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2104 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2105 $tagline .= '#' . $tag . ' ';
2108 $text = $text . ' ' . $tagline;
2111 $text = self::convertForUriId(0, $text, self::CONNECTORS);
2114 // If a link is followed by a quote then there should be a newline before it
2115 // Maybe we should make this newline at every time before a quote.
2116 $text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
2118 // Now convert HTML to Markdown
2119 $text = HTML::toMarkdown($text);
2121 // Libertree has a problem with escaped hashtags.
2122 $text = str_replace(['\#'], ['#'], $text);
2124 // Remove any leading or trailing whitespace, as this will mess up
2125 // the Diaspora signature verification and cause the item to disappear
2126 $text = trim($text);
2128 if ($for_diaspora) {
2129 $url_search_string = "^\[\]";
2130 $text = preg_replace_callback(
2131 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
2132 [self::class, 'bbCodeMention2DiasporaCallback'],
2137 Hook::callAll('bb2diaspora', $text);
2139 DI::profiler()->stopRecording();
2144 * Pull out all #hashtags and @person tags from $string.
2146 * We also get @person@domain.com - which would make
2147 * the regex quite complicated as tags can also
2148 * end a sentence. So we'll run through our results
2149 * and strip the period from any tags which end with one.
2150 * Returns array of tags found, or empty array.
2152 * @param string $string Post content
2153 * @return array List of tag and person names
2155 public static function getTags(string $string): array
2157 DI::profiler()->startRecording('rendering');
2160 self::performWithEscapedTags($string, ['noparse', 'pre', 'code', 'img', 'attachment'], function ($string) use (&$ret) {
2161 // Convert hashtag links to hashtags
2162 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2 ', $string);
2164 // Force line feeds at bbtags
2165 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
2167 // ignore anything in a bbtag
2168 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
2170 // Match full names against @tags including the space between first and last
2171 // We will look these up afterward to see if they are full names or not recognisable.
2173 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
2174 foreach ($matches[1] as $match) {
2175 if (strstr($match, ']')) {
2176 // we might be inside a bbcode color tag - leave it alone
2180 if (substr($match, -1, 1) === '.') {
2181 $ret[] = substr($match, 0, -1);
2188 // Otherwise pull out single word tags. These can be @nickname, @first_last
2191 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?\']*[^\^ \x0D\x0A,;:?!\'.])/', $string, $matches)) {
2192 foreach ($matches[1] as $match) {
2193 if (strstr($match, ']')) {
2194 // we might be inside a bbcode color tag - leave it alone
2198 // try not to catch url fragments
2199 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
2208 DI::profiler()->stopRecording();
2209 return array_unique($ret);
2213 * Expand tags to URLs, checks the tag is at the start of a line or preceded by a non-word character
2215 * @param string $body HTML/BBCode
2216 * @return string body with expanded tags
2218 public static function expandTags(string $body): string
2220 return preg_replace_callback(
2221 "/(?<=\W|^)([!#@])([^\^ \x0D\x0A,;:?'\"]*[^\^ \x0D\x0A,;:?!'\".])/",
2222 function (array $match) {
2223 switch ($match[1]) {
2226 $contact = Contact::getByURL($match[2]);
2227 if (!empty($contact)) {
2228 return $match[1] . '[url=' . $contact['url'] . ']' . $contact['name'] . '[/url]';
2230 return $match[1] . $match[2];
2236 return $match[1] . '[url=' . DI::baseUrl() . '/search?tag=' . $match[2] . ']' . $match[2] . '[/url]';
2244 * Perform a custom function on a text after having escaped blocks enclosed in the provided tag list.
2246 * @param string $text HTML/BBCode
2247 * @param array $tagList A list of tag names, e.g ['noparse', 'nobb', 'pre']
2248 * @param callable $callback
2250 * @see Strings::performWithEscapedBlocks
2252 public static function performWithEscapedTags(string $text, array $tagList, callable $callback): string
2254 $tagList = array_map('preg_quote', $tagList);
2256 return Strings::performWithEscapedBlocks($text, '#\[(?:' . implode('|', $tagList) . ').*?\[/(?:' . implode('|', $tagList) . ')]#ism', $callback);
2260 * Replaces mentions in the provided message body in BBCode links for the provided user and network if any
2262 * @param string $body HTML/BBCode
2263 * @param int $profile_uid Profile user id
2264 * @param string $network Network name
2265 * @return string HTML/BBCode with inserted images
2266 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2267 * @throws \ImagickException
2269 public static function setMentions(string $body, $profile_uid = 0, $network = '')
2271 DI::profiler()->startRecording('rendering');
2272 $body = self::performWithEscapedTags($body, ['noparse', 'pre', 'code', 'img'], function ($body) use ($profile_uid, $network) {
2273 $tags = self::getTags($body);
2277 foreach ($tags as $tag) {
2278 $tag_type = substr($tag, 0, 1);
2280 if ($tag_type == Tag::TAG_CHARACTER[Tag::HASHTAG]) {
2285 * If we already tagged 'Robert Johnson', don't try and tag 'Robert'.
