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\Smilies;
31 use Friendica\Core\Hook;
32 use Friendica\Core\Logger;
33 use Friendica\Core\Protocol;
34 use Friendica\Core\Renderer;
35 use Friendica\Core\System;
37 use Friendica\Model\Contact;
38 use Friendica\Model\Event;
39 use Friendica\Model\Photo;
40 use Friendica\Model\Tag;
41 use Friendica\Network\Probe;
42 use Friendica\Object\Image;
43 use Friendica\Protocol\Activity;
44 use Friendica\Util\Images;
45 use Friendica\Util\Map;
46 use Friendica\Util\Network;
47 use Friendica\Util\ParseUrl;
48 use Friendica\Util\Proxy as ProxyUtils;
49 use Friendica\Util\Strings;
50 use Friendica\Util\XML;
61 const ACTIVITYPUB = 9;
64 * Fetches attachment data that were generated the old way
66 * @param string $body Message body
68 * 'type' -> Message type ('link', 'video', 'photo')
69 * 'text' -> Text before the shared message
70 * 'after' -> Text after the shared message
71 * 'image' -> Preview image of the message
72 * 'url' -> Url to the attached message
73 * 'title' -> Title of the attachment
74 * 'description' -> Description of the attachment
75 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
77 private static function getOldAttachmentData($body)
81 // Simplify image codes
82 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
84 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
85 foreach ($attached as $data) {
86 if (!in_array($data[1], ['type-link', 'type-video', 'type-photo'])) {
90 $post['type'] = substr($data[1], 5);
92 $pos = strpos($body, $data[0]);
94 $post['text'] = trim(substr($body, 0, $pos));
95 $post['after'] = trim(substr($body, $pos + strlen($data[0])));
97 $post['text'] = trim(str_replace($data[0], '', $body));
101 $attacheddata = $data[2];
103 if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
105 $picturedata = Images::getInfoFromURLCached($matches[1]);
108 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
109 $post['image'] = $matches[1];
111 $post['preview'] = $matches[1];
116 if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
117 $post['url'] = $matches[1];
118 $post['title'] = $matches[2];
120 if (!empty($post['url']) && (in_array($post['type'], ['link', 'video']))
121 && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
122 $post['url'] = $matches[1];
125 // Search for description
126 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
127 $post['description'] = $matches[1];
135 * Fetches attachment data that were generated with the "attachment" element
137 * @param string $body Message body
139 * 'type' -> Message type ('link', 'video', 'photo')
140 * 'text' -> Text before the shared message
141 * 'after' -> Text after the shared message
142 * 'image' -> Preview image of the message
143 * 'url' -> Url to the attached message
144 * 'title' -> Title of the attachment
145 * 'description' -> Description of the attachment
146 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
148 public static function getAttachmentData($body)
160 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
161 return self::getOldAttachmentData($body);
164 $attributes = $match[2];
166 $data['text'] = trim($match[1]);
169 preg_match("/type='(.*?)'/ism", $attributes, $matches);
170 if (!empty($matches[1])) {
171 $type = strtolower($matches[1]);
174 preg_match('/type="(.*?)"/ism', $attributes, $matches);
175 if (!empty($matches[1])) {
176 $type = strtolower($matches[1]);
183 if (!in_array($type, ['link', 'audio', 'photo', 'video'])) {
188 $data['type'] = $type;
192 preg_match("/url='(.*?)'/ism", $attributes, $matches);
193 if (!empty($matches[1])) {
197 preg_match('/url="(.*?)"/ism', $attributes, $matches);
198 if (!empty($matches[1])) {
203 $data['url'] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
207 preg_match("/title='(.*?)'/ism", $attributes, $matches);
208 if (!empty($matches[1])) {
209 $title = $matches[1];
212 preg_match('/title="(.*?)"/ism', $attributes, $matches);
213 if (!empty($matches[1])) {
214 $title = $matches[1];
218 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
219 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
220 $title = str_replace(['[', ']'], ['[', ']'], $title);
221 $data['title'] = $title;
225 preg_match("/image='(.*?)'/ism", $attributes, $matches);
226 if (!empty($matches[1])) {
227 $image = $matches[1];
230 preg_match('/image="(.*?)"/ism', $attributes, $matches);
231 if (!empty($matches[1])) {
232 $image = $matches[1];
236 $data['image'] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
240 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
241 if (!empty($matches[1])) {
242 $preview = $matches[1];
245 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
246 if (!empty($matches[1])) {
247 $preview = $matches[1];
250 if ($preview != '') {
251 $data['preview'] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
254 $data['description'] = trim($match[3]);
256 $data['after'] = trim($match[4]);
261 public static function getAttachedData($body, $item = [])
265 - type: link, video, photo
273 $has_title = !empty($item['title']);
274 $plink = $item['plink'] ?? '';
275 $post = self::getAttachmentData($body);
277 // Get all linked images with alternative image description
278 if (preg_match_all("/\[img=([^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
279 foreach ($pictures as $picture) {
280 if (Photo::isLocal($picture[1])) {
281 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2]];
284 if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
285 $post['image_description'] = $post['images'][0]['description'];
289 if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
290 foreach ($pictures as $picture) {
291 if (Photo::isLocal($picture[1])) {
292 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => ''];
297 // if nothing is found, it maybe having an image.
298 if (!isset($post['type'])) {
299 // Simplify image codes
300 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
301 $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
302 $post['text'] = $body;
304 if (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $body, $pictures, PREG_SET_ORDER)) {
305 if ((count($pictures) == 1) && !$has_title) {
306 if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
307 // Replace the preview picture with the real picture
308 $url = str_replace('-1.', '-0.', $pictures[0][2]);
309 $data = ['url' => $url, 'type' => 'photo'];
311 // Checking, if the link goes to a picture
312 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
316 // Sometimes photo posts to the own album are not detected at the start.
