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\Object\Image;
42 use Friendica\Protocol\Activity;
43 use Friendica\Util\Images;
44 use Friendica\Util\Map;
45 use Friendica\Util\ParseUrl;
46 use Friendica\Util\Proxy as ProxyUtils;
47 use Friendica\Util\Strings;
48 use Friendica\Util\XML;
59 const ACTIVITYPUB = 9;
62 * Fetches attachment data that were generated the old way
64 * @param string $body Message body
66 * 'type' -> Message type ('link', 'video', 'photo')
67 * 'text' -> Text before the shared message
68 * 'after' -> Text after the shared message
69 * 'image' -> Preview image of the message
70 * 'url' -> Url to the attached message
71 * 'title' -> Title of the attachment
72 * 'description' -> Description of the attachment
73 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
75 private static function getOldAttachmentData($body)
79 // Simplify image codes
80 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
82 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
83 foreach ($attached as $data) {
84 if (!in_array($data[1], ['type-link', 'type-video', 'type-photo'])) {
88 $post['type'] = substr($data[1], 5);
90 $pos = strpos($body, $data[0]);
92 $post['text'] = trim(substr($body, 0, $pos));
93 $post['after'] = trim(substr($body, $pos + strlen($data[0])));
95 $post['text'] = trim(str_replace($data[0], '', $body));
99 $attacheddata = $data[2];
101 if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
103 $picturedata = Images::getInfoFromURLCached($matches[1]);
106 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
107 $post['image'] = $matches[1];
109 $post['preview'] = $matches[1];
114 if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
115 $post['url'] = $matches[1];
116 $post['title'] = $matches[2];
118 if (!empty($post['url']) && (in_array($post['type'], ['link', 'video']))
119 && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
120 $post['url'] = $matches[1];
123 // Search for description
124 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
125 $post['description'] = $matches[1];
133 * Fetches attachment data that were generated with the "attachment" element
135 * @param string $body Message body
137 * 'type' -> Message type ('link', 'video', 'photo')
138 * 'text' -> Text before the shared message
139 * 'after' -> Text after the shared message
140 * 'image' -> Preview image of the message
141 * 'url' -> Url to the attached message
142 * 'title' -> Title of the attachment
143 * 'description' -> Description of the attachment
144 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
146 public static function getAttachmentData($body)
154 'provider_name' => '',
155 'provider_url' => '',
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]);
258 $parts = parse_url($data['url']);
259 if (!empty($parts['scheme']) && !empty($parts['host'])) {
260 $data['provider_name'] = $parts['host'];
261 $data['provider_url'] = $parts['scheme'] . '://' . $parts['host'];
263 if (!empty($parts['port'])) {
264 $data['provider_url'] .= ':' . $parts['port'];
271 public static function getAttachedData($body, $item = [])
275 - type: link, video, photo
283 $has_title = !empty($item['title']);
284 $plink = $item['plink'] ?? '';
285 $post = self::getAttachmentData($body);
287 // Get all linked images with alternative image description
288 if (preg_match_all("/\[img=([^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
289 foreach ($pictures as $picture) {
290 if (Photo::isLocal($picture[1])) {
291 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2]];
294 if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
295 $post['image_description'] = $post['images'][0]['description'];
299 if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
300 foreach ($pictures as $picture) {
301 if (Photo::isLocal($picture[1])) {
302 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => ''];
307 // if nothing is found, it maybe having an image.
308 if (!isset($post['type'])) {
309 // Simplify image codes
310 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
311 $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
312 $post['text'] = $body;
314 if (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $body, $pictures, PREG_SET_ORDER)) {
315 if ((count($pictures) == 1) && !$has_title) {
316 if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
317 // Replace the preview picture with the real picture
318 $url = str_replace('-1.', '-0.', $pictures[0][2]);
319 $data = ['url' => $url, 'type' => 'photo'];
321 // Checking, if the link goes to a picture
322 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
326 // Sometimes photo posts to the own album are not detected at the start.
327 // So we seem to cannot use the cache for these cases. That's strange.
328 if (($data['type'] != 'photo') && strstr($pictures[0][1], "/photos/")) {
329 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
332 if ($data['type'] == 'photo') {
333 $post['type'] = 'photo';
334 if (isset($data['images'][0])) {
335 $post['image'] = $data['images'][0]['src'];
336 $post['url'] = $data['url'];
338 $post['image'] = $data['url'];
341 $post['preview'] = $pictures[0][2];
342 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
344 $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
345 if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
346 $post['type'] = 'photo';
347 $post['image'] = $pictures[0][1];
348 $post['preview'] = $pictures[0][2];
349 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
352 } elseif (count($pictures) > 0) {
353 $post['type'] = 'link';
354 $post['url'] = $plink;
355 $post['image'] = $pictures[0][2];
356 $post['text'] = $body;
358 foreach ($pictures as $picture) {
359 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
362 } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
363 if ((count($pictures) == 1) && !$has_title) {
364 $post['type'] = 'photo';
365 $post['image'] = $pictures[0][1];
366 $post['text'] = str_replace($pictures[0][0], '', $body);
367 } elseif (count($pictures) > 0) {
368 $post['type'] = 'link';
369 $post['url'] = $plink;
370 $post['image'] = $pictures[0][1];
371 $post['text'] = $body;
373 foreach ($pictures as $picture) {
374 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
379 // Test for the external links
380 preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER);
381 preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $links2, PREG_SET_ORDER);
383 $links = array_merge($links1, $links2);
385 // If there is only a single one, then use it.
386 // This should cover link posts via API.
