3 * @copyright Copyright (C) 2020, Friendica
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\Content\Text\HTMLPurifier_URIScheme_cid;
33 use Friendica\Core\Hook;
34 use Friendica\Core\Logger;
35 use Friendica\Core\Protocol;
36 use Friendica\Core\Renderer;
37 use Friendica\Core\System;
39 use Friendica\Model\Contact;
40 use Friendica\Model\Event;
41 use Friendica\Model\Photo;
42 use Friendica\Model\Tag;
43 use Friendica\Object\Image;
44 use Friendica\Protocol\Activity;
45 use Friendica\Util\Images;
46 use Friendica\Util\Map;
47 use Friendica\Util\ParseUrl;
48 use Friendica\Util\Proxy as ProxyUtils;
49 use Friendica\Util\Strings;
50 use Friendica\Util\XML;
54 // Update this value to the current date whenever changes are made to BBCode::convert
55 const VERSION = '2020-12-18-video-embeds';
64 const ACTIVITYPUB = 9;
67 * Fetches attachment data that were generated the old way
69 * @param string $body Message body
71 * 'type' -> Message type ('link', 'video', 'photo')
72 * 'text' -> Text before the shared message
73 * 'after' -> Text after the shared message
74 * 'image' -> Preview image of the message
75 * 'url' -> Url to the attached message
76 * 'title' -> Title of the attachment
77 * 'description' -> Description of the attachment
78 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
80 private static function getOldAttachmentData($body)
84 // Simplify image codes
85 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
87 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
88 foreach ($attached as $data) {
89 if (!in_array($data[1], ['type-link', 'type-video', 'type-photo'])) {
93 $post['type'] = substr($data[1], 5);
95 $pos = strpos($body, $data[0]);
97 $post['text'] = trim(substr($body, 0, $pos));
98 $post['after'] = trim(substr($body, $pos + strlen($data[0])));
100 $post['text'] = trim(str_replace($data[0], '', $body));
104 $attacheddata = $data[2];
106 if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
108 $picturedata = Images::getInfoFromURLCached($matches[1]);
111 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
112 $post['image'] = $matches[1];
114 $post['preview'] = $matches[1];
119 if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
120 $post['url'] = $matches[1];
121 $post['title'] = $matches[2];
123 if (!empty($post['url']) && (in_array($post['type'], ['link', 'video']))
124 && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
125 $post['url'] = $matches[1];
128 // Search for description
129 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
130 $post['description'] = $matches[1];
138 * Fetches attachment data that were generated with the "attachment" element
140 * @param string $body Message body
142 * 'type' -> Message type ('link', 'video', 'photo')
143 * 'text' -> Text before the shared message
144 * 'after' -> Text after the shared message
145 * 'image' -> Preview image of the message
146 * 'url' -> Url to the attached message
147 * 'title' -> Title of the attachment
148 * 'description' -> Description of the attachment
149 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
151 public static function getAttachmentData($body)
159 'provider_name' => '',
160 'provider_url' => '',
165 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
166 return self::getOldAttachmentData($body);
169 $attributes = $match[2];
171 $data['text'] = trim($match[1]);
174 preg_match("/type='(.*?)'/ism", $attributes, $matches);
175 if (!empty($matches[1])) {
176 $type = strtolower($matches[1]);
179 preg_match('/type="(.*?)"/ism', $attributes, $matches);
180 if (!empty($matches[1])) {
181 $type = strtolower($matches[1]);
188 if (!in_array($type, ['link', 'audio', 'photo', 'video'])) {
193 $data['type'] = $type;
197 preg_match("/url='(.*?)'/ism", $attributes, $matches);
198 if (!empty($matches[1])) {
202 preg_match('/url="(.*?)"/ism', $attributes, $matches);
203 if (!empty($matches[1])) {
208 $data['url'] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
212 preg_match("/title='(.*?)'/ism", $attributes, $matches);
213 if (!empty($matches[1])) {
214 $title = $matches[1];
217 preg_match('/title="(.*?)"/ism', $attributes, $matches);
218 if (!empty($matches[1])) {
219 $title = $matches[1];
223 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
224 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
225 $title = str_replace(['[', ']'], ['[', ']'], $title);
226 $data['title'] = $title;
230 preg_match("/image='(.*?)'/ism", $attributes, $matches);
231 if (!empty($matches[1])) {
232 $image = $matches[1];
235 preg_match('/image="(.*?)"/ism', $attributes, $matches);
236 if (!empty($matches[1])) {
237 $image = $matches[1];
241 $data['image'] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
245 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
246 if (!empty($matches[1])) {
247 $preview = $matches[1];
250 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
251 if (!empty($matches[1])) {
252 $preview = $matches[1];
255 if ($preview != '') {
256 $data['preview'] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
259 $data['description'] = trim($match[3]);
261 $data['after'] = trim($match[4]);
263 $parts = parse_url($data['url']);
264 if (!empty($parts['scheme']) && !empty($parts['host'])) {
265 $data['provider_name'] = $parts['host'];
266 $data['provider_url'] = $parts['scheme'] . '://' . $parts['host'];
268 if (!empty($parts['port'])) {
269 $data['provider_url'] .= ':' . $parts['port'];
276 public static function getAttachedData($body, $item = [])
280 - type: link, video, photo
288 $has_title = !empty($item['title']);
289 $plink = $item['plink'] ?? '';
290 $post = self::getAttachmentData($body);
292 // Get all linked images with alternative image description
293 if (preg_match_all("/\[img=([^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
294 foreach ($pictures as $picture) {
295 if (Photo::isLocal($picture[1])) {
296 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2]];
299 if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
300 $post['image_description'] = $post['images'][0]['description'];
304 if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
305 foreach ($pictures as $picture) {
306 if (Photo::isLocal($picture[1])) {
307 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => ''];
312 // if nothing is found, it maybe having an image.
313 if (!isset($post['type'])) {
314 // Simplify image codes
315 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
316 $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
317 $post['text'] = $body;
319 if (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $body, $pictures, PREG_SET_ORDER)) {
320 if ((count($pictures) == 1) && !$has_title) {
321 if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
322 // Replace the preview picture with the real picture
323 $url = str_replace('-1.', '-0.', $pictures[0][2]);
324 $data = ['url' => $url, 'type' => 'photo'];
326 // Checking, if the link goes to a picture
327 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
331 // Sometimes photo posts to the own album are not detected at the start.
332 // So we seem to cannot use the cache for these cases. That's strange.
