3 * @file src/Content/Text/BBCode.php
6 namespace Friendica\Content\Text;
11 use Friendica\Content\OEmbed;
12 use Friendica\Content\Smilies;
13 use Friendica\Core\Config;
14 use Friendica\Core\Hook;
15 use Friendica\Core\L10n;
16 use Friendica\Core\Logger;
17 use Friendica\Core\Protocol;
18 use Friendica\Core\Renderer;
19 use Friendica\Core\System;
21 use Friendica\Model\Contact;
22 use Friendica\Model\Event;
23 use Friendica\Model\Photo;
24 use Friendica\Network\Probe;
25 use Friendica\Object\Image;
26 use Friendica\Protocol\Activity;
27 use Friendica\Util\Images;
28 use Friendica\Util\Map;
29 use Friendica\Util\Network;
30 use Friendica\Util\ParseUrl;
31 use Friendica\Util\Proxy as ProxyUtils;
32 use Friendica\Util\Strings;
33 use Friendica\Util\XML;
38 * @brief Fetches attachment data that were generated the old way
40 * @param string $body Message body
42 * 'type' -> Message type ('link', 'video', 'photo')
43 * 'text' -> Text before the shared message
44 * 'after' -> Text after the shared message
45 * 'image' -> Preview image of the message
46 * 'url' -> Url to the attached message
47 * 'title' -> Title of the attachment
48 * 'description' -> Description of the attachment
49 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
51 private static function getOldAttachmentData($body)
55 // Simplify image codes
56 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
58 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
59 foreach ($attached as $data) {
60 if (!in_array($data[1], ['type-link', 'type-video', 'type-photo'])) {
64 $post['type'] = substr($data[1], 5);
66 $pos = strpos($body, $data[0]);
68 $post['text'] = trim(substr($body, 0, $pos));
69 $post['after'] = trim(substr($body, $pos + strlen($data[0])));
71 $post['text'] = trim(str_replace($data[0], '', $body));
75 $attacheddata = $data[2];
77 if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
79 $picturedata = Images::getInfoFromURLCached($matches[1]);
82 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
83 $post['image'] = $matches[1];
85 $post['preview'] = $matches[1];
90 if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
91 $post['url'] = $matches[1];
92 $post['title'] = $matches[2];
94 if (!empty($post['url']) && (in_array($post['type'], ['link', 'video']))
95 && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
96 $post['url'] = $matches[1];
99 // Search for description
100 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
101 $post['description'] = $matches[1];
109 * @brief Fetches attachment data that were generated with the "attachment" element
111 * @param string $body Message body
113 * 'type' -> Message type ('link', 'video', 'photo')
114 * 'text' -> Text before the shared message
115 * 'after' -> Text after the shared message
116 * 'image' -> Preview image of the message
117 * 'url' -> Url to the attached message
118 * 'title' -> Title of the attachment
119 * 'description' -> Description of the attachment
120 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
122 public static function getAttachmentData($body)
134 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
135 return self::getOldAttachmentData($body);
138 $attributes = $match[2];
140 $data['text'] = trim($match[1]);
143 preg_match("/type='(.*?)'/ism", $attributes, $matches);
144 if (!empty($matches[1])) {
145 $type = strtolower($matches[1]);
148 preg_match('/type="(.*?)"/ism', $attributes, $matches);
149 if (!empty($matches[1])) {
150 $type = strtolower($matches[1]);
157 if (!in_array($type, ['link', 'audio', 'photo', 'video'])) {
162 $data['type'] = $type;
166 preg_match("/url='(.*?)'/ism", $attributes, $matches);
167 if (!empty($matches[1])) {
171 preg_match('/url="(.*?)"/ism', $attributes, $matches);
172 if (!empty($matches[1])) {
177 $data['url'] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
181 preg_match("/title='(.*?)'/ism", $attributes, $matches);
182 if (!empty($matches[1])) {
183 $title = $matches[1];
186 preg_match('/title="(.*?)"/ism', $attributes, $matches);
187 if (!empty($matches[1])) {
188 $title = $matches[1];
192 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
193 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
194 $title = str_replace(['[', ']'], ['[', ']'], $title);
195 $data['title'] = $title;
199 preg_match("/image='(.*?)'/ism", $attributes, $matches);
200 if (!empty($matches[1])) {
201 $image = $matches[1];
204 preg_match('/image="(.*?)"/ism', $attributes, $matches);
205 if (!empty($matches[1])) {
206 $image = $matches[1];
210 $data['image'] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
214 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
215 if (!empty($matches[1])) {
216 $preview = $matches[1];
219 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
220 if (!empty($matches[1])) {
221 $preview = $matches[1];
224 if ($preview != '') {
225 $data['preview'] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
228 $data['description'] = trim($match[3]);
230 $data['after'] = trim($match[4]);
235 public static function getAttachedData($body, $item = [])
239 - type: link, video, photo
247 $has_title = !empty($item['title']);
248 $plink = $item['plink'] ?? '';
249 $post = self::getAttachmentData($body);
251 // Get all linked images with alternative image description
252 if (preg_match_all("/\[img=([^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
253 foreach ($pictures as $picture) {
254 if (Photo::isLocal($picture[1])) {
255 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2]];
258 if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
259 $post['image_description'] = $post['images'][0]['description'];
263 if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
264 foreach ($pictures as $picture) {
265 if (Photo::isLocal($picture[1])) {
266 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => ''];
271 // if nothing is found, it maybe having an image.