317 // So we seem to cannot use the cache for these cases. That's strange.
318 if (($data['type'] != 'photo') && strstr($pictures[0][1], "/photos/")) {
319 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
322 if ($data['type'] == 'photo') {
323 $post['type'] = 'photo';
324 if (isset($data['images'][0])) {
325 $post['image'] = $data['images'][0]['src'];
326 $post['url'] = $data['url'];
328 $post['image'] = $data['url'];
331 $post['preview'] = $pictures[0][2];
332 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
334 $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
335 if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
336 $post['type'] = 'photo';
337 $post['image'] = $pictures[0][1];
338 $post['preview'] = $pictures[0][2];
339 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
342 } elseif (count($pictures) > 0) {
343 $post['type'] = 'link';
344 $post['url'] = $plink;
345 $post['image'] = $pictures[0][2];
346 $post['text'] = $body;
348 foreach ($pictures as $picture) {
349 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
352 } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
353 if ((count($pictures) == 1) && !$has_title) {
354 $post['type'] = 'photo';
355 $post['image'] = $pictures[0][1];
356 $post['text'] = str_replace($pictures[0][0], '', $body);
357 } elseif (count($pictures) > 0) {
358 $post['type'] = 'link';
359 $post['url'] = $plink;
360 $post['image'] = $pictures[0][1];
361 $post['text'] = $body;
363 foreach ($pictures as $picture) {
364 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
369 // Test for the external links
370 preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER);
371 preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $links2, PREG_SET_ORDER);
373 $links = array_merge($links1, $links2);
375 // If there is only a single one, then use it.
376 // This should cover link posts via API.
377 if ((count($links) == 1) && !isset($post['preview']) && !$has_title) {
378 $post['type'] = 'link';
379 $post['url'] = $links[0][1];
382 // Simplify "video" element
383 $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
385 // Now count the number of external media links
386 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
387 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
388 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER);
389 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $links4, PREG_SET_ORDER);
391 // Add them to the other external links
392 $links = array_merge($links, $links1, $links2, $links3, $links4);
394 // Are there more than one?
395 if (count($links) > 1) {
396 // The post will be the type "text", which means a blog post
397 unset($post['type']);
398 $post['url'] = $plink;
401 if (!isset($post['type'])) {
402 $post['type'] = "text";
403 $post['text'] = trim($body);
405 } elseif (isset($post['url']) && ($post['type'] == 'video')) {
406 $data = ParseUrl::getSiteinfoCached($post['url'], true);
408 if (isset($data['images'][0])) {
409 $post['image'] = $data['images'][0]['src'];
417 * Remove [attachment] BBCode and replaces it with a regular [url]
419 * @param string $body
420 * @param boolean $no_link_desc No link description
422 * @return string with replaced body
424 public static function removeAttachment($body, $no_link_desc = false)
426 return preg_replace_callback("/\s*\[attachment (.*)\](.*?)\[\/attachment\]\s*/ism",
427 function ($match) use ($no_link_desc) {
428 $attach_data = self::getAttachmentData($match[0]);
429 if (empty($attach_data['url'])) {
431 } elseif (empty($attach_data['title']) || $no_link_desc) {
432 return "\n[url]" . $attach_data['url'] . "[/url]\n";
434 return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
440 * Converts a BBCode text into plaintext
443 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
447 public static function toPlaintext($text, $keep_urls = true)
449 $naked_text = HTML::toPlaintext(self::convert($text, false, 0, true), 0, !$keep_urls);
454 private static function proxyUrl($image, $simplehtml = self::INTERNAL)
456 // Only send proxied pictures to API and for internal display
457 if (in_array($simplehtml, [self::INTERNAL, self::API])) {
458 return ProxyUtils::proxifyUrl($image);
465 * This function changing the visual size (not the real size) of images.
466 * The function does not work for pictures with an alternate text description.
467 * This could only be changed by using some new "img" BBCode format.
469 * @param string $srctext The body with images
470 * @return string The body with possibly scaled images
472 public static function scaleExternalImages(string $srctext)
476 // Simplify image links
477 $s = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $s);
480 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
482 foreach ($matches as $mtch) {
483 Logger::log('scale_external_image: ' . $mtch[1]);
485 $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
486 if (stristr($mtch[1], $hostname)) {
490 $curlResult = Network::curl($mtch[1], true);
491 if (!$curlResult->isSuccess()) {
495 $i = $curlResult->getBody();
496 $type = $curlResult->getContentType();
497 $type = Images::getMimeTypeByData($i, $mtch[1], $type);
500 $Image = new Image($i, $type);
501 if ($Image->isValid()) {
502 $orig_width = $Image->getWidth();
503 $orig_height = $Image->getHeight();
505 if ($orig_width > 640 || $orig_height > 640) {
506 $Image->scaleDown(640);
507 $new_width = $Image->getWidth();
508 $new_height = $Image->getHeight();
509 Logger::info('External images scaled', ['orig_width' => $orig_width, 'new_width' => $new_width, 'orig_height' => $orig_height, 'new_height' => $new_height, 'match' => $mtch[0]]);
512 '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
516 Logger::info('New string', ['image' => $s]);
527 * Truncates imported message body string length to max_import_size
529 * The purpose of this function is to apply system message length limits to
530 * imported messages without including any embedded photos in the length
532 * @param string $body
534 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
536 public static function limitBodySize($body)
538 $maxlen = DI::config()->get('config', 'max_import_size', 0);
540 // If the length of the body, including the embedded images, is smaller
541 // than the maximum, then don't waste time looking for the images