333 if (($data['type'] != 'photo') && strstr($pictures[0][1], "/photos/")) {
334 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
337 if ($data['type'] == 'photo') {
338 $post['type'] = 'photo';
339 if (isset($data['images'][0])) {
340 $post['image'] = $data['images'][0]['src'];
341 $post['url'] = $data['url'];
343 $post['image'] = $data['url'];
346 $post['preview'] = $pictures[0][2];
347 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
349 $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
350 if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
351 $post['type'] = 'photo';
352 $post['image'] = $pictures[0][1];
353 $post['preview'] = $pictures[0][2];
354 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
357 } elseif (count($pictures) > 0) {
358 $post['type'] = 'link';
359 $post['url'] = $plink;
360 $post['image'] = $pictures[0][2];
361 $post['text'] = $body;
363 foreach ($pictures as $picture) {
364 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
367 } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
368 if ((count($pictures) == 1) && !$has_title) {
369 $post['type'] = 'photo';
370 $post['image'] = $pictures[0][1];
371 $post['text'] = str_replace($pictures[0][0], '', $body);
372 } elseif (count($pictures) > 0) {
373 $post['type'] = 'link';
374 $post['url'] = $plink;
375 $post['image'] = $pictures[0][1];
376 $post['text'] = $body;
378 foreach ($pictures as $picture) {
379 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
384 // Test for the external links
385 preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER);
386 preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $links2, PREG_SET_ORDER);
388 $links = array_merge($links1, $links2);
390 // If there is only a single one, then use it.
391 // This should cover link posts via API.
392 if ((count($links) == 1) && !isset($post['preview']) && !$has_title) {
393 $post['type'] = 'link';
394 $post['url'] = $links[0][1];
397 // Simplify "video" element
398 $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
400 // Now count the number of external media links
401 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
402 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
403 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER);
404 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $links4, PREG_SET_ORDER);
406 // Add them to the other external links
407 $links = array_merge($links, $links1, $links2, $links3, $links4);
409 // Are there more than one?
410 if (count($links) > 1) {
411 // The post will be the type "text", which means a blog post
412 unset($post['type']);
413 $post['url'] = $plink;
416 if (!isset($post['type'])) {
417 $post['type'] = "text";
418 $post['text'] = trim($body);
420 } elseif (isset($post['url']) && ($post['type'] == 'video')) {
421 $data = ParseUrl::getSiteinfoCached($post['url'], true);
423 if (isset($data['images'][0])) {
424 $post['image'] = $data['images'][0]['src'];
432 * Remove [attachment] BBCode and replaces it with a regular [url]
434 * @param string $body
435 * @param boolean $no_link_desc No link description
437 * @return string with replaced body
439 public static function removeAttachment($body, $no_link_desc = false)
441 return preg_replace_callback("/\s*\[attachment (.*)\](.*?)\[\/attachment\]\s*/ism",
442 function ($match) use ($no_link_desc) {
443 $attach_data = self::getAttachmentData($match[0]);
444 if (empty($attach_data['url'])) {
446 } elseif (empty($attach_data['title']) || $no_link_desc) {
447 return "\n[url]" . $attach_data['url'] . "[/url]\n";
449 return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
455 * Converts a BBCode text into plaintext
458 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
462 public static function toPlaintext($text, $keep_urls = true)
464 $naked_text = HTML::toPlaintext(self::convert($text, false, 0, true), 0, !$keep_urls);
469 private static function proxyUrl($image, $simplehtml = self::INTERNAL)
471 // Only send proxied pictures to API and for internal display
472 if (in_array($simplehtml, [self::INTERNAL, self::API])) {
473 return ProxyUtils::proxifyUrl($image);
480 * This function changing the visual size (not the real size) of images.
481 * The function does not work for pictures with an alternate text description.
482 * This could only be changed by using some new "img" BBCode format.
484 * @param string $srctext The body with images
485 * @return string The body with possibly scaled images
487 public static function scaleExternalImages(string $srctext)
491 // Simplify image links
492 $s = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $s);
495 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
497 foreach ($matches as $mtch) {
498 Logger::log('scale_external_image: ' . $mtch[1]);
500 $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
501 if (stristr($mtch[1], $hostname)) {
505 $curlResult = DI::httpRequest()->get($mtch[1]);
506 if (!$curlResult->isSuccess()) {
510 $i = $curlResult->getBody();
511 $type = $curlResult->getContentType();
512 $type = Images::getMimeTypeByData($i, $mtch[1], $type);
515 $Image = new Image($i, $type);
516 if ($Image->isValid()) {
517 $orig_width = $Image->getWidth();
518 $orig_height = $Image->getHeight();
520 if ($orig_width > 640 || $orig_height > 640) {
521 $Image->scaleDown(640);
522 $new_width = $Image->getWidth();
523 $new_height = $Image->getHeight();
524 Logger::info('External images scaled', ['orig_width' => $orig_width, 'new_width' => $new_width, 'orig_height' => $orig_height, 'new_height' => $new_height, 'match' => $mtch[0]]);
527 '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
531 Logger::info('New string', ['image' => $s]);
542 * Truncates imported message body string length to max_import_size
544 * The purpose of this function is to apply system message length limits to
545 * imported messages without including any embedded photos in the length
547 * @param string $body
549 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
551 public static function limitBodySize($body)
553 $maxlen = DI::config()->get('config', 'max_import_size', 0);
555 // If the length of the body, including the embedded images, is smaller
556 // than the maximum, then don't waste time looking for the images
557 if ($maxlen && (strlen($body) > $maxlen)) {
559 Logger::info('the total body length exceeds the limit', ['maxlen' => $maxlen, 'body_len' => strlen($body)]);
565 $img_start = strpos($orig_body, '[img');
566 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
567 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
568 while (($img_st_close !== false) && ($img_end !== false)) {
570 $img_st_close++; // make it point to AFTER the closing bracket
571 $img_end += $img_start;
572 $img_end += strlen('[/img]');
574 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
575 // This is an embedded image
577 if (($textlen + $img_start) > $maxlen) {
578 if ($textlen < $maxlen) {
579 Logger::info('the limit happens before an embedded image');
580 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
584 $new_body = $new_body . substr($orig_body, 0, $img_start);
585 $textlen += $img_start;
588 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
591 if (($textlen + $img_end) > $maxlen) {
592 if ($textlen < $maxlen) {
593 Logger::info('the limit happens before the end of a non-embedded image');
594 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
598 $new_body = $new_body . substr($orig_body, 0, $img_end);
599 $textlen += $img_end;
602 $orig_body = substr($orig_body, $img_end);
604 if ($orig_body === false) {
605 // in case the body ends on a closing image tag
609 $img_start = strpos($orig_body, '[img');
610 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
611 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
614 if (($textlen + strlen($orig_body)) > $maxlen) {
615 if ($textlen < $maxlen) {
616 Logger::info('the limit happens after the end of the last image');
617 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