2286 * Robert Johnson should be first in the $tags array
2288 foreach ($tagged as $nextTag) {
2289 if (stristr($nextTag, $tag . ' ')) {
2294 if (($success = Item::replaceTag($body, $profile_uid, $tag, $network)) && $success['replaced']) {
2302 DI::profiler()->stopRecording();
2307 * @param string $author Author display name
2308 * @param string $profile Author profile URL
2309 * @param string $avatar Author profile picture URL
2310 * @param string $link Post source URL
2311 * @param string $posted Post created date
2312 * @param string|null $guid Post guid (if any)
2313 * @param string|null $uri Post uri (if any)
2315 * @TODO Rewrite to handle over whole record array
2317 public static function getShareOpeningTag(string $author, string $profile, string $avatar, string $link, string $posted, string $guid = null, string $uri = null): string
2319 DI::profiler()->startRecording('rendering');
2320 $header = "[share author='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $author) .
2321 "' profile='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $profile) .
2322 "' avatar='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $avatar) .
2323 "' link='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $link) .
2324 "' posted='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $posted);
2327 $header .= "' guid='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $guid);
2331 $header .= "' message_id='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $uri);
2336 DI::profiler()->stopRecording();
2341 * Returns the BBCode relevant to embed the provided URL in a post body.
2342 * For media type, it will return [img], [video] and [audio] tags.
2343 * For regular web pages, it will either output a [bookmark] tag if title and description were provided,
2344 * an [attachment] tag or a simple [url] tag depending on $tryAttachment.
2346 * @param string $url
2347 * @param bool $tryAttachment
2348 * @param string|null $title
2349 * @param string|null $description
2350 * @param string|null $tags
2352 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2353 * @see ParseUrl::getSiteinfoCached
2355 public static function embedURL(string $url, bool $tryAttachment = true, string $title = null, string $description = null, string $tags = null): string
2357 DI::profiler()->startRecording('rendering');
2358 DI::logger()->info($url);
2360 // If there is already some content information submitted we don't
2361 // need to parse the url for content.
2362 if (!empty($title) && !empty($description)) {
2363 $title = str_replace(["\r", "\n"], ['', ''], $title);
2365 $description = '[quote]' . trim($description) . '[/quote]' . "\n";
2368 if (!empty($tags)) {
2369 $arr_tags = ParseUrl::convertTagsToArray($tags);
2370 if (count($arr_tags)) {
2371 $str_tags = "\n" . implode(' ', $arr_tags) . "\n";
2375 $result = sprintf('[bookmark=%s]%s[/bookmark]%s', $url, ($title) ? $title : $url, $description) . $str_tags;
2377 DI::logger()->info('(unparsed): returns: ' . $result);
2379 DI::profiler()->stopRecording();
2383 $siteinfo = ParseUrl::getSiteinfoCached($url);
2385 if (in_array($siteinfo['type'], ['image', 'video', 'audio'])) {
2386 switch ($siteinfo['type']) {
2388 $bbcode = "\n" . '[video]' . $url . '[/video]' . "\n";
2391 $bbcode = "\n" . '[audio]' . $url . '[/audio]' . "\n";
2394 $bbcode = "\n" . '[img]' . $url . '[/img]' . "\n";
2398 DI::profiler()->stopRecording();
2402 unset($siteinfo['keywords']);
2404 // Bypass attachment if parse url for a comment
2405 if (!$tryAttachment) {
2406 DI::profiler()->stopRecording();
2407 return "\n" . '[url=' . $url . ']' . ($siteinfo['title'] ?? $url) . '[/url]';
2410 // Format it as BBCode attachment
2411 $bbcode = "\n" . PageInfo::getFooterFromData($siteinfo);
2412 DI::profiler()->stopRecording();