387 if ((count($links) == 1) && !isset($post['preview']) && !$has_title) {
388 $post['type'] = 'link';
389 $post['url'] = $links[0][1];
392 // Simplify "video" element
393 $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
395 // Now count the number of external media links
396 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
397 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
398 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER);
399 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $links4, PREG_SET_ORDER);
401 // Add them to the other external links
402 $links = array_merge($links, $links1, $links2, $links3, $links4);
404 // Are there more than one?
405 if (count($links) > 1) {
406 // The post will be the type "text", which means a blog post
407 unset($post['type']);
408 $post['url'] = $plink;
411 if (!isset($post['type'])) {
412 $post['type'] = "text";
413 $post['text'] = trim($body);
415 } elseif (isset($post['url']) && ($post['type'] == 'video')) {
416 $data = ParseUrl::getSiteinfoCached($post['url'], true);
418 if (isset($data['images'][0])) {
419 $post['image'] = $data['images'][0]['src'];
427 * Remove [attachment] BBCode and replaces it with a regular [url]
429 * @param string $body
430 * @param boolean $no_link_desc No link description
432 * @return string with replaced body
434 public static function removeAttachment($body, $no_link_desc = false)
436 return preg_replace_callback("/\s*\[attachment (.*)\](.*?)\[\/attachment\]\s*/ism",
437 function ($match) use ($no_link_desc) {
438 $attach_data = self::getAttachmentData($match[0]);
439 if (empty($attach_data['url'])) {
441 } elseif (empty($attach_data['title']) || $no_link_desc) {
442 return "\n[url]" . $attach_data['url'] . "[/url]\n";
444 return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
450 * Converts a BBCode text into plaintext
453 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
457 public static function toPlaintext($text, $keep_urls = true)
459 $naked_text = HTML::toPlaintext(self::convert($text, false, 0, true), 0, !$keep_urls);
464 private static function proxyUrl($image, $simplehtml = self::INTERNAL)
466 // Only send proxied pictures to API and for internal display
467 if (in_array($simplehtml, [self::INTERNAL, self::API])) {
468 return ProxyUtils::proxifyUrl($image);
475 * This function changing the visual size (not the real size) of images.
476 * The function does not work for pictures with an alternate text description.
477 * This could only be changed by using some new "img" BBCode format.
479 * @param string $srctext The body with images
480 * @return string The body with possibly scaled images
482 public static function scaleExternalImages(string $srctext)
486 // Simplify image links
487 $s = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $s);
490 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
492 foreach ($matches as $mtch) {
493 Logger::log('scale_external_image: ' . $mtch[1]);
495 $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
496 if (stristr($mtch[1], $hostname)) {
500 $curlResult = DI::httpRequest()->get($mtch[1]);
501 if (!$curlResult->isSuccess()) {
505 $i = $curlResult->getBody();
506 $type = $curlResult->getContentType();
507 $type = Images::getMimeTypeByData($i, $mtch[1], $type);
510 $Image = new Image($i, $type);
511 if ($Image->isValid()) {
512 $orig_width = $Image->getWidth();
513 $orig_height = $Image->getHeight();
515 if ($orig_width > 640 || $orig_height > 640) {
516 $Image->scaleDown(640);
517 $new_width = $Image->getWidth();
518 $new_height = $Image->getHeight();
519 Logger::info('External images scaled', ['orig_width' => $orig_width, 'new_width' => $new_width, 'orig_height' => $orig_height, 'new_height' => $new_height, 'match' => $mtch[0]]);
522 '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
526 Logger::info('New string', ['image' => $s]);
537 * Truncates imported message body string length to max_import_size
539 * The purpose of this function is to apply system message length limits to
540 * imported messages without including any embedded photos in the length
542 * @param string $body
544 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
546 public static function limitBodySize($body)
548 $maxlen = DI::config()->get('config', 'max_import_size', 0);
550 // If the length of the body, including the embedded images, is smaller
551 // than the maximum, then don't waste time looking for the images
552 if ($maxlen && (strlen($body) > $maxlen)) {
554 Logger::info('the total body length exceeds the limit', ['maxlen' => $maxlen, 'body_len' => strlen($body)]);
560 $img_start = strpos($orig_body, '[img');
561 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
562 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
563 while (($img_st_close !== false) && ($img_end !== false)) {
565 $img_st_close++; // make it point to AFTER the closing bracket
566 $img_end += $img_start;
567 $img_end += strlen('[/img]');
569 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
570 // This is an embedded image
572 if (($textlen + $img_start) > $maxlen) {
573 if ($textlen < $maxlen) {
574 Logger::info('the limit happens before an embedded image');
575 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
579 $new_body = $new_body . substr($orig_body, 0, $img_start);
580 $textlen += $img_start;
583 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
586 if (($textlen + $img_end) > $maxlen) {
587 if ($textlen < $maxlen) {
588 Logger::info('the limit happens before the end of a non-embedded image');
589 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
593 $new_body = $new_body . substr($orig_body, 0, $img_end);
594 $textlen += $img_end;
597 $orig_body = substr($orig_body, $img_end);
599 if ($orig_body === false) {
600 // in case the body ends on a closing image tag
604 $img_start = strpos($orig_body, '[img');
605 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
606 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
609 if (($textlen + strlen($orig_body)) > $maxlen) {
610 if ($textlen < $maxlen) {
611 Logger::info('the limit happens after the end of the last image');
612 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
615 Logger::info('the text size with embedded images extracted did not violate the limit');
616 $new_body = $new_body . $orig_body;
626 * Processes [attachment] tags
628 * Note: Can produce a [bookmark] tag in the returned string
630 * @param string $text
631 * @param integer $simplehtml
632 * @param bool $tryoembed
634 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
636 private static function convertAttachment($text, $simplehtml = self::INTERNAL, $tryoembed = true)
638 $data = self::getAttachmentData($text);
639 if (empty($data) || empty($data['url'])) {
643 if (isset($data['title'])) {
644 $data['title'] = strip_tags($data['title']);
645 $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
647 $data['title'] = null;
650 if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
651 $data['preview'] = $data['image'];
657 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
658 $return = OEmbed::getHTML($data['url'], $data['title']);