542 if ($maxlen && (strlen($body) > $maxlen)) {
544 Logger::info('the total body length exceeds the limit', ['maxlen' => $maxlen, 'body_len' => strlen($body)]);
550 $img_start = strpos($orig_body, '[img');
551 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
552 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
553 while (($img_st_close !== false) && ($img_end !== false)) {
555 $img_st_close++; // make it point to AFTER the closing bracket
556 $img_end += $img_start;
557 $img_end += strlen('[/img]');
559 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
560 // This is an embedded image
562 if (($textlen + $img_start) > $maxlen) {
563 if ($textlen < $maxlen) {
564 Logger::info('the limit happens before an embedded image');
565 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
569 $new_body = $new_body . substr($orig_body, 0, $img_start);
570 $textlen += $img_start;
573 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
576 if (($textlen + $img_end) > $maxlen) {
577 if ($textlen < $maxlen) {
578 Logger::info('the limit happens before the end of a non-embedded image');
579 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
583 $new_body = $new_body . substr($orig_body, 0, $img_end);
584 $textlen += $img_end;
587 $orig_body = substr($orig_body, $img_end);
589 if ($orig_body === false) {
590 // in case the body ends on a closing image tag
594 $img_start = strpos($orig_body, '[img');
595 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
596 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
599 if (($textlen + strlen($orig_body)) > $maxlen) {
600 if ($textlen < $maxlen) {
601 Logger::info('the limit happens after the end of the last image');
602 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
605 Logger::info('the text size with embedded images extracted did not violate the limit');
606 $new_body = $new_body . $orig_body;
616 * Processes [attachment] tags
618 * Note: Can produce a [bookmark] tag in the returned string
620 * @param string $text
621 * @param integer $simplehtml
622 * @param bool $tryoembed
624 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
626 private static function convertAttachment($text, $simplehtml = self::INTERNAL, $tryoembed = true)
628 $data = self::getAttachmentData($text);
629 if (empty($data) || empty($data['url'])) {
633 if (isset($data['title'])) {
634 $data['title'] = strip_tags($data['title']);
635 $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
637 $data['title'] = null;
640 if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
641 $data['preview'] = $data['image'];
647 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
648 $return = OEmbed::getHTML($data['url'], $data['title']);
650 throw new Exception('OEmbed is disabled for this attachment.');
652 } catch (Exception $e) {
653 $data['title'] = ($data['title'] ?? '') ?: $data['url'];
655 if ($simplehtml != self::CONNECTORS) {
656 $return = sprintf('<div class="type-%s">', $data['type']);
659 if (!empty($data['title']) && !empty($data['url'])) {
660 if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
661 $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), $data['title']);
663 if (!empty($data['image'])) {
664 $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), $data['title']);
665 } elseif (!empty($data['preview'])) {
666 $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']);
668 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
672 if (!empty($data['description']) && $data['description'] != $data['title']) {
673 // Sanitize the HTML by converting it to BBCode
674 $bbcode = HTML::toBBCode($data['description']);
675 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
678 if (!empty($data['url'])) {
679 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
682 if ($simplehtml != self::CONNECTORS) {
687 return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
690 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
692 $data = self::getAttachmentData($Text);
697 return $data['text'] . ($data['after'] ?? '');
700 $title = htmlentities($data['title'] ?? '', ENT_QUOTES, 'UTF-8', false);
701 $text = htmlentities($data['text'], ENT_QUOTES, 'UTF-8', false);
702 if ($plaintext || (($title != '') && strstr($text, $title))) {
703 $data['title'] = $data['url'];
704 } elseif (($text != '') && strstr($title, $text)) {
705 $data['text'] = $data['title'];
706 $data['title'] = $data['url'];
709 if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
710 return $data['title'] . $data['after'];
713 // If the link already is included in the post, don't add it again
714 if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
715 return $data['text'] . $data['after'];
718 $text = $data['text'];
720 if (!empty($data['url']) && !empty($data['title'])) {
721 $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
722 } elseif (!empty($data['url'])) {
723 $text .= "\n[url]" . $data['url'] . '[/url]';
726 return $text . "\n" . $data['after'];
730 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
732 * @param array $match Array with the matching values
733 * @return string reformatted link including HTML codes
735 private static function convertUrlForActivityPubCallback($match)
739 if (isset($match[2]) && ($match[1] != $match[2])) {
743 $parts = parse_url($url);
744 if (!isset($parts['scheme'])) {
748 return self::convertUrlForActivityPub($url);
752 * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
754 * @param string $url URL that is about to be reformatted
755 * @return string reformatted link including HTML codes
757 private static function convertUrlForActivityPub($url)
759 $html = '<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>';
760 return sprintf($html, $url, self::getStyledURL($url));
764 * Converts an URL in a nicer format (without the scheme and possibly shortened)
766 * @param string $url URL that is about to be reformatted
767 * @return string reformatted link
769 private static function getStyledURL($url)
771 $parts = parse_url($url);
772 $scheme = $parts['scheme'] . '://';
773 $styled_url = str_replace($scheme, '', $url);
775 if (strlen($styled_url) > 30) {
776 $styled_url = substr($styled_url, 0, 30) . "…";
783 * [noparse][i]italic[/i][/noparse] turns into
784 * [noparse][ i ]italic[ /i ][/noparse],
785 * to hide them from parser.