272 if (!isset($post['type'])) {
273 // Simplify image codes
274 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
275 $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
276 $post['text'] = $body;
278 if (preg_match_all("(\[url=(.*?)\]\s*\[img\](.*?)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
279 if ((count($pictures) == 1) && !$has_title) {
280 if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
281 // Replace the preview picture with the real picture
282 $url = str_replace('-1.', '-0.', $pictures[0][2]);
283 $data = ['url' => $url, 'type' => 'photo'];
285 // Checking, if the link goes to a picture
286 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
290 // Sometimes photo posts to the own album are not detected at the start.
291 // So we seem to cannot use the cache for these cases. That's strange.
292 if (($data['type'] != 'photo') && strstr($pictures[0][1], "/photos/")) {
293 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
296 if ($data['type'] == 'photo') {
297 $post['type'] = 'photo';
298 if (isset($data['images'][0])) {
299 $post['image'] = $data['images'][0]['src'];
300 $post['url'] = $data['url'];
302 $post['image'] = $data['url'];
305 $post['preview'] = $pictures[0][2];
306 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
308 $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
309 if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
310 $post['type'] = 'photo';
311 $post['image'] = $pictures[0][1];
312 $post['preview'] = $pictures[0][2];
313 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
316 } elseif (count($pictures) > 0) {
317 $post['type'] = 'link';
318 $post['url'] = $plink;
319 $post['image'] = $pictures[0][2];
320 $post['text'] = $body;
322 foreach ($pictures as $picture) {
323 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
326 } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
327 if ((count($pictures) == 1) && !$has_title) {
328 $post['type'] = 'photo';
329 $post['image'] = $pictures[0][1];
330 $post['text'] = str_replace($pictures[0][0], '', $body);
331 } elseif (count($pictures) > 0) {
332 $post['type'] = 'link';
333 $post['url'] = $plink;
334 $post['image'] = $pictures[0][1];
335 $post['text'] = $body;
337 foreach ($pictures as $picture) {
338 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
343 // Test for the external links
344 preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER);
345 preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $links2, PREG_SET_ORDER);
347 $links = array_merge($links1, $links2);
349 // If there is only a single one, then use it.
350 // This should cover link posts via API.
351 if ((count($links) == 1) && !isset($post['preview']) && !$has_title) {
352 $post['type'] = 'link';
353 $post['url'] = $links[0][1];
356 // Simplify "video" element
357 $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
359 // Now count the number of external media links
360 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
361 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
362 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER);
363 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $links4, PREG_SET_ORDER);
365 // Add them to the other external links
366 $links = array_merge($links, $links1, $links2, $links3, $links4);
368 // Are there more than one?
369 if (count($links) > 1) {
370 // The post will be the type "text", which means a blog post
371 unset($post['type']);
372 $post['url'] = $plink;
375 if (!isset($post['type'])) {
376 $post['type'] = "text";
377 $post['text'] = trim($body);
379 } elseif (isset($post['url']) && ($post['type'] == 'video')) {
380 $data = ParseUrl::getSiteinfoCached($post['url'], true);
382 if (isset($data['images'][0])) {
383 $post['image'] = $data['images'][0]['src'];
391 * Remove [attachment] BBCode and replaces it with a regular [url]
393 * @param string $body
394 * @param boolean $no_link_desc No link description
396 * @return string with replaced body
398 public static function removeAttachment($body, $no_link_desc = false)
400 return preg_replace_callback("/\s*\[attachment (.*)\](.*?)\[\/attachment\]\s*/ism",
401 function ($match) use ($no_link_desc) {
402 $attach_data = self::getAttachmentData($match[0]);
403 if (empty($attach_data['url'])) {
405 } elseif (empty($attach_data['title']) || $no_link_desc) {
406 return "\n[url]" . $attach_data['url'] . "[/url]\n";
408 return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
414 * @brief Converts a BBCode text into plaintext
417 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
421 public static function toPlaintext($text, $keep_urls = true)
423 $naked_text = HTML::toPlaintext(BBCode::convert($text, false, 0, true), 0, !$keep_urls);
428 private static function proxyUrl($image, $simplehtml = false)
430 // Only send proxied pictures to API and for internal display
431 if (in_array($simplehtml, [false, 2])) {
432 return ProxyUtils::proxifyUrl($image);
438 public static function scaleExternalImages($srctext)
443 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
445 foreach ($matches as $mtch) {
446 Logger::log('scale_external_image: ' . $mtch[1]);
448 $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
449 if (stristr($mtch[1], $hostname)) {
453 $i = Network::fetchUrl($mtch[1]);
458 // guess mimetype from headers or filename
459 $type = Images::guessType($mtch[1], true);
462 $Image = new Image($i, $type);
463 if ($Image->isValid()) {
464 $orig_width = $Image->getWidth();
465 $orig_height = $Image->getHeight();
467 if ($orig_width > 640 || $orig_height > 640) {
468 $Image->scaleDown(640);
469 $new_width = $Image->getWidth();
470 $new_height = $Image->getHeight();
471 Logger::log('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], Logger::DEBUG);
474 '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
478 Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG);
489 * The purpose of this function is to apply system message length limits to
490 * imported messages without including any embedded photos in the length
492 * @brief Truncates imported message body string length to max_import_size
493 * @param string $body
495 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
497 public static function limitBodySize($body)
499 $maxlen = Config::get('config', 'max_import_size', 0);
501 // If the length of the body, including the embedded images, is smaller
502 // than the maximum, then don't waste time looking for the images
503 if ($maxlen && (strlen($body) > $maxlen)) {
505 Logger::log('the total body length exceeds the limit', Logger::DEBUG);
511 $img_start = strpos($orig_body, '[img');
512 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
513 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
514 while (($img_st_close !== false) && ($img_end !== false)) {
516 $img_st_close++; // make it point to AFTER the closing bracket
517 $img_end += $img_start;
518 $img_end += strlen('