660 throw new Exception('OEmbed is disabled for this attachment.');
662 } catch (Exception $e) {
663 $data['title'] = ($data['title'] ?? '') ?: $data['url'];
665 if ($simplehtml != self::CONNECTORS) {
666 $return = sprintf('<div class="type-%s">', $data['type']);
669 if (!empty($data['title']) && !empty($data['url'])) {
670 if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
671 $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), $data['title']);
673 if (!empty($data['image'])) {
674 $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), $data['title']);
675 } elseif (!empty($data['preview'])) {
676 $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']);
678 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
682 if (!empty($data['description']) && $data['description'] != $data['title']) {
683 // Sanitize the HTML by converting it to BBCode
684 $bbcode = HTML::toBBCode($data['description']);
685 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
688 if (!empty($data['url'])) {
689 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
692 if ($simplehtml != self::CONNECTORS) {
697 return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
700 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
702 $data = self::getAttachmentData($Text);
707 return $data['text'] . ($data['after'] ?? '');
710 $title = htmlentities($data['title'] ?? '', ENT_QUOTES, 'UTF-8', false);
711 $text = htmlentities($data['text'], ENT_QUOTES, 'UTF-8', false);
712 if ($plaintext || (($title != '') && strstr($text, $title))) {
713 $data['title'] = $data['url'];
714 } elseif (($text != '') && strstr($title, $text)) {
715 $data['text'] = $data['title'];
716 $data['title'] = $data['url'];
719 if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
720 return $data['title'] . $data['after'];
723 // If the link already is included in the post, don't add it again
724 if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
725 return $data['text'] . $data['after'];
728 $text = $data['text'];
730 if (!empty($data['url']) && !empty($data['title'])) {
731 $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
732 } elseif (!empty($data['url'])) {
733 $text .= "\n[url]" . $data['url'] . '[/url]';
736 return $text . "\n" . $data['after'];
740 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
742 * @param array $match Array with the matching values
743 * @return string reformatted link including HTML codes
745 private static function convertUrlForActivityPubCallback($match)
749 if (isset($match[2]) && ($match[1] != $match[2])) {
753 $parts = parse_url($url);
754 if (!isset($parts['scheme'])) {
758 return self::convertUrlForActivityPub($url);
762 * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
764 * @param string $url URL that is about to be reformatted
765 * @return string reformatted link including HTML codes
767 private static function convertUrlForActivityPub($url)
769 $html = '<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>';
770 return sprintf($html, $url, self::getStyledURL($url));
774 * Converts an URL in a nicer format (without the scheme and possibly shortened)
776 * @param string $url URL that is about to be reformatted
777 * @return string reformatted link
779 private static function getStyledURL($url)
781 $parts = parse_url($url);
782 $scheme = $parts['scheme'] . '://';
783 $styled_url = str_replace($scheme, '', $url);
785 if (strlen($styled_url) > 30) {
786 $styled_url = substr($styled_url, 0, 30) . "…";
793 * [noparse][i]italic[/i][/noparse] turns into
794 * [noparse][ i ]italic[ /i ][/noparse],
795 * to hide them from parser.
797 private static function escapeNoparseCallback($match)
799 $whole_match = $match[0];
800 $captured = $match[1];
801 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
802 $new_str = str_replace($captured, $spacefied, $whole_match);
807 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
808 * now turns back and the [noparse] tags are trimed
809 * returning [i]italic[/i]
811 private static function unescapeNoparseCallback($match)
813 $captured = $match[1];
814 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
819 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
822 * @param string $text Text to search
823 * @param string $name Tag name
824 * @param int $occurrences Number of first occurrences to skip
825 * @return boolean|array
827 public static function getTagPosition($text, $name, $occurrences = 0)
829 if ($occurrences < 0) {
834 for ($i = 0; $i <= $occurrences; $i++) {
835 if ($start_open !== false) {
836 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
840 if ($start_open === false) {
844 $start_equal = strpos($text, '=', $start_open);
845 $start_close = strpos($text, ']', $start_open);
847 if ($start_close === false) {
853 $end_open = strpos($text, '[/' . $name . ']', $start_close);
855 if ($end_open === false) {
861 'open' => $start_open,
862 'close' => $start_close
866 'close' => $end_open + strlen('[/' . $name . ']')
870 if ($start_equal !== false) {
871 $res['start']['equal'] = $start_equal + 1;
878 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
880 * @param string $pattern Preg pattern string
881 * @param string $replace Preg replace string
882 * @param string $name BBCode tag name
883 * @param string $text Text to search
886 public static function pregReplaceInTag($pattern, $replace, $name, $text)
889 $pos = self::getTagPosition($text, $name, $occurrences);
890 while ($pos !== false && $occurrences++ < 1000) {
891 $start = substr($text, 0, $pos['start']['open']);
892 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
893 $end = substr($text, $pos['end']['close']);
894 if ($end === false) {
898 $subject = preg_replace($pattern, $replace, $subject);
899 $text = $start . $subject . $end;
901 $pos = self::getTagPosition($text, $name, $occurrences);
907 private static function extractImagesFromItemBody($body)
914 $img_start = strpos($orig_body, '[img');
915 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
916 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
917 while (($img_st_close !== false) && ($img_end !== false)) {
918 $img_st_close++; // make it point to AFTER the closing bracket
919 $img_end += $img_start;
921 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
922 // This is an embedded image
923 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
924 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
928 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
931 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
933 if ($orig_body === false) {
934 // in case the body ends on a closing image tag
938 $img_start = strpos($orig_body, '[img');
939 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
940 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
943 $new_body = $new_body . $orig_body;
945 return ['body' => $new_body, 'images' => $saved_image];
948 private static function interpolateSavedImagesIntoItemBody($body, array $images)
953 foreach ($images as $image) {
954 // We're depending on the property of 'foreach' (specified on the PHP website) that
955 // it loops over the array starting from the first element and going sequentially
956 // to the last element