787 private static function escapeNoparseCallback($match)
789 $whole_match = $match[0];
790 $captured = $match[1];
791 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
792 $new_str = str_replace($captured, $spacefied, $whole_match);
797 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
798 * now turns back and the [noparse] tags are trimed
799 * returning [i]italic[/i]
801 private static function unescapeNoparseCallback($match)
803 $captured = $match[1];
804 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
809 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
812 * @param string $text Text to search
813 * @param string $name Tag name
814 * @param int $occurrences Number of first occurrences to skip
815 * @return boolean|array
817 public static function getTagPosition($text, $name, $occurrences = 0)
819 if ($occurrences < 0) {
824 for ($i = 0; $i <= $occurrences; $i++) {
825 if ($start_open !== false) {
826 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
830 if ($start_open === false) {
834 $start_equal = strpos($text, '=', $start_open);
835 $start_close = strpos($text, ']', $start_open);
837 if ($start_close === false) {
843 $end_open = strpos($text, '[/' . $name . ']', $start_close);
845 if ($end_open === false) {
851 'open' => $start_open,
852 'close' => $start_close
856 'close' => $end_open + strlen('[/' . $name . ']')
860 if ($start_equal !== false) {
861 $res['start']['equal'] = $start_equal + 1;
868 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
870 * @param string $pattern Preg pattern string
871 * @param string $replace Preg replace string
872 * @param string $name BBCode tag name
873 * @param string $text Text to search
876 public static function pregReplaceInTag($pattern, $replace, $name, $text)
879 $pos = self::getTagPosition($text, $name, $occurrences);
880 while ($pos !== false && $occurrences++ < 1000) {
881 $start = substr($text, 0, $pos['start']['open']);
882 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
883 $end = substr($text, $pos['end']['close']);
884 if ($end === false) {
888 $subject = preg_replace($pattern, $replace, $subject);
889 $text = $start . $subject . $end;
891 $pos = self::getTagPosition($text, $name, $occurrences);
897 private static function extractImagesFromItemBody($body)
904 $img_start = strpos($orig_body, '[img');
905 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
906 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
907 while (($img_st_close !== false) && ($img_end !== false)) {
908 $img_st_close++; // make it point to AFTER the closing bracket
909 $img_end += $img_start;
911 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
912 // This is an embedded image
913 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
914 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
918 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
921 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
923 if ($orig_body === false) {
924 // in case the body ends on a closing image tag
928 $img_start = strpos($orig_body, '[img');
929 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
930 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
933 $new_body = $new_body . $orig_body;
935 return ['body' => $new_body, 'images' => $saved_image];
938 private static function interpolateSavedImagesIntoItemBody($body, array $images)
943 foreach ($images as $image) {
944 // We're depending on the property of 'foreach' (specified on the PHP website) that
945 // it loops over the array starting from the first element and going sequentially
946 // to the last element
947 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
948 '<img src="' . self::proxyUrl($image) . '" alt="' . DI::l10n()->t('Image/photo') . '" />', $newbody);
956 * This function converts a [share] block to text according to a provided callback function whose signature is:
958 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
961 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
963 * - $author_contact is a contact record array
964 * - $content is the inner content of the [share] block
965 * - $is_quote_share indicates whether there's any content before the [share] block
966 * - Return value is the string that should replace the [share] block in the provided text
968 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
970 * @param string $text A BBCode string
971 * @param callable $callback
972 * @return string The BBCode string with all [share] blocks replaced
974 public static function convertShare($text, callable $callback)
976 $return = preg_replace_callback(
977 "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
978 function ($match) use ($callback) {
979 $attribute_string = $match[2];
981 foreach (['author', 'profile', 'avatar', 'link', 'posted', 'guid'] as $field) {
982 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
983 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
986 $author_contact = Contact::getByURL($attributes['profile'], 0, ['url', 'addr', 'name', 'micro'], false);
987 $author_contact['url'] = ($author_contact['url'] ?? $attributes['profile']);
988 $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']);
990 $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author'];
991 $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
992 $attributes['profile'] = ($author_contact['url'] ?? '') ?: $attributes['profile'];
994 if ($attributes['avatar']) {
995 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
998 return $match[1] . $callback($attributes, $author_contact, $match[3], trim($match[1]) != '');
1007 * Default [share] tag conversion callback
1009 * Note: Can produce a [bookmark] tag in the output
1011 * @see BBCode::convertShare()
1012 * @param array $attributes [share] block attribute values
1013 * @param array $author_contact Contact row of the shared author
1014 * @param string $content Inner content of the [share] block
1015 * @param boolean $is_quote_share Whether there is content before the [share] block
1016 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
1018 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1020 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
1022 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
1024 switch ($simplehtml) {
1026 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
1028 case self::DIASPORA:
1029 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
1030 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
1032 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
1034 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
1035 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
1038 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
1040 if (empty($attributes['posted']) && !empty($attributes['link'])) {
1041 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
1046 case self::CONNECTORS:
1047 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
1048 $headline .= DI::l10n()->t('<a href="%1$s" target="_blank" rel="noopener noreferrer">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
1049 $headline .= ':</b></p>' . "\n";
1051 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
1055 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
1057 case self::ACTIVITYPUB:
1058 $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>';
1059 $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