620 Logger::info('the text size with embedded images extracted did not violate the limit');
621 $new_body = $new_body . $orig_body;
631 * Processes [attachment] tags
633 * Note: Can produce a [bookmark] tag in the returned string
635 * @param string $text
636 * @param integer $simplehtml
637 * @param bool $tryoembed
639 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
641 private static function convertAttachment($text, $simplehtml = self::INTERNAL, $tryoembed = true)
643 $data = self::getAttachmentData($text);
644 if (empty($data) || empty($data['url'])) {
648 if (isset($data['title'])) {
649 $data['title'] = strip_tags($data['title']);
650 $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
652 $data['title'] = null;
655 if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
656 $data['preview'] = $data['image'];
662 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
663 $return = OEmbed::getHTML($data['url'], $data['title']);
665 throw new Exception('OEmbed is disabled for this attachment.');
667 } catch (Exception $e) {
668 $data['title'] = ($data['title'] ?? '') ?: $data['url'];
670 if ($simplehtml != self::CONNECTORS) {
671 $return = sprintf('<div class="type-%s">', $data['type']);
674 if (!empty($data['title']) && !empty($data['url'])) {
675 if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
676 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
678 if (!empty($data['image'])) {
679 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br>', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
680 } elseif (!empty($data['preview'])) {
681 $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), $data['title']);
683 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
687 if (!empty($data['description']) && $data['description'] != $data['title']) {
688 // Sanitize the HTML by converting it to BBCode
689 $bbcode = HTML::toBBCode($data['description']);
690 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
693 if (!empty($data['url'])) {
694 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
697 if ($simplehtml != self::CONNECTORS) {
702 return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
705 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
707 $data = self::getAttachmentData($Text);
712 return $data['text'] . ($data['after'] ?? '');
715 $title = htmlentities($data['title'] ?? '', ENT_QUOTES, 'UTF-8', false);
716 $text = htmlentities($data['text'], ENT_QUOTES, 'UTF-8', false);
717 if ($plaintext || (($title != '') && strstr($text, $title))) {
718 $data['title'] = $data['url'];
719 } elseif (($text != '') && strstr($title, $text)) {
720 $data['text'] = $data['title'];
721 $data['title'] = $data['url'];
724 if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
725 return $data['title'] . $data['after'];
728 // If the link already is included in the post, don't add it again
729 if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
730 return $data['text'] . $data['after'];
733 $text = $data['text'];
735 if (!empty($data['url']) && !empty($data['title'])) {
736 $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
737 } elseif (!empty($data['url'])) {
738 $text .= "\n[url]" . $data['url'] . '[/url]';
741 return $text . "\n" . $data['after'];
745 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
747 * @param array $match Array with the matching values
748 * @return string reformatted link including HTML codes
750 private static function convertUrlForActivityPubCallback($match)
754 if (isset($match[2]) && ($match[1] != $match[2])) {
758 $parts = parse_url($url);
759 if (!isset($parts['scheme'])) {
763 return self::convertUrlForActivityPub($url);
767 * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
769 * @param string $url URL that is about to be reformatted
770 * @return string reformatted link including HTML codes
772 private static function convertUrlForActivityPub($url)
774 $html = '<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>';
775 return sprintf($html, $url, self::getStyledURL($url));
779 * Converts an URL in a nicer format (without the scheme and possibly shortened)
781 * @param string $url URL that is about to be reformatted
782 * @return string reformatted link
784 private static function getStyledURL($url)
786 $parts = parse_url($url);
787 $scheme = $parts['scheme'] . '://';
788 $styled_url = str_replace($scheme, '', $url);
790 if (strlen($styled_url) > 30) {
791 $styled_url = substr($styled_url, 0, 30) . "…";
798 * [noparse][i]italic[/i][/noparse] turns into
799 * [noparse][ i ]italic[ /i ][/noparse],
800 * to hide them from parser.
802 private static function escapeNoparseCallback($match)
804 $whole_match = $match[0];
805 $captured = $match[1];
806 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
807 $new_str = str_replace($captured, $spacefied, $whole_match);
812 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
813 * now turns back and the [noparse] tags are trimed
814 * returning [i]italic[/i]
816 private static function unescapeNoparseCallback($match)
818 $captured = $match[1];
819 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
824 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
827 * @param string $text Text to search
828 * @param string $name Tag name
829 * @param int $occurrences Number of first occurrences to skip
830 * @return boolean|array
832 public static function getTagPosition($text, $name, $occurrences = 0)
834 if ($occurrences < 0) {
839 for ($i = 0; $i <= $occurrences; $i++) {
840 if ($start_open !== false) {
841 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
845 if ($start_open === false) {
849 $start_equal = strpos($text, '=', $start_open);
850 $start_close = strpos($text, ']', $start_open);
852 if ($start_close === false) {
858 $end_open = strpos($text, '[/' . $name . ']', $start_close);
860 if ($end_open === false) {
866 'open' => $start_open,
867 'close' => $start_close
871 'close' => $end_open + strlen('[/' . $name . ']')
875 if ($start_equal !== false) {
876 $res['start']['equal'] = $start_equal + 1;
883 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
885 * @param string $pattern Preg pattern string
886 * @param string $replace Preg replace string
887 * @param string $name BBCode tag name
888 * @param string $text Text to search
891 public static function pregReplaceInTag($pattern, $replace, $name, $text)
894 $pos = self::getTagPosition($text, $name, $occurrences);
895 while ($pos !== false && $occurrences++ < 1000) {
896 $start = substr($text, 0, $pos['start']['open']);
897 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
898 $end = substr($text, $pos['end']['close']);
899 if ($end === false) {
903 $subject = preg_replace($pattern, $replace, $subject);
904 $text = $start . $subject . $end;
906 $pos = self::getTagPosition($text, $name, $occurrences);
912 private static function extractImagesFromItemBody($body)
919 $img_start = strpos($orig_body, '[img');
920 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
921 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
922 while (($img_st_close !== false) && ($img_end !== false)) {
923 $img_st_close++; // make it point to AFTER the closing bracket
924 $img_end += $img_start;
926 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
927 // This is an embedded image
928 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
929 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
933 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
936 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
938 if ($orig_body === false) {
939 // in case the body ends on a closing image tag
943 $img_start = strpos($orig_body, '[img');
944 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
945 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
948 $new_body = $new_body . $orig_body;
950 return ['body' => $new_body, 'images' => $saved_image];
953 private static function interpolateSavedImagesIntoItemBody($body, array $images)
958 foreach ($images as $image) {