957 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
958 '<img src="' . self::proxyUrl($image) . '" alt="' . DI::l10n()->t('Image/photo') . '" />', $newbody);
966 * This function converts a [share] block to text according to a provided callback function whose signature is:
968 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
971 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
973 * - $author_contact is a contact record array
974 * - $content is the inner content of the [share] block
975 * - $is_quote_share indicates whether there's any content before the [share] block
976 * - Return value is the string that should replace the [share] block in the provided text
978 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
980 * @param string $text A BBCode string
981 * @param callable $callback
982 * @return string The BBCode string with all [share] blocks replaced
984 public static function convertShare($text, callable $callback)
986 $return = preg_replace_callback(
987 "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
988 function ($match) use ($callback) {
989 $attribute_string = $match[2];
991 foreach (['author', 'profile', 'avatar', 'link', 'posted', 'guid'] as $field) {
992 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
993 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
996 $author_contact = Contact::getByURL($attributes['profile'], false, ['url', 'addr', 'name', 'micro']);
997 $author_contact['url'] = ($author_contact['url'] ?? $attributes['profile']);
998 $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']);
1000 $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author'];
1001 $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
1002 $attributes['profile'] = ($author_contact['url'] ?? '') ?: $attributes['profile'];
1004 if ($attributes['avatar']) {
1005 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
1008 return $match[1] . $callback($attributes, $author_contact, $match[3], trim($match[1]) != '');
1017 * Default [share] tag conversion callback
1019 * Note: Can produce a [bookmark] tag in the output
1021 * @see BBCode::convertShare()
1022 * @param array $attributes [share] block attribute values
1023 * @param array $author_contact Contact row of the shared author
1024 * @param string $content Inner content of the [share] block
1025 * @param boolean $is_quote_share Whether there is content before the [share] block
1026 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
1028 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1030 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
1032 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
1034 switch ($simplehtml) {
1036 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
1038 case self::DIASPORA:
1039 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
1040 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
1042 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
1044 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
1045 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
1048 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
1050 if (empty($attributes['posted']) && !empty($attributes['link'])) {
1051 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
1056 case self::CONNECTORS:
1057 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
1058 $headline .= DI::l10n()->t('<a href="%1$s" target="_blank" rel="noopener noreferrer">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
1059 $headline .= ':</b></p>' . "\n";
1061 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
1065 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
1067 case self::ACTIVITYPUB:
1068 $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>';
1069 $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
1072 $text = ($is_quote_share? "\n" : '');
1074 $contact = Contact::getByURL($attributes['profile'], false, ['network']);
1075 $network = $contact['network'] ?? Protocol::PHANTOM;
1077 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1078 $text .= Renderer::replaceMacros($tpl, [
1079 '$profile' => $attributes['profile'],
1080 '$avatar' => $attributes['avatar'],
1081 '$author' => $attributes['author'],
1082 '$link' => $attributes['link'],
1083 '$link_title' => DI::l10n()->t('link to source'),
1084 '$posted' => $attributes['posted'],
1085 '$guid' => $attributes['guid'],
1086 '$network_name' => ContactSelector::networkToName($network, $attributes['profile']),
1087 '$network_icon' => ContactSelector::networkToIcon($network, $attributes['profile']),
1088 '$content' => self::setMentions(trim($content), 0, $network),
1096 private static function removePictureLinksCallback($match)
1098 $cache_key = 'remove:' . $match[1];
1099 $text = DI::cache()->get($cache_key);
1101 if (is_null($text)) {
1104 $stamp1 = microtime(true);
1106 $ch = @curl_init($match[1]);
1107 @curl_setopt($ch, CURLOPT_NOBODY, true);
1108 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1109 @curl_setopt($ch, CURLOPT_USERAGENT, DI::httpRequest()->getUserAgent());
1111 $curl_info = @curl_getinfo($ch);
1113 DI::profiler()->saveTimestamp($stamp1, "network");
1115 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1116 $text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
1118 $text = "[url=" . $match[2] . ']' . $match[2] . "[/url]";
1120 // if its not a picture then look if its a page that contains a picture link
1121 $body = DI::httpRequest()->fetch($match[1]);
1123 $doc = new DOMDocument();
1124 @$doc->loadHTML($body);
1125 $xpath = new DOMXPath($doc);
1126 $list = $xpath->query("//meta[@name]");
1127 foreach ($list as $node) {
1130 if ($node->attributes->length) {
1131 foreach ($node->attributes as $attribute) {
1132 $attr[$attribute->name] = $attribute->value;
1136 if (strtolower($attr['name']) == 'twitter:image') {
1137 $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
1141 DI::cache()->set($cache_key, $text);
1147 private static function expandLinksCallback($match)
1149 if (($match[3] == '') || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1150 return ($match[1] . "[url]" . $match[2] . "[/url]");
1152 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1156 private static function cleanPictureLinksCallback($match)
1160 // When the picture link is the own photo path then we can avoid fetching the link
1161 $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
1162 if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
1163 if (!empty($match[3])) {
1164 $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
1166 $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
1171 $cache_key = 'clean:' . $match[1];
1172 $text = DI::cache()->get($cache_key);
1173 if (!is_null($text)) {
1177 // Only fetch the header, not the content
1178 $stamp1 = microtime(true);
1180 $ch = @curl_init($match[1]);
1181 @curl_setopt($ch, CURLOPT_NOBODY, true);
1182 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1183 @curl_setopt($ch, CURLOPT_USERAGENT, DI::httpRequest()->getUserAgent());
1185 $curl_info = @curl_getinfo($ch);
1187 DI::profiler()->saveTimestamp($stamp1, "network");
1189 // if its a link to a picture then embed this picture