1062 $text = ($is_quote_share? "\n" : '');
1064 $contact = Contact::getByURL($attributes['profile'], 0, ['network'], false);
1065 $network = $contact['network'] ?? Protocol::PHANTOM;
1067 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1068 $text .= Renderer::replaceMacros($tpl, [
1069 '$profile' => $attributes['profile'],
1070 '$avatar' => $attributes['avatar'],
1071 '$author' => $attributes['author'],
1072 '$link' => $attributes['link'],
1073 '$link_title' => DI::l10n()->t('link to source'),
1074 '$posted' => $attributes['posted'],
1075 '$guid' => $attributes['guid'],
1076 '$network_name' => ContactSelector::networkToName($network, $attributes['profile']),
1077 '$network_icon' => ContactSelector::networkToIcon($network, $attributes['profile']),
1078 '$content' => self::setMentions(trim($content), 0, $network),
1086 private static function removePictureLinksCallback($match)
1088 $cache_key = 'remove:' . $match[1];
1089 $text = DI::cache()->get($cache_key);
1091 if (is_null($text)) {
1094 $stamp1 = microtime(true);
1096 $ch = @curl_init($match[1]);
1097 @curl_setopt($ch, CURLOPT_NOBODY, true);
1098 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1099 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1101 $curl_info = @curl_getinfo($ch);
1103 DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
1105 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1106 $text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
1108 $text = "[url=" . $match[2] . ']' . $match[2] . "[/url]";
1110 // if its not a picture then look if its a page that contains a picture link
1111 $body = Network::fetchUrl($match[1]);
1113 $doc = new DOMDocument();
1114 @$doc->loadHTML($body);
1115 $xpath = new DOMXPath($doc);
1116 $list = $xpath->query("//meta[@name]");
1117 foreach ($list as $node) {
1120 if ($node->attributes->length) {
1121 foreach ($node->attributes as $attribute) {
1122 $attr[$attribute->name] = $attribute->value;
1126 if (strtolower($attr['name']) == 'twitter:image') {
1127 $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
1131 DI::cache()->set($cache_key, $text);
1137 private static function expandLinksCallback($match)
1139 if (($match[3] == '') || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1140 return ($match[1] . "[url]" . $match[2] . "[/url]");
1142 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1146 private static function cleanPictureLinksCallback($match)
1150 // When the picture link is the own photo path then we can avoid fetching the link
1151 $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
1152 if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
1153 if (!empty($match[3])) {
1154 $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
1156 $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
1161 $cache_key = 'clean:' . $match[1];
1162 $text = DI::cache()->get($cache_key);
1163 if (!is_null($text)) {
1167 // Only fetch the header, not the content
1168 $stamp1 = microtime(true);
1170 $ch = @curl_init($match[1]);
1171 @curl_setopt($ch, CURLOPT_NOBODY, true);
1172 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1173 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1175 $curl_info = @curl_getinfo($ch);
1177 DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
1179 // if its a link to a picture then embed this picture
1180 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1181 $text = '[img]' . $match[1] . '[/img]';
1183 if (!empty($match[3])) {
1184 $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
1186 $text = '[img]' . $match[2] . '[/img]';
1189 // if its not a picture then look if its a page that contains a picture link
1190 $body = Network::fetchUrl($match[1]);
1192 $doc = new DOMDocument();
1193 @$doc->loadHTML($body);
1194 $xpath = new DOMXPath($doc);
1195 $list = $xpath->query("//meta[@name]");
1196 foreach ($list as $node) {
1198 if ($node->attributes->length) {
1199 foreach ($node->attributes as $attribute) {
1200 $attr[$attribute->name] = $attribute->value;
1204 if (strtolower($attr['name']) == "twitter:image") {
1205 if (!empty($match[3])) {
1206 $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
1208 $text = "[img]" . $attr['content'] . "[/img]";
1213 DI::cache()->set($cache_key, $text);
1218 public static function cleanPictureLinks($text)
1220 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1221 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $return);
1226 * Converts a BBCode message to HTML message
1228 * BBcode 2 HTML was written by WAY2WEB.net
1229 * extended to work with Mistpark/Friendica - Mike Macgirvin
1231 * Simple HTML values meaning:
1232 * - 0: Friendica display
1234 * - 2: Used for Windows Phone push, Friendica API
1235 * - 3: Used before converting to Markdown in bb2diaspora.php
1236 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1239 * - 7: Used for dfrn, OStatus
1240 * - 8: Used for Twitter, WP backlink text setting
1243 * @param string $text
1244 * @param bool $try_oembed
1245 * @param int $simple_html
1246 * @param bool $for_plaintext
1248 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1250 public static function convert(string $text = null, $try_oembed = true, $simple_html = self::INTERNAL, $for_plaintext = false)
1252 // Accounting for null default column values
1253 if (is_null($text) || $text === '') {
1259 $text = self::performWithEscapedTags($text, ['code'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
1260 $text = self::performWithEscapedTags($text, ['noparse', 'nobb', 'pre'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
1262 * preg_match_callback function to replace potential Oembed tags with Oembed content
1264 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1266 * $match[2] = $title or absent
1268 $try_oembed_callback = function ($match)
1271 $title = $match[2] ?? null;
1274 $return = OEmbed::getHTML($url, $title);
1275 } catch (Exception $ex) {
1276 $return = $match[0];
1284 // Remove the abstract element. It is a non visible element.
1285 $text = self::stripAbstract($text);
1287 // Move new lines outside of tags
1288 $text = preg_replace("#\[(\w*)](\n*)#ism", '$2[$1]', $text);
1289 $text = preg_replace("#(\n*)\[/(\w*)]#ism", '[/$2]$1', $text);
1291 // Extract the private images which use data urls since preg has issues with
1292 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1293 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1295 $extracted = self::extractImagesFromItemBody($text);
1296 $text = $extracted['body'];
1297 $saved_image = $extracted['images'];
1299 // If we find any event code, turn it into an event.
1300 // After we're finished processing the bbcode we'll
1301 // replace all of the event code with a reformatted version.
1303 $ev = Event::fromBBCode($text);
1305 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1306 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1308 $text = str_replace("<", "<", $text);
1309 $text = str_replace(">", ">", $text);
1311 // remove some newlines before the general conversion
1312 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1313 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1315 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1317 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1320 // Convert new line chars to html <br /> tags
1322 // nlbr seems to be hopelessly messed up
1323 // $Text = nl2br($Text);
1325 // We'll emulate it.