959 // We're depending on the property of 'foreach' (specified on the PHP website) that
960 // it loops over the array starting from the first element and going sequentially
961 // to the last element
962 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
963 '<img src="' . self::proxyUrl($image) . '" alt="' . DI::l10n()->t('Image/photo') . '" />', $newbody);
971 * This function converts a [share] block to text according to a provided callback function whose signature is:
973 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
976 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
978 * - $author_contact is a contact record array
979 * - $content is the inner content of the [share] block
980 * - $is_quote_share indicates whether there's any content before the [share] block
981 * - Return value is the string that should replace the [share] block in the provided text
983 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
985 * @param string $text A BBCode string
986 * @param callable $callback
987 * @return string The BBCode string with all [share] blocks replaced
989 public static function convertShare($text, callable $callback)
991 $return = preg_replace_callback(
992 "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
993 function ($match) use ($callback) {
994 $attribute_string = $match[2];
996 foreach (['author', 'profile', 'avatar', 'link', 'posted', 'guid'] as $field) {
997 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
998 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
1001 $author_contact = Contact::getByURL($attributes['profile'], false, ['url', 'addr', 'name', 'micro']);
1002 $author_contact['url'] = ($author_contact['url'] ?? $attributes['profile']);
1003 $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']);
1005 $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author'];
1006 $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
1007 $attributes['profile'] = ($author_contact['url'] ?? '') ?: $attributes['profile'];
1009 if ($attributes['avatar']) {
1010 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
1013 $content = preg_replace(Strings::autoLinkRegEx(), '<a href="$1">$1</a>', $match[3]);
1015 return $match[1] . $callback($attributes, $author_contact, $content, trim($match[1]) != '');
1024 * Default [share] tag conversion callback
1026 * Note: Can produce a [bookmark] tag in the output
1028 * @see BBCode::convertShare()
1029 * @param array $attributes [share] block attribute values
1030 * @param array $author_contact Contact row of the shared author
1031 * @param string $content Inner content of the [share] block
1032 * @param boolean $is_quote_share Whether there is content before the [share] block
1033 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
1035 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1037 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
1039 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
1041 switch ($simplehtml) {
1043 $text = ($is_quote_share? '<br>' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
1045 case self::DIASPORA:
1046 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
1047 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
1049 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
1051 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
1052 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
1055 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
1057 if (empty($attributes['posted']) && !empty($attributes['link'])) {
1058 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
1063 case self::CONNECTORS:
1064 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
1065 $headline .= DI::l10n()->t('<a href="%1$s" target="_blank" rel="noopener noreferrer">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
1066 $headline .= ':</b></p>' . "\n";
1068 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
1072 $text = ($is_quote_share? '<br>' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
1074 case self::ACTIVITYPUB:
1075 $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>';
1076 $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
1079 $text = ($is_quote_share? "\n" : '');
1081 $contact = Contact::getByURL($attributes['profile'], false, ['network']);
1082 $network = $contact['network'] ?? Protocol::PHANTOM;
1084 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1085 $text .= Renderer::replaceMacros($tpl, [
1086 '$profile' => $attributes['profile'],
1087 '$avatar' => $attributes['avatar'],
1088 '$author' => $attributes['author'],
1089 '$link' => $attributes['link'],
1090 '$link_title' => DI::l10n()->t('link to source'),
1091 '$posted' => $attributes['posted'],
1092 '$guid' => $attributes['guid'],
1093 '$network_name' => ContactSelector::networkToName($network, $attributes['profile']),
1094 '$network_icon' => ContactSelector::networkToIcon($network, $attributes['profile']),
1095 '$content' => self::setMentions(trim($content), 0, $network),
1103 private static function removePictureLinksCallback($match)
1105 $cache_key = 'remove:' . $match[1];
1106 $text = DI::cache()->get($cache_key);
1108 if (is_null($text)) {
1111 $stamp1 = microtime(true);
1113 $ch = @curl_init($match[1]);
1114 @curl_setopt($ch, CURLOPT_NOBODY, true);
1115 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1116 @curl_setopt($ch, CURLOPT_USERAGENT, DI::httpRequest()->getUserAgent());
1118 $curl_info = @curl_getinfo($ch);
1120 DI::profiler()->saveTimestamp($stamp1, "network");
1122 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1123 $text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
1125 $text = "[url=" . $match[2] . ']' . $match[2] . "[/url]";
1127 // if its not a picture then look if its a page that contains a picture link
1128 $body = DI::httpRequest()->fetch($match[1]);
1130 $doc = new DOMDocument();
1131 @$doc->loadHTML($body);
1132 $xpath = new DOMXPath($doc);
1133 $list = $xpath->query("//meta[@name]");
1134 foreach ($list as $node) {
1137 if ($node->attributes->length) {
1138 foreach ($node->attributes as $attribute) {
1139 $attr[$attribute->name] = $attribute->value;
1143 if (strtolower($attr['name']) == 'twitter:image') {
1144 $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
1148 DI::cache()->set($cache_key, $text);
1154 private static function expandLinksCallback($match)
1156 if (($match[3] == '') || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1157 return ($match[1] . "[url]" . $match[2] . "[/url]");
1159 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1163 private static function cleanPictureLinksCallback($match)
1167 // When the picture link is the own photo path then we can avoid fetching the link
1168 $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
1169 if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
1170 if (!empty($match[3])) {
1171 $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
1173 $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
1178 $cache_key = 'clean:' . $match[1];
1179 $text = DI::cache()->get($cache_key);
1180 if (!is_null($text)) {
1184 // Only fetch the header, not the content
1185 $stamp1 = microtime(true);
1187 $ch = @curl_init($match[1]);
1188 @curl_setopt($ch, CURLOPT_NOBODY, true);
1189 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1190 @curl_setopt($ch, CURLOPT_USERAGENT, DI::httpRequest()->getUserAgent());
1192 $curl_info = @curl_getinfo($ch);
1194 DI::profiler()->saveTimestamp($stamp1, "network");
1196 // if its a link to a picture then embed this picture
1197 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1198 $text = '[img]' . $match[1] . '[/img]';
1200 if (!empty($match[3])) {
1201 $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
1203 $text = '[img]' . $match[2] . '[/img]';
1206 // if its not a picture then look if its a page that contains a picture link
1207 $body = DI::httpRequest()->fetch($match[1]);
1209 $doc = new DOMDocument();
1210 @$doc->loadHTML($body);
1211 $xpath = new DOMXPath($doc);
1212 $list = $xpath->query("//meta[@name]");