1190 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1191 $text = '[img]' . $match[1] . '[/img]';
1193 if (!empty($match[3])) {
1194 $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
1196 $text = '[img]' . $match[2] . '[/img]';
1199 // if its not a picture then look if its a page that contains a picture link
1200 $body = DI::httpRequest()->fetch($match[1]);
1202 $doc = new DOMDocument();
1203 @$doc->loadHTML($body);
1204 $xpath = new DOMXPath($doc);
1205 $list = $xpath->query("//meta[@name]");
1206 foreach ($list as $node) {
1208 if ($node->attributes->length) {
1209 foreach ($node->attributes as $attribute) {
1210 $attr[$attribute->name] = $attribute->value;
1214 if (strtolower($attr['name']) == "twitter:image") {
1215 if (!empty($match[3])) {
1216 $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
1218 $text = "[img]" . $attr['content'] . "[/img]";
1223 DI::cache()->set($cache_key, $text);
1228 public static function cleanPictureLinks($text)
1230 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1231 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $return);
1235 public static function removeLinks(string $bbcode)
1237 $bbcode = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $1 ', $bbcode);
1238 $bbcode = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $bbcode);
1240 $bbcode = preg_replace('/[@!#]\[url\=.*?\].*?\[\/url\]/ism', '', $bbcode);
1241 $bbcode = preg_replace("/\[url=[^\[\]]*\](.*)\[\/url\]/Usi", ' $1 ', $bbcode);
1242 $bbcode = preg_replace('/[@!#]?\[url.*?\[\/url\]/ism', '', $bbcode);
1247 * Converts a BBCode message to HTML message
1249 * BBcode 2 HTML was written by WAY2WEB.net
1250 * extended to work with Mistpark/Friendica - Mike Macgirvin
1252 * Simple HTML values meaning:
1253 * - 0: Friendica display
1255 * - 2: Used for Windows Phone push, Friendica API
1256 * - 3: Used before converting to Markdown in bb2diaspora.php
1257 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1260 * - 7: Used for dfrn, OStatus
1261 * - 8: Used for Twitter, WP backlink text setting
1264 * @param string $text
1265 * @param bool $try_oembed
1266 * @param int $simple_html
1267 * @param bool $for_plaintext
1269 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1271 public static function convert(string $text = null, $try_oembed = true, $simple_html = self::INTERNAL, $for_plaintext = false)
1273 // Accounting for null default column values
1274 if (is_null($text) || $text === '') {
1280 $text = self::performWithEscapedTags($text, ['code'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
1281 $text = self::performWithEscapedTags($text, ['noparse', 'nobb', 'pre'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
1283 * preg_match_callback function to replace potential Oembed tags with Oembed content
1285 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1287 * $match[2] = $title or absent
1289 $try_oembed_callback = function ($match)
1292 $title = $match[2] ?? null;
1295 $return = OEmbed::getHTML($url, $title);
1296 } catch (Exception $ex) {
1297 $return = $match[0];
1305 // Remove the abstract element. It is a non visible element.
1306 $text = self::stripAbstract($text);
1308 // Move new lines outside of tags
1309 $text = preg_replace("#\[(\w*)](\n*)#ism", '$2[$1]', $text);
1310 $text = preg_replace("#(\n*)\[/(\w*)]#ism", '[/$2]$1', $text);
1312 // Extract the private images which use data urls since preg has issues with
1313 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1314 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1316 $extracted = self::extractImagesFromItemBody($text);
1317 $text = $extracted['body'];
1318 $saved_image = $extracted['images'];
1320 // If we find any event code, turn it into an event.
1321 // After we're finished processing the bbcode we'll
1322 // replace all of the event code with a reformatted version.
1324 $ev = Event::fromBBCode($text);
1326 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1327 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1329 $text = str_replace("<", "<", $text);
1330 $text = str_replace(">", ">", $text);
1332 // remove some newlines before the general conversion
1333 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1334 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1336 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1338 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1341 // Convert new line chars to html <br /> tags
1343 // nlbr seems to be hopelessly messed up
1344 // $Text = nl2br($Text);
1346 // We'll emulate it.
1348 $text = trim($text);
1349 $text = str_replace("\r\n", "\n", $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 // removing multiplicated newlines
1371 if (DI::config()->get('system', 'remove_multiplicated_lines')) {
1372 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1373 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1374 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1375 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1378 $text = str_replace($search, $replace, $text);
1379 } while ($oldtext != $text);
1382 /// @todo Have a closer look at the different html modes
1383 // Handle attached links or videos
1384 if ($simple_html == self::ACTIVITYPUB) {
1385 $text = self::removeAttachment($text);
1386 } elseif (!in_array($simple_html, [self::INTERNAL, self::CONNECTORS])) {
1387 $text = self::removeAttachment($text, true);
1389 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1392 // leave open the posibility of [map=something]
1393 // this is replaced in Item::prepareBody() which has knowledge of the item location
1394 if (strpos($text, '[/map]') !== false) {
1395 $text = preg_replace_callback(
1396 "/\[map\](.*?)\[\/map\]/ism",
1397 function ($match) use ($simple_html) {
1398 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1404 if (strpos($text, '[map=') !== false) {
1405 $text = preg_replace_callback(
1406 "/\[map=(.*?)\]/ism",
1407 function ($match) use ($simple_html) {
1408 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1414 if (strpos($text, '[map]') !== false) {
1415 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1418 // Check for headers
1419 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1420 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1421 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1422 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1423 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1424 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1426 // Check for paragraph
1427 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1429 // Check for bold text
1430 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1432 // Check for Italics text
1433 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1435 // Check for Underline text
1436 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1438 // Check for strike-through text
1439 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1441 // Check for over-line text
1442 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1444 // Check for colored text
1445 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1447 // Check for sized text
1448 // [size=50] --> font-size: 50px (with the unit).