1327 $text = trim($text);
1328 $text = str_replace("\r\n", "\n", $text);
1330 // Remove linefeeds inside of the table elements. See issue #6799
1331 $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
1332 "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
1333 "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
1334 "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "];
1335 $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
1336 "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
1337 "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
1338 "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"];
1341 $text = str_replace($search, $replace, $text);
1342 } while ($oldtext != $text);
1344 // Replace these here only once
1345 $search = ["\n[table]", "[/table]\n"];
1346 $replace = ["[table]", "[/table]"];
1347 $text = str_replace($search, $replace, $text);
1349 // removing multiplicated newlines
1350 if (DI::config()->get('system', 'remove_multiplicated_lines')) {
1351 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1352 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1353 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1354 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1357 $text = str_replace($search, $replace, $text);
1358 } while ($oldtext != $text);
1361 /// @todo Have a closer look at the different html modes
1362 // Handle attached links or videos
1363 if ($simple_html == self::ACTIVITYPUB) {
1364 $text = self::removeAttachment($text);
1365 } elseif (!in_array($simple_html, [self::INTERNAL, self::CONNECTORS])) {
1366 $text = self::removeAttachment($text, true);
1368 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1371 // leave open the posibility of [map=something]
1372 // this is replaced in Item::prepareBody() which has knowledge of the item location
1373 if (strpos($text, '[/map]') !== false) {
1374 $text = preg_replace_callback(
1375 "/\[map\](.*?)\[\/map\]/ism",
1376 function ($match) use ($simple_html) {
1377 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1383 if (strpos($text, '[map=') !== false) {
1384 $text = preg_replace_callback(
1385 "/\[map=(.*?)\]/ism",
1386 function ($match) use ($simple_html) {
1387 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1393 if (strpos($text, '[map]') !== false) {
1394 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1397 // Check for headers
1398 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1399 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1400 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1401 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1402 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1403 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1405 // Check for paragraph
1406 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1408 // Check for bold text
1409 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1411 // Check for Italics text
1412 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1414 // Check for Underline text
1415 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1417 // Check for strike-through text
1418 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1420 // Check for over-line text
1421 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1423 // Check for colored text
1424 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1426 // Check for sized text
1427 // [size=50] --> font-size: 50px (with the unit).
1428 if ($simple_html != self::DIASPORA) {
1429 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1430 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1432 // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
1433 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text);
1437 // Check for centered text
1438 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1440 // Check for list text
1441 $text = str_replace("[*]", "<li>", $text);
1443 // Check for style sheet commands
1444 $text = preg_replace_callback(
1445 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1447 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1452 // Check for CSS classes
1453 $text = preg_replace_callback(
1454 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1456 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1461 // handle nested lists
1464 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1465 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1466 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1467 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1468 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1469 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1470 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1471 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1472 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1473 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1474 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1475 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1476 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1477 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1480 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1481 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1482 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1483 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1485 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1486 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1488 $text = str_replace('[hr]', '<hr />', $text);
1490 if (!$for_plaintext) {
1493 // Escaping BBCodes susceptible to contain rogue URL we don'' want the autolinker to catch
1494 $text = preg_replace_callback('#\[(url|img|audio|video|youtube|vimeo|share|attachment|iframe|bookmark).+?\[/\1\]#ism',
1495 function ($matches) use (&$escaped) {
1496 $return = '{escaped-' . count($escaped) . '}';
1497 $escaped[] = $matches[0];
1504 // Autolinker for isolated URLs
1505 $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1507 // Restoring escaped blocks
1508 $text = preg_replace_callback('/{escaped-([0-9]+)}/iU',
1509 function ($matches) use ($escaped) {
1510 return $escaped[intval($matches[1])] ?? $matches[0];
1516 // This is actually executed in Item::prepareBody()
1518 $nosmile = strpos($text, '[nosmile]') !== false;
1519 $text = str_replace('[nosmile]', '', $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);
1619 $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism", '<video src="$1" controls="controls" width="' . $a->videowidth . '" height="' . $a->videoheight . '" loop="true"><a href="$1">$1</a></video>', $text);
1620 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $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",
1628 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1631 // html5 video and audio
1635 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1637 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1640 // Youtube extensions
1642 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1643 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1644 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1647 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1648 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1649 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1652 $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);
1654 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1655 '<a href="https://www.youtube.com/watch?v=$1" target="_blank" rel="noopener noreferrer">https://www.youtube.com/watch?v=$1</a>', $text);
1659 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1660 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1663 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1664 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1667 $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);
1669 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1670 '<a href="https://vimeo.com/$1" target="_blank" rel="noopener noreferrer">https://vimeo.com/$1</a>', $text);
1674 $text = OEmbed::BBCode2HTML($text);
1676 // Avoid triple linefeeds through oembed
1677 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1679 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1680 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1681 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1682 // start which is always required). Allow desc with a missing summary for compatibility.