1213 foreach ($list as $node) {
1215 if ($node->attributes->length) {
1216 foreach ($node->attributes as $attribute) {
1217 $attr[$attribute->name] = $attribute->value;
1221 if (strtolower($attr['name']) == "twitter:image") {
1222 if (!empty($match[3])) {
1223 $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
1225 $text = "[img]" . $attr['content'] . "[/img]";
1230 DI::cache()->set($cache_key, $text);
1235 public static function cleanPictureLinks($text)
1237 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1238 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $return);
1242 public static function removeLinks(string $bbcode)
1244 $bbcode = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $1 ', $bbcode);
1245 $bbcode = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $bbcode);
1247 $bbcode = preg_replace('/[@!#]\[url\=.*?\].*?\[\/url\]/ism', '', $bbcode);
1248 $bbcode = preg_replace("/\[url=[^\[\]]*\](.*)\[\/url\]/Usi", ' $1 ', $bbcode);
1249 $bbcode = preg_replace('/[@!#]?\[url.*?\[\/url\]/ism', '', $bbcode);
1254 * Converts a BBCode message to HTML message
1256 * BBcode 2 HTML was written by WAY2WEB.net
1257 * extended to work with Mistpark/Friendica - Mike Macgirvin
1259 * Simple HTML values meaning:
1260 * - 0: Friendica display
1262 * - 2: Used for Windows Phone push, Friendica API
1263 * - 3: Used before converting to Markdown in bb2diaspora.php
1264 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1267 * - 7: Used for dfrn, OStatus
1268 * - 8: Used for Twitter, WP backlink text setting
1271 * @param string $text
1272 * @param bool $try_oembed
1273 * @param int $simple_html
1274 * @param bool $for_plaintext
1276 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1278 public static function convert(string $text = null, $try_oembed = true, $simple_html = self::INTERNAL, $for_plaintext = false)
1280 // Accounting for null default column values
1281 if (is_null($text) || $text === '') {
1285 Hook::callAll('bbcode', $text);
1289 $text = self::performWithEscapedTags($text, ['code'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
1290 $text = self::performWithEscapedTags($text, ['noparse', 'nobb', 'pre'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
1292 * preg_match_callback function to replace potential Oembed tags with Oembed content
1294 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1296 * $match[2] = $title or absent
1298 $try_oembed_callback = function ($match)
1301 $title = $match[2] ?? null;
1304 $return = OEmbed::getHTML($url, $title);
1305 } catch (Exception $ex) {
1306 $return = $match[0];
1312 // Remove the abstract element. It is a non visible element.
1313 $text = self::stripAbstract($text);
1315 // Line ending normalisation
1316 $text = str_replace("\r\n", "\n", $text);
1318 // Move new lines outside of tags
1319 $text = preg_replace("#\[(\w*)](\n*)#ism", '$2[$1]', $text);
1320 $text = preg_replace("#(\n*)\[/(\w*)]#ism", '[/$2]$1', $text);
1322 // Extract the private images which use data urls since preg has issues with
1323 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1324 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1326 $extracted = self::extractImagesFromItemBody($text);
1327 $text = $extracted['body'];
1328 $saved_image = $extracted['images'];
1330 // If we find any event code, turn it into an event.
1331 // After we're finished processing the bbcode we'll
1332 // replace all of the event code with a reformatted version.
1334 $ev = Event::fromBBCode($text);
1336 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1337 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1339 $text = str_replace("<", "<", $text);
1340 $text = str_replace(">", ">", $text);
1342 // remove some newlines before the general conversion
1343 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1344 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1346 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1348 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1351 // Remove linefeeds inside of the table elements. See issue #6799
1352 $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
1353 "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
1354 "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
1355 "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "];
1356 $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
1357 "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
1358 "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
1359 "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"];
1362 $text = str_replace($search, $replace, $text);
1363 } while ($oldtext != $text);
1365 // Replace these here only once
1366 $search = ["\n[table]", "[/table]\n"];
1367 $replace = ["[table]", "[/table]"];
1368 $text = str_replace($search, $replace, $text);
1370 // Trim new lines regardless of the system.remove_multiplicated_lines config value
1371 $text = trim($text, "\n");
1373 // removing multiplicated newlines
1374 if (DI::config()->get('system', 'remove_multiplicated_lines')) {
1375 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[*]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1376 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1377 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[*]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1378 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1381 $text = str_replace($search, $replace, $text);
1382 } while ($oldtext != $text);
1385 // Add HTML new lines
1386 $text = str_replace("\n", '<br>', $text);
1388 /// @todo Have a closer look at the different html modes
1389 // Handle attached links or videos
1390 if ($simple_html == self::ACTIVITYPUB) {
1391 $text = self::removeAttachment($text);
1392 } elseif (!in_array($simple_html, [self::INTERNAL, self::CONNECTORS])) {
1393 $text = self::removeAttachment($text, true);
1395 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1398 $nosmile = strpos($text, '[nosmile]') !== false;
1399 $text = str_replace('[nosmile]', '', $text);
1401 // Replace non graphical smilies for external posts
1402 if (!$nosmile && !$for_plaintext) {
1403 $text = self::performWithEscapedTags($text, ['img'], function ($text) {
1404 return Smilies::replace($text);
1408 // leave open the posibility of [map=something]
1409 // this is replaced in Item::prepareBody() which has knowledge of the item location
1410 if (strpos($text, '[/map]') !== false) {
1411 $text = preg_replace_callback(
1412 "/\[map\](.*?)\[\/map\]/ism",
1413 function ($match) use ($simple_html) {
1414 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1420 if (strpos($text, '[map=') !== false) {
1421 $text = preg_replace_callback(
1422 "/\[map=(.*?)\]/ism",
1423 function ($match) use ($simple_html) {
1424 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1430 if (strpos($text, '[map]') !== false) {
1431 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1434 // Check for headers
1435 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1436 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1437 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1438 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1439 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1440 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1442 // Check for paragraph
1443 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1445 // Check for bold text
1446 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1448 // Check for Italics text
1449 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1451 // Check for Underline text
1452 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1454 // Check for strike-through text
1455 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1457 // Check for over-line text
1458 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1460 // Check for colored text
1461 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1463 // Check for sized text
1464 // [size=50] --> font-size: 50px (with the unit).