1449 if ($simple_html != self::DIASPORA) {
1450 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1451 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1453 // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
1454 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text);
1458 // Check for centered text
1459 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1461 // Check for list text
1462 $text = str_replace("[*]", "<li>", $text);
1464 // Check for style sheet commands
1465 $text = preg_replace_callback(
1466 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1468 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1473 // Check for CSS classes
1474 $text = preg_replace_callback(
1475 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1477 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1482 // handle nested lists
1485 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1486 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1487 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1488 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1489 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1490 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1491 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1492 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1493 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1494 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1495 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1496 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1497 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1498 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1501 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1502 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1503 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1504 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1506 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1507 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1509 $text = str_replace('[hr]', '<hr />', $text);
1511 if (!$for_plaintext) {
1512 $text = self::performWithEscapedTags($text, ['url', 'img', 'audio', 'video', 'youtube', 'vimeo', 'share', 'attachment', 'iframe', 'bookmark'], function ($text) {
1513 return preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1517 // This is actually executed in Item::prepareBody()
1519 $nosmile = strpos($text, '[nosmile]') !== false;
1520 $text = str_replace('[nosmile]', '', $text);
1522 // Check for font change text
1523 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1525 // Declare the format for [spoiler] layout
1526 $SpoilerLayout = '<details class="spoiler"><summary>' . DI::l10n()->t('Click to open/close') . '</summary>$1</details>';
1528 // Check for [spoiler] text
1529 // handle nested quotes
1531 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1532 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", $SpoilerLayout, $text);
1535 // Check for [spoiler=Title] text
1537 // handle nested quotes
1539 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1540 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1541 '<details class="spoiler"><summary>$1</summary>$2</details>',
1545 // Declare the format for [quote] layout
1546 $QuoteLayout = '<blockquote>$1</blockquote>';
1548 // Check for [quote] text
1549 // handle nested quotes
1551 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1552 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1555 // Check for [quote=Author] text
1557 $t_wrote = DI::l10n()->t('$1 wrote:');
1559 // handle nested quotes
1561 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1562 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1563 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1568 // [img=widthxheight]image source[/img]
1569 $text = preg_replace_callback(
1570 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1571 function ($matches) use ($simple_html) {
1572 if (strpos($matches[3], "data:image/") === 0) {
1576 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1577 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1582 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1583 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1585 $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/ism",
1586 function ($matches) use ($simple_html) {
1587 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1588 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1589 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '" title="' . $matches[2] . '">';
1594 // [img]pathtoimage[/img]
1595 $text = preg_replace_callback(
1596 "/\[img\](.*?)\[\/img\]/ism",
1597 function ($matches) use ($simple_html) {
1598 if (strpos($matches[1], "data:image/") === 0) {
1602 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1603 return "[img]" . $matches[1] . "[/img]";
1608 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1609 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1611 $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);
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 />', $text);
1613 //$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);
1615 // Simplify "video" element
1616 $text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
1620 $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism",
1621 '<video src="$1" controls="controls" width="' . $a->videowidth . '" height="' . $a->videoheight . '" loop="true"><a href="$1">$1</a></video>', $text);
1622 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1623 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1624 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
1626 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1627 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $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 // html5 video and audio
1639 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1641 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1644 // Youtube extensions
1646 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1647 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1648 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1651 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1652 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1653 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1656 $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);
1658 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1659 '<a href="https://www.youtube.com/watch?v=$1" target="_blank" rel="noopener noreferrer">https://www.youtube.com/watch?v=$1</a>', $text);
1663 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1664 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1667 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1668 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1671 $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);
1673 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1674 '<a href="https://vimeo.com/$1" target="_blank" rel="noopener noreferrer">https://vimeo.com/$1</a>', $text);
1678 $text = OEmbed::BBCode2HTML($text);
1680 // Avoid triple linefeeds through oembed
1681 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1683 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1684 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1685 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1686 // start which is always required). Allow desc with a missing summary for compatibility.