1684 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1685 $sub = Event::getHTML($ev, $simple_html);
1687 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1688 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1689 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1690 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1691 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1692 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1693 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1696 // Replace non graphical smilies for external posts
1697 if (!$nosmile && !$for_plaintext) {
1698 $text = Smilies::replace($text);
1701 if (!$for_plaintext && DI::config()->get('system', 'big_emojis') && ($simple_html != self::DIASPORA)) {
1702 $conv = html_entity_decode(str_replace([' ', "\n", "\r"], '', $text));
1703 // Emojis are always 4 byte Unicode characters
1704 if (!empty($conv) && (strlen($conv) / mb_strlen($conv) == 4)) {
1705 $text = '<span style="font-size: xx-large; line-height: initial;">' . $text . '</span>';
1709 if (!$for_plaintext) {
1710 if (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1711 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1712 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1715 $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1716 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1719 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1721 // Remove all hashtag addresses
1722 if ($simple_html && !in_array($simple_html, [self::DIASPORA, self::OSTATUS, self::ACTIVITYPUB])) {
1723 $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1724 } elseif ($simple_html == self::DIASPORA) {
1725 // The ! is converted to @ since Diaspora only understands the @
1726 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1727 '@<a href="$2">$3</a>',
1729 } elseif (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1730 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1731 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1733 } elseif (!$simple_html) {
1734 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1735 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1739 // Bookmarks in red - will be converted to bookmarks in friendica
1740 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1741 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1742 $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1743 "[bookmark=$1]$2[/bookmark]", $text);
1745 if (in_array($simple_html, [self::API, self::OSTATUS, self::TWITTER])) {
1746 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1747 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1748 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1751 // Perform URL Search
1753 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1756 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1758 // Handle Diaspora posts
1759 $text = preg_replace_callback(
1760 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1762 return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1766 $text = preg_replace_callback(
1767 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1769 return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1773 // Server independent link to posts and comments
1774 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1775 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1776 $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
1780 * - #[url=<anything>]<term>[/url]
1781 * - [url=<anything>]#<term>[/url]
1783 $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) use ($simple_html) {
1784 if ($simple_html == BBCode::ACTIVITYPUB) {
1785 return '<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1786 . '" data-tag="' . XML::escape($matches[1]) . '" rel="tag ugc">#'
1787 . XML::escape($matches[1]) . '</a>';
1789 return '#<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1790 . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
1791 . XML::escape($matches[1]) . '</a>';
1795 // We need no target="_blank" rel="noopener noreferrer" for local links
1796 // convert links start with DI::baseUrl() as local link without the target="_blank" rel="noopener noreferrer" attribute
1797 $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
1798 $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1799 $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1801 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1802 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1804 // Red compatibility, though the link can't be authenticated on Friendica
1805 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1808 // we may need to restrict this further if it picks up too many strays
1809 // link acct:user@host to a webfinger profile redirector
1811 $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);
1813 // Perform MAIL Search
1814 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1815 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1817 /// @todo What is the meaning of these lines?
1818 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1819 $text = preg_replace('/\&\#039\;/', '\'', $text);
1821 // Currently deactivated, it made problems with " inside of alt texts.
1822 //$text = preg_replace('/\"\;/', '"', $text);
1824 // fix any escaped ampersands that may have been converted into links
1825 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1827 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1828 $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:'];
1830 array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
1832 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1833 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $text);
1835 // sanitize href attributes (only allowlisted protocols URLs)
1836 // default value for backward compatibility
1837 $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
1839 // Always allowed protocol even if config isn't set or not including it
1840 $allowed_link_protocols[] = '//';
1841 $allowed_link_protocols[] = 'http://';
1842 $allowed_link_protocols[] = 'https://';
1843 $allowed_link_protocols[] = 'redir/';
1845 array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
1847 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1848 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
1851 $text = self::convertShare(
1853 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1854 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1858 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1861 }); // Escaped noparse, nobb, pre
1863 // Remove escaping tags
1864 $text = preg_replace("/\[noparse\](.*?)\[\/noparse\]/ism", '\1', $text);
1865 $text = preg_replace("/\[nobb\](.*?)\[\/nobb\]/ism", '\1', $text);
1867 // Additionally, [pre] tags preserve spaces
1868 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", function ($match) {
1869 return str_replace(' ', ' ', $match[1]);
1875 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1876 function ($matches) {
1877 if (strpos($matches[2], "\n") !== false) {
1878 $return = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
1880 $return = '<code>' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
1888 // Clean up the HTML by loading and saving the HTML with the DOM.
1889 // Bad structured html can break a whole page.
1890 // For performance reasons do it only with activated item cache or at export.
1891 if (!$try_oembed || (get_itemcachepath() != '')) {
1892 $doc = new DOMDocument();
1893 $doc->preserveWhiteSpace = false;
1895 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1897 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1898 $encoding = '<?xml encoding="UTF-8">';
1899 @$doc->loadHTML($encoding . $doctype . '<html><body>' . $text . '</body></html>');
1900 $doc->encoding = 'UTF-8';
1901 $text = $doc->saveHTML();
1902 $text = str_replace(['<html><body>', '</body></html>', $doctype, $encoding], ['', '', '', ''], $text);
1904 $text = str_replace('<br></li>', '</li>', $text);
1906 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1909 // Clean up some useless linebreaks in lists
1910 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1911 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1912 //$Text = str_replace('</li><br />', '</li>', $Text);
1913 //$Text = str_replace('<br /><li>', '<li>', $Text);
1914 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1916 Hook::callAll('bbcode', $text);
1922 * Strips the "abstract" tag from the provided text
1924 * @param string $text The text with BBCode
1925 * @return string The same text - but without "abstract" element
1927 public static function stripAbstract($text)
1929 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1930 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1936 * Returns the value of the "abstract" element
1938 * @param string $text The text that maybe contains the element
1939 * @param string $addon The addon for which the abstract is meant for
1940 * @return string The abstract
1942 public static function getAbstract($text, $addon = '')
1946 $addon = strtolower($addon);
1948 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1949 foreach ($results AS $result) {
1950 $abstracts[strtolower($result[1])] = $result[2];
1954 if (isset($abstracts[$addon])) {
1955 $abstract = $abstracts[$addon];
1958 if ($abstract == '' && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1959 $abstract = $result[1];
1966 * Callback function to replace a Friendica style mention in a mention for Diaspora
1968 * @param array $match Matching values for the callback
1969 * [1] = Mention type (! or @)
1972 * @return string Replaced mention
1973 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1974 * @throws \ImagickException
1976 private static function bbCodeMention2DiasporaCallback($match)
1978 $contact = Contact::getDetailsByURL($match[3]);
1980 if (empty($contact['addr'])) {
1981 $contact = Probe::uri($match[3]);
1984 if (empty($contact['addr'])) {
1988 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
1993 * Converts a BBCode text into Markdown
1995 * This function converts a BBCode item body to be sent to Markdown-enabled
1996 * systems like Diaspora and Libertree
1998 * @param string $text
1999 * @param bool $for_diaspora Diaspora requires more changes than Libertree
2001 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2003 public static function toMarkdown($text, $for_diaspora = true)
2005 $original_text = $text;
2007 // Since Diaspora is creating a summary for links, this function removes them before posting
2008 if ($for_diaspora) {
2009 $text = self::removeShareInformation($text);
2013 * Transform #tags, strip off the [url] and replace spaces with underscore
2015 $url_search_string = "^\[\]";
2016 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
2017 function ($matches) {
2018 return '#' . str_replace(' ', '_', $matches[2]);
2023 // Converting images with size parameters to simple images. Markdown doesn't know it.