1465 if ($simple_html != self::DIASPORA) {
1466 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", '<span style="font-size:$1px;line-height:normal;">$2</span>', $text);
1467 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", '<span style="font-size:$1;line-height:normal;">$2</span>', $text);
1469 // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
1470 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text);
1474 // Check for centered text
1475 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", '<div style="text-align:center;">$1</div>', $text);
1477 // Check for list text
1478 $text = str_replace("[*]", "<li>", $text);
1480 // Check for style sheet commands
1481 $text = preg_replace("(\[style=(.*?)\](.*?)\[\/style\])ism", '<span style="$1">$2</span>', $text);
1483 // Check for CSS classes
1484 $text = preg_replace("(\[class=(.*?)\](.*?)\[\/class\])ism", '<span class="$1">$2</span>', $text);
1486 // handle nested lists
1489 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1490 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1491 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1492 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1493 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1494 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1495 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1496 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1497 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1498 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1499 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1500 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1501 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1502 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1505 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1506 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1507 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1508 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1510 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1511 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1513 $text = str_replace('[hr]', '<hr />', $text);
1515 if (!$for_plaintext) {
1516 $text = self::performWithEscapedTags($text, ['url', 'img', 'audio', 'video', 'youtube', 'vimeo', 'share', 'attachment', 'iframe', 'bookmark'], function ($text) {
1517 return preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1521 // Check for font change text
1522 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1524 // Declare the format for [spoiler] layout
1525 $SpoilerLayout = '<details class="spoiler"><summary>' . DI::l10n()->t('Click to open/close') . '</summary>$1</details>';
1527 // Check for [spoiler] text
1528 // handle nested quotes
1530 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1531 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", $SpoilerLayout, $text);
1534 // Check for [spoiler=Title] text
1536 // handle nested quotes
1538 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1539 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1540 '<details class="spoiler"><summary>$1</summary>$2</details>',
1544 // Declare the format for [quote] layout
1545 $QuoteLayout = '<blockquote>$1</blockquote>';
1547 // Check for [quote] text
1548 // handle nested quotes
1550 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1551 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1554 // Check for [quote=Author] text
1556 $t_wrote = DI::l10n()->t('$1 wrote:');
1558 // handle nested quotes
1560 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1561 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1562 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1567 // [img=widthxheight]image source[/img]
1568 $text = preg_replace_callback(
1569 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1570 function ($matches) use ($simple_html) {
1571 if (strpos($matches[3], "data:image/") === 0) {
1575 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1576 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1581 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1582 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1584 $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/ism",
1585 function ($matches) use ($simple_html) {
1586 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1587 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1588 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '" title="' . $matches[2] . '">';
1593 // [img]pathtoimage[/img]
1594 $text = preg_replace_callback(
1595 "/\[img\](.*?)\[\/img\]/ism",
1596 function ($matches) use ($simple_html) {
1597 if (strpos($matches[1], "data:image/") === 0) {
1601 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1602 return "[img]" . $matches[1] . "[/img]";
1607 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1608 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1610 $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);
1611 $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>', $text);
1612 //$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);
1614 // Simplify "video" element
1615 $text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
1618 // html5 video and audio
1619 $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism",
1620 '<video src="$1" controls width="' . $a->videowidth . '" height="' . $a->videoheight . '" loop="true"><a href="$1">$1</a></video>', $text);
1622 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1623 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1625 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1626 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1627 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", '<audio src="$1" controls><a href="$1">$1</a></audio>', $text);
1629 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1630 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1631 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1632 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1635 // Backward compatibility, [iframe] support has been removed in version 2020.12
1636 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1638 // Youtube extensions
1640 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1641 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1642 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1645 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1646 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1647 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1650 $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);
1652 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1653 '<a href="https://www.youtube.com/watch?v=$1" target="_blank" rel="noopener noreferrer">https://www.youtube.com/watch?v=$1</a>', $text);
1657 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1658 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1661 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1662 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1665 $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);
1667 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1668 '<a href="https://vimeo.com/$1" target="_blank" rel="noopener noreferrer">https://vimeo.com/$1</a>', $text);
1672 $text = OEmbed::BBCode2HTML($text);
1674 // Avoid triple linefeeds through oembed
1675 $text = str_replace("<br style='clear:left'></span><br><br>", "<br style='clear:left'></span><br>", $text);
1677 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1678 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1679 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1680 // start which is always required). Allow desc with a missing summary for compatibility.
1682 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1683 $sub = Event::getHTML($ev, $simple_html);
1685 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1686 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1687 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1688 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1689 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1690 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1691 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1694 if (!$for_plaintext && DI::config()->get('system', 'big_emojis') && ($simple_html != self::DIASPORA)) {
1695 $conv = html_entity_decode(str_replace([' ', "\n", "\r"], '', $text));
1696 // Emojis are always 4 byte Unicode characters
1697 if (!empty($conv) && (strlen($conv) / mb_strlen($conv) == 4)) {
1698 $text = '<span style="font-size: xx-large; line-height: normal;">' . $text . '</span>';
1702 if (!$for_plaintext) {
1703 if (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1704 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1705 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1708 $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1709 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1712 // Remove all hashtag addresses
1713 if ($simple_html && !in_array($simple_html, [self::DIASPORA, self::OSTATUS, self::ACTIVITYPUB])) {
1714 $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1715 } elseif ($simple_html == self::DIASPORA) {
1716 // The ! is converted to @ since Diaspora only understands the @
1717 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1718 '@<a href="$2">$3</a>',
1720 } elseif (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1721 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1722 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1724 } elseif (!$simple_html) {
1725 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1726 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1730 // Bookmarks in red - will be converted to bookmarks in friendica
1731 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1732 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1733 $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1734 "[bookmark=$1]$2[/bookmark]", $text);
1736 if (in_array($simple_html, [self::API, self::OSTATUS, self::TWITTER])) {
1737 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1738 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1739 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1742 // Perform URL Search
1744 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1747 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1749 // Handle Diaspora posts
1750 $text = preg_replace_callback(
1751 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1753 return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1757 $text = preg_replace_callback(
1758 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1760 return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1764 // Server independent link to posts and comments
1765 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1766 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1767 $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
1771 * - #[url=<anything>]<term>[/url]
1772 * - [url=<anything>]#<term>[/url]
1774 $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) use ($simple_html) {
1775 if ($simple_html == BBCode::ACTIVITYPUB) {
1776 return '<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1777 . '" data-tag="' . XML::escape($matches[1]) . '" rel="tag ugc">#'
1778 . XML::escape($matches[1]) . '</a>';
1780 return '#<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1781 . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
1782 . XML::escape($matches[1]) . '</a>';
1786 // We need no target="_blank" rel="noopener noreferrer" for local links
1787 // convert links start with DI::baseUrl() as local link without the target="_blank" rel="noopener noreferrer" attribute
1788 $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
1789 $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1790 $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1792 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1793 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1795 // Red compatibility, though the link can't be authenticated on Friendica
1796 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1799 // we may need to restrict this further if it picks up too many strays
1800 // link acct:user@host to a webfinger profile redirector
1802 $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);
1804 // Perform MAIL Search
1805 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1806 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1808 /// @todo What is the meaning of these lines?
1809 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1810 $text = preg_replace('/\&\#039\;/', '\'', $text);
1812 // Currently deactivated, it made problems with " inside of alt texts.
1813 //$text = preg_replace('/\"\;/', '"', $text);
1815 // fix any escaped ampersands that may have been converted into links
1816 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1818 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1819 $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:'];
1821 array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
1823 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1824 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $text);
1826 // sanitize href attributes (only allowlisted protocols URLs)
1827 // default value for backward compatibility
1828 $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
1830 // Always allowed protocol even if config isn't set or not including it
1831 $allowed_link_protocols[] = '//';
1832 $allowed_link_protocols[] = 'http://';
1833 $allowed_link_protocols[] = 'https://';
1834 $allowed_link_protocols[] = 'redir/';
1836 array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
1838 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1839 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
1842 $text = self::convertShare(
1844 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1845 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1849 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1852 }); // Escaped noparse, nobb, pre
1854 // Remove escaping tags and replace new lines that remain
1855 $text = preg_replace_callback('/\[(noparse|nobb)](.*?)\[\/\1]/ism', function ($match) {
1856 return str_replace("\n", "<br>", $match[2]);
1859 // Additionally, [pre] tags preserve spaces
1860 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", function ($match) {
1861 return str_replace([' ', "\n"], [' ', "<br>"], htmlentities($match[1], ENT_NOQUOTES,'UTF-8'));
1867 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1868 function ($matches) {
1869 if (strpos($matches[2], "\n") !== false) {
1870 $return = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlentities(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
1872 $return = '<code>' . htmlentities($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
1880 \HTMLPurifier_URISchemeRegistry::instance()->register('cid', new HTMLPurifier_URIScheme_cid());
1882 $config = \HTMLPurifier_HTML5Config::createDefault();
1883 $config->set('HTML.Doctype', 'HTML5');
1884 $config->set('HTML.SafeIframe', true);
1885 $config->set('URI.SafeIframeRegexp', '%^(?:
1886 https://www.youtube.com/embed/
1888 https://player.vimeo.com/video/
1890 ' . DI::baseUrl() . '/oembed/ # Has to change with the source in Content\Oembed::iframe
1892 $config->set('Attr.AllowedRel', [
1893 'noreferrer' => true,
1896 $config->set('Attr.AllowedFrameTargets', [
1900 $HTMLPurifier = new \HTMLPurifier($config);
1901 $text = $HTMLPurifier->purify($text);
1907 * Strips the "abstract" tag from the provided text
1909 * @param string $text The text with BBCode
1910 * @return string The same text - but without "abstract" element
1912 public static function stripAbstract($text)
1914 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1915 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1921 * Returns the value of the "abstract" element
1923 * @param string $text The text that maybe contains the element
1924 * @param string $addon The addon for which the abstract is meant for
1925 * @return string The abstract
1927 public static function getAbstract($text, $addon = '')
1931 $addon = strtolower($addon);
1933 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1934 foreach ($results AS $result) {
1935 $abstracts[strtolower($result[1])] = $result[2];
1939 if (isset($abstracts[$addon])) {
1940 $abstract = $abstracts[$addon];
1943 if ($abstract == '' && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1944 $abstract = $result[1];
1951 * Callback function to replace a Friendica style mention in a mention for Diaspora
1953 * @param array $match Matching values for the callback
1954 * [1] = Mention type (! or @)
1957 * @return string Replaced mention
1958 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1959 * @throws \ImagickException
1961 private static function bbCodeMention2DiasporaCallback($match)
1963 $contact = Contact::getByURL($match[3], false, ['addr']);
1964 if (empty($contact['addr'])) {
1968 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
1973 * Converts a BBCode text into Markdown
1975 * This function converts a BBCode item body to be sent to Markdown-enabled
1976 * systems like Diaspora and Libertree
1978 * @param string $text
1979 * @param bool $for_diaspora Diaspora requires more changes than Libertree
1981 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1983 public static function toMarkdown($text, $for_diaspora = true)
1985 $original_text = $text;
1987 // Since Diaspora is creating a summary for links, this function removes them before posting
1988 if ($for_diaspora) {
1989 $text = self::removeShareInformation($text);
1993 * Transform #tags, strip off the [url] and replace spaces with underscore
1995 $url_search_string = "^\[\]";
1996 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
1997 function ($matches) {
1998 return '#' . str_replace(' ', '_', $matches[2]);
2003 // Converting images with size parameters to simple images. Markdown doesn't know it.
2004 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2006 // Convert it to HTML - don't try oembed
2007 if ($for_diaspora) {
2008 $text = self::convert($text, false, self::DIASPORA);
2010 // Add all tags that maybe were removed
2011 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2013 foreach ($tags[2] as $tag) {
2014 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2015 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2016 $tagline .= '#' . $tag . ' ';
2019 $text = $text . " " . $tagline;
2022 $text = self::convert($text, false, self::CONNECTORS);
2025 // If a link is followed by a quote then there should be a newline before it
2026 // Maybe we should make this newline at every time before a quote.
2027 $text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
2029 $stamp1 = microtime(true);
2031 // Now convert HTML to Markdown
2032 $text = HTML::toMarkdown($text);
2034 DI::profiler()->saveTimestamp($stamp1, "parser");
2036 // Libertree has a problem with escaped hashtags.