1688 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1689 $sub = Event::getHTML($ev, $simple_html);
1691 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1692 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1693 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1694 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1695 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1696 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1697 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1700 // Replace non graphical smilies for external posts
1701 if (!$nosmile && !$for_plaintext) {
1702 $text = Smilies::replace($text);
1705 if (!$for_plaintext && DI::config()->get('system', 'big_emojis') && ($simple_html != self::DIASPORA)) {
1706 $conv = html_entity_decode(str_replace([' ', "\n", "\r"], '', $text));
1707 // Emojis are always 4 byte Unicode characters
1708 if (!empty($conv) && (strlen($conv) / mb_strlen($conv) == 4)) {
1709 $text = '<span style="font-size: xx-large; line-height: initial;">' . $text . '</span>';
1713 if (!$for_plaintext) {
1714 if (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1715 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1716 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1719 $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1720 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1723 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1725 // Remove all hashtag addresses
1726 if ($simple_html && !in_array($simple_html, [self::DIASPORA, self::OSTATUS, self::ACTIVITYPUB])) {
1727 $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1728 } elseif ($simple_html == self::DIASPORA) {
1729 // The ! is converted to @ since Diaspora only understands the @
1730 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1731 '@<a href="$2">$3</a>',
1733 } elseif (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1734 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1735 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1737 } elseif (!$simple_html) {
1738 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1739 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1743 // Bookmarks in red - will be converted to bookmarks in friendica
1744 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1745 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1746 $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1747 "[bookmark=$1]$2[/bookmark]", $text);
1749 if (in_array($simple_html, [self::API, self::OSTATUS, self::TWITTER])) {
1750 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1751 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1752 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1755 // Perform URL Search
1757 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1760 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1762 // Handle Diaspora posts
1763 $text = preg_replace_callback(
1764 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1766 return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1770 $text = preg_replace_callback(
1771 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1773 return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1777 // Server independent link to posts and comments
1778 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1779 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1780 $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
1784 * - #[url=<anything>]<term>[/url]
1785 * - [url=<anything>]#<term>[/url]
1787 $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) use ($simple_html) {
1788 if ($simple_html == BBCode::ACTIVITYPUB) {
1789 return '<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1790 . '" data-tag="' . XML::escape($matches[1]) . '" rel="tag ugc">#'
1791 . XML::escape($matches[1]) . '</a>';
1793 return '#<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1794 . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
1795 . XML::escape($matches[1]) . '</a>';
1799 // We need no target="_blank" rel="noopener noreferrer" for local links
1800 // convert links start with DI::baseUrl() as local link without the target="_blank" rel="noopener noreferrer" attribute
1801 $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
1802 $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1803 $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1805 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1806 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1808 // Red compatibility, though the link can't be authenticated on Friendica
1809 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1812 // we may need to restrict this further if it picks up too many strays
1813 // link acct:user@host to a webfinger profile redirector
1815 $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);
1817 // Perform MAIL Search
1818 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1819 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1821 /// @todo What is the meaning of these lines?
1822 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1823 $text = preg_replace('/\&\#039\;/', '\'', $text);
1825 // Currently deactivated, it made problems with " inside of alt texts.
1826 //$text = preg_replace('/\"\;/', '"', $text);
1828 // fix any escaped ampersands that may have been converted into links
1829 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1831 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1832 $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:'];
1834 array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
1836 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1837 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $text);
1839 // sanitize href attributes (only allowlisted protocols URLs)
1840 // default value for backward compatibility
1841 $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
1843 // Always allowed protocol even if config isn't set or not including it
1844 $allowed_link_protocols[] = '//';
1845 $allowed_link_protocols[] = 'http://';
1846 $allowed_link_protocols[] = 'https://';
1847 $allowed_link_protocols[] = 'redir/';
1849 array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
1851 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1852 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
1855 $text = self::convertShare(
1857 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1858 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1862 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1865 }); // Escaped noparse, nobb, pre
1867 // Remove escaping tags
1868 $text = preg_replace("/\[noparse\](.*?)\[\/noparse\]/ism", '\1', $text);
1869 $text = preg_replace("/\[nobb\](.*?)\[\/nobb\]/ism", '\1', $text);
1871 // Additionally, [pre] tags preserve spaces
1872 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", function ($match) {
1873 return str_replace(' ', ' ', $match[1]);
1879 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1880 function ($matches) {
1881 if (strpos($matches[2], "\n") !== false) {
1882 $return = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
1884 $return = '<code>' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
1892 // Clean up the HTML by loading and saving the HTML with the DOM.
1893 // Bad structured html can break a whole page.
1894 // For performance reasons do it only with activated item cache or at export.
1895 if (!$try_oembed || (get_itemcachepath() != '')) {
1896 $doc = new DOMDocument();
1897 $doc->preserveWhiteSpace = false;
1899 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1901 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1902 $encoding = '<?xml encoding="UTF-8">';
1903 @$doc->loadHTML($encoding . $doctype . '<html><body>' . $text . '</body></html>');
1904 $doc->encoding = 'UTF-8';
1905 $text = $doc->saveHTML();
1906 $text = str_replace(['<html><body>', '</body></html>', $doctype, $encoding], ['', '', '', ''], $text);
1908 $text = str_replace('<br></li>', '</li>', $text);
1910 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1913 // Clean up some useless linebreaks in lists
1914 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1915 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1916 //$Text = str_replace('</li><br />', '</li>', $Text);
1917 //$Text = str_replace('<br /><li>', '<li>', $Text);
1918 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1920 Hook::callAll('bbcode', $text);
1926 * Strips the "abstract" tag from the provided text
1928 * @param string $text The text with BBCode
1929 * @return string The same text - but without "abstract" element
1931 public static function stripAbstract($text)
1933 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1934 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1940 * Returns the value of the "abstract" element
1942 * @param string $text The text that maybe contains the element
1943 * @param string $addon The addon for which the abstract is meant for
1944 * @return string The abstract
1946 public static function getAbstract($text, $addon = '')
1950 $addon = strtolower($addon);
1952 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1953 foreach ($results AS $result) {
1954 $abstracts[strtolower($result[1])] = $result[2];
1958 if (isset($abstracts[$addon])) {
1959 $abstract = $abstracts[$addon];
1962 if ($abstract == '' && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1963 $abstract = $result[1];
1970 * Callback function to replace a Friendica style mention in a mention for Diaspora
1972 * @param array $match Matching values for the callback
1973 * [1] = Mention type (! or @)
1976 * @return string Replaced mention
1977 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1978 * @throws \ImagickException
1980 private static function bbCodeMention2DiasporaCallback($match)
1982 $contact = Contact::getByURL($match[3], false, ['addr']);
1983 if (empty($contact['addr'])) {
1987 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
1992 * Converts a BBCode text into Markdown
1994 * This function converts a BBCode item body to be sent to Markdown-enabled
1995 * systems like Diaspora and Libertree
1997 * @param string $text
1998 * @param bool $for_diaspora Diaspora requires more changes than Libertree
2000 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2002 public static function toMarkdown($text, $for_diaspora = true)
2004 $original_text = $text;
2006 // Since Diaspora is creating a summary for links, this function removes them before posting
2007 if ($for_diaspora) {
2008 $text = self::removeShareInformation($text);
2012 * Transform #tags, strip off the [url] and replace spaces with underscore
2014 $url_search_string = "^\[\]";
2015 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
2016 function ($matches) {
2017 return '#' . str_replace(' ', '_', $matches[2]);
2022 // Converting images with size parameters to simple images. Markdown doesn't know it.