2024 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2026 // Convert it to HTML - don't try oembed
2027 if ($for_diaspora) {
2028 $text = self::convert($text, false, self::DIASPORA);
2030 // Add all tags that maybe were removed
2031 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2033 foreach ($tags[2] as $tag) {
2034 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2035 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2036 $tagline .= '#' . $tag . ' ';
2039 $text = $text . " " . $tagline;
2042 $text = self::convert($text, false, self::CONNECTORS);
2045 // If a link is followed by a quote then there should be a newline before it
2046 // Maybe we should make this newline at every time before a quote.
2047 $text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
2049 $stamp1 = microtime(true);
2051 // Now convert HTML to Markdown
2052 $text = HTML::toMarkdown($text);
2054 DI::profiler()->saveTimestamp($stamp1, "parser", System::callstack());
2056 // Libertree has a problem with escaped hashtags.
2057 $text = str_replace(['\#'], ['#'], $text);
2059 // Remove any leading or trailing whitespace, as this will mess up
2060 // the Diaspora signature verification and cause the item to disappear
2061 $text = trim($text);
2063 if ($for_diaspora) {
2064 $url_search_string = "^\[\]";
2065 $text = preg_replace_callback(
2066 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
2067 ['self', 'bbCodeMention2DiasporaCallback'],
2072 Hook::callAll('bb2diaspora', $text);
2078 * Pull out all #hashtags and @person tags from $string.
2080 * We also get @person@domain.com - which would make
2081 * the regex quite complicated as tags can also
2082 * end a sentence. So we'll run through our results
2083 * and strip the period from any tags which end with one.
2084 * Returns array of tags found, or empty array.
2086 * @param string $string Post content
2088 * @return array List of tag and person names
2090 public static function getTags($string)
2094 BBCode::performWithEscapedTags($string, ['noparse', 'pre', 'code'], function ($string) use (&$ret) {
2095 // Convert hashtag links to hashtags
2096 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2 ', $string);
2098 // Force line feeds at bbtags
2099 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
2101 // ignore anything in a bbtag
2102 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
2104 // Match full names against @tags including the space between first and last
2105 // We will look these up afterward to see if they are full names or not recognisable.
2107 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
2108 foreach ($matches[1] as $match) {
2109 if (strstr($match, ']')) {
2110 // we might be inside a bbcode color tag - leave it alone
2114 if (substr($match, -1, 1) === '.') {
2115 $ret[] = substr($match, 0, -1);
2122 // Otherwise pull out single word tags. These can be @nickname, @first_last
2125 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?\']*[^\^ \x0D\x0A,;:?!\'.])/', $string, $matches)) {
2126 foreach ($matches[1] as $match) {
2127 if (strstr($match, ']')) {
2128 // we might be inside a bbcode color tag - leave it alone
2132 // ignore strictly numeric tags like #1
2133 if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
2137 // try not to catch url fragments
2138 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
2147 return array_unique($ret);
2151 * Perform a custom function on a text after having escaped blocks enclosed in the provided tag list.
2153 * @param string $text
2154 * @param array $tagList A list of tag names, e.g ['noparse', 'nobb', 'pre']
2155 * @param callable $callback
2158 *@see Strings::performWithEscapedBlocks
2161 public static function performWithEscapedTags(string $text, array $tagList, callable $callback)
2163 $tagList = array_map('preg_quote', $tagList);
2165 return Strings::performWithEscapedBlocks($text, '#\[(?:' . implode('|', $tagList) . ').*?\[/(?:' . implode('|', $tagList) . ')]#ism', $callback);
2169 * Replaces mentions in the provided message body for the provided user and network if any
2172 * @param $profile_uid
2175 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2176 * @throws \ImagickException
2178 public static function setMentions($body, $profile_uid = 0, $network = '')
2180 BBCode::performWithEscapedTags($body, ['noparse', 'pre', 'code', 'img'], function ($body) use ($profile_uid, $network) {
2181 $tags = BBCode::getTags($body);
2186 foreach ($tags as $tag) {
2187 $tag_type = substr($tag, 0, 1);
2189 if ($tag_type == Tag::TAG_CHARACTER[Tag::HASHTAG]) {
2194 * If we already tagged 'Robert Johnson', don't try and tag 'Robert'.
2195 * Robert Johnson should be first in the $tags array
2197 foreach ($tagged as $nextTag) {
2198 if (stristr($nextTag, $tag . ' ')) {
2203 $success = Item::replaceTag($body, $inform, $profile_uid, $tag, $network);
2205 if ($success['replaced']) {
2217 * @param string $author Author display name
2218 * @param string $profile Author profile URL
2219 * @param string $avatar Author profile picture URL
2220 * @param string $link Post source URL
2221 * @param string $posted Post created date
2222 * @param string|null $guid Post guid (if any)
2224 * @TODO Rewrite to handle over whole record array
2226 public static function getShareOpeningTag(string $author, string $profile, string $avatar, string $link, string $posted, string $guid = null)
2228 $header = "[share author='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $author) .
2229 "' profile='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $profile) .
2230 "' avatar='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $avatar) .
2231 "' link='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $link) .
2232 "' posted='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $posted);
2235 $header .= "' guid='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $guid);