2037 $text = str_replace(['\#'], ['#'], $text);
2039 // Remove any leading or trailing whitespace, as this will mess up
2040 // the Diaspora signature verification and cause the item to disappear
2041 $text = trim($text);
2043 if ($for_diaspora) {
2044 $url_search_string = "^\[\]";
2045 $text = preg_replace_callback(
2046 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
2047 ['self', 'bbCodeMention2DiasporaCallback'],
2052 Hook::callAll('bb2diaspora', $text);
2058 * Pull out all #hashtags and @person tags from $string.
2060 * We also get @person@domain.com - which would make
2061 * the regex quite complicated as tags can also
2062 * end a sentence. So we'll run through our results
2063 * and strip the period from any tags which end with one.
2064 * Returns array of tags found, or empty array.
2066 * @param string $string Post content
2068 * @return array List of tag and person names
2070 public static function getTags($string)
2074 BBCode::performWithEscapedTags($string, ['noparse', 'pre', 'code', 'img'], function ($string) use (&$ret) {
2075 // Convert hashtag links to hashtags
2076 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2 ', $string);
2078 // Force line feeds at bbtags
2079 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
2081 // ignore anything in a bbtag
2082 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
2084 // Match full names against @tags including the space between first and last
2085 // We will look these up afterward to see if they are full names or not recognisable.
2087 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
2088 foreach ($matches[1] as $match) {
2089 if (strstr($match, ']')) {
2090 // we might be inside a bbcode color tag - leave it alone
2094 if (substr($match, -1, 1) === '.') {
2095 $ret[] = substr($match, 0, -1);
2102 // Otherwise pull out single word tags. These can be @nickname, @first_last
2105 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?\']*[^\^ \x0D\x0A,;:?!\'.])/', $string, $matches)) {
2106 foreach ($matches[1] as $match) {
2107 if (strstr($match, ']')) {
2108 // we might be inside a bbcode color tag - leave it alone
2112 // try not to catch url fragments
2113 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
2122 return array_unique($ret);
2126 * Perform a custom function on a text after having escaped blocks enclosed in the provided tag list.
2128 * @param string $text
2129 * @param array $tagList A list of tag names, e.g ['noparse', 'nobb', 'pre']
2130 * @param callable $callback
2133 *@see Strings::performWithEscapedBlocks
2136 public static function performWithEscapedTags(string $text, array $tagList, callable $callback)
2138 $tagList = array_map('preg_quote', $tagList);
2140 return Strings::performWithEscapedBlocks($text, '#\[(?:' . implode('|', $tagList) . ').*?\[/(?:' . implode('|', $tagList) . ')]#ism', $callback);
2144 * Replaces mentions in the provided message body for the provided user and network if any
2147 * @param $profile_uid
2150 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2151 * @throws \ImagickException
2153 public static function setMentions($body, $profile_uid = 0, $network = '')
2155 BBCode::performWithEscapedTags($body, ['noparse', 'pre', 'code', 'img'], function ($body) use ($profile_uid, $network) {
2156 $tags = BBCode::getTags($body);
2161 foreach ($tags as $tag) {
2162 $tag_type = substr($tag, 0, 1);
2164 if ($tag_type == Tag::TAG_CHARACTER[Tag::HASHTAG]) {
2169 * If we already tagged 'Robert Johnson', don't try and tag 'Robert'.
2170 * Robert Johnson should be first in the $tags array
2172 foreach ($tagged as $nextTag) {
2173 if (stristr($nextTag, $tag . ' ')) {
2178 $success = Item::replaceTag($body, $inform, $profile_uid, $tag, $network);
2180 if ($success['replaced']) {
2192 * @param string $author Author display name
2193 * @param string $profile Author profile URL
2194 * @param string $avatar Author profile picture URL
2195 * @param string $link Post source URL
2196 * @param string $posted Post created date
2197 * @param string|null $guid Post guid (if any)
2199 * @TODO Rewrite to handle over whole record array
2201 public static function getShareOpeningTag(string $author, string $profile, string $avatar, string $link, string $posted, string $guid = null)
2203 $header = "[share author='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $author) .
2204 "' profile='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $profile) .
2205 "' avatar='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $avatar) .
2206 "' link='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $link) .
2207 "' posted='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $posted);
2210 $header .= "' guid='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $guid);
2219 * Returns the BBCode relevant to embed the provided URL in a post body.
2220 * For media type, it will return [img], [video] and [audio] tags.
2221 * For regular web pages, it will either output a [bookmark] tag if title and description were provided,
2222 * an [attachment] tag or a simple [url] tag depending on $tryAttachment.
2224 * @param string $url
2225 * @param bool $tryAttachment
2226 * @param string|null $title
2227 * @param string|null $description
2228 * @param string|null $tags
2230 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2231 *@see ParseUrl::getSiteinfoCached
2234 public static function embedURL(string $url, bool $tryAttachment = true, string $title = null, string $description = null, string $tags = null): string
2236 DI::logger()->info($url);
2238 // If there is already some content information submitted we don't
2239 // need to parse the url for content.
2240 if (!empty($title) && !empty($description)) {
2241 $title = str_replace(["\r", "\n"], ['', ''], $title);
2243 $description = '[quote]' . trim($description) . '[/quote]' . "\n";
2246 if (!empty($tags)) {
2247 $arr_tags = ParseUrl::convertTagsToArray($tags);
2248 if (count($arr_tags)) {
2249 $str_tags = "\n" . implode(' ', $arr_tags) . "\n";
2253 $result = sprintf('[bookmark=%s]%s[/bookmark]%s', $url, ($title) ? $title : $url, $description) . $str_tags;
2255 DI::logger()->info('(unparsed): returns: ' . $result);
2260 $siteinfo = ParseUrl::getSiteinfoCached($url);
2262 if (in_array($siteinfo['type'], ['image', 'video', 'audio'])) {
2263 switch ($siteinfo['type']) {
2265 $bbcode = "\n" . '[video]' . $url . '[/video]' . "\n";
2268 $bbcode = "\n" . '[audio]' . $url . '[/audio]' . "\n";
2271 $bbcode = "\n" . '[img]' . $url . '[/img]' . "\n";
2278 unset($siteinfo['keywords']);
2280 // Bypass attachment if parse url for a comment
2281 if (!$tryAttachment) {
2282 return "\n" . '[url=' . $url . ']' . $siteinfo['title'] . '[/url]';
2285 // Format it as BBCode attachment
2286 return "\n" . PageInfo::getFooterFromData($siteinfo);