2023 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2025 // Convert it to HTML - don't try oembed
2026 if ($for_diaspora) {
2027 $text = self::convert($text, false, self::DIASPORA);
2029 // Add all tags that maybe were removed
2030 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2032 foreach ($tags[2] as $tag) {
2033 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2034 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2035 $tagline .= '#' . $tag . ' ';
2038 $text = $text . " " . $tagline;
2041 $text = self::convert($text, false, self::CONNECTORS);
2044 // If a link is followed by a quote then there should be a newline before it
2045 // Maybe we should make this newline at every time before a quote.
2046 $text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
2048 $stamp1 = microtime(true);
2050 // Now convert HTML to Markdown
2051 $text = HTML::toMarkdown($text);
2053 DI::profiler()->saveTimestamp($stamp1, "parser");
2055 // Libertree has a problem with escaped hashtags.
2056 $text = str_replace(['\#'], ['#'], $text);
2058 // Remove any leading or trailing whitespace, as this will mess up
2059 // the Diaspora signature verification and cause the item to disappear
2060 $text = trim($text);
2062 if ($for_diaspora) {
2063 $url_search_string = "^\[\]";
2064 $text = preg_replace_callback(
2065 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
2066 ['self', 'bbCodeMention2DiasporaCallback'],
2071 Hook::callAll('bb2diaspora', $text);
2077 * Pull out all #hashtags and @person tags from $string.
2079 * We also get @person@domain.com - which would make
2080 * the regex quite complicated as tags can also
2081 * end a sentence. So we'll run through our results
2082 * and strip the period from any tags which end with one.
2083 * Returns array of tags found, or empty array.
2085 * @param string $string Post content
2087 * @return array List of tag and person names
2089 public static function getTags($string)
2093 BBCode::performWithEscapedTags($string, ['noparse', 'pre', 'code'], function ($string) use (&$ret) {
2094 // Convert hashtag links to hashtags
2095 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2 ', $string);
2097 // Force line feeds at bbtags
2098 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
2100 // ignore anything in a bbtag
2101 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
2103 // Match full names against @tags including the space between first and last
2104 // We will look these up afterward to see if they are full names or not recognisable.
2106 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
2107 foreach ($matches[1] as $match) {
2108 if (strstr($match, ']')) {
2109 // we might be inside a bbcode color tag - leave it alone
2113 if (substr($match, -1, 1) === '.') {
2114 $ret[] = substr($match, 0, -1);
2121 // Otherwise pull out single word tags. These can be @nickname, @first_last
2124 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?\']*[^\^ \x0D\x0A,;:?!\'.])/', $string, $matches)) {
2125 foreach ($matches[1] as $match) {
2126 if (strstr($match, ']')) {
2127 // we might be inside a bbcode color tag - leave it alone
2131 // ignore strictly numeric tags like #1
2132 if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
2136 // try not to catch url fragments
2137 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
2146 return array_unique($ret);
2150 * Perform a custom function on a text after having escaped blocks enclosed in the provided tag list.
2152 * @param string $text
2153 * @param array $tagList A list of tag names, e.g ['noparse', 'nobb', 'pre']
2154 * @param callable $callback
2157 *@see Strings::performWithEscapedBlocks
2160 public static function performWithEscapedTags(string $text, array $tagList, callable $callback)
2162 $tagList = array_map('preg_quote', $tagList);
2164 return Strings::performWithEscapedBlocks($text, '#\[(?:' . implode('|', $tagList) . ').*?\[/(?:' . implode('|', $tagList) . ')]#ism', $callback);
2168 * Replaces mentions in the provided message body for the provided user and network if any
2171 * @param $profile_uid
2174 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2175 * @throws \ImagickException
2177 public static function setMentions($body, $profile_uid = 0, $network = '')
2179 BBCode::performWithEscapedTags($body, ['noparse', 'pre', 'code', 'img'], function ($body) use ($profile_uid, $network) {
2180 $tags = BBCode::getTags($body);
2185 foreach ($tags as $tag) {
2186 $tag_type = substr($tag, 0, 1);
2188 if ($tag_type == Tag::TAG_CHARACTER[Tag::HASHTAG]) {
2193 * If we already tagged 'Robert Johnson', don't try and tag 'Robert'.
2194 * Robert Johnson should be first in the $tags array
2196 foreach ($tagged as $nextTag) {
2197 if (stristr($nextTag, $tag . ' ')) {
2202 $success = Item::replaceTag($body, $inform, $profile_uid, $tag, $network);
2204 if ($success['replaced']) {
2216 * @param string $author Author display name
2217 * @param string $profile Author profile URL
2218 * @param string $avatar Author profile picture URL
2219 * @param string $link Post source URL
2220 * @param string $posted Post created date
2221 * @param string|null $guid Post guid (if any)
2223 * @TODO Rewrite to handle over whole record array
2225 public static function getShareOpeningTag(string $author, string $profile, string $avatar, string $link, string $posted, string $guid = null)
2227 $header = "[share author='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $author) .
2228 "' profile='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $profile) .
2229 "' avatar='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $avatar) .
2230 "' link='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $link) .
2231 "' posted='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $posted);
2234 $header .= "' guid='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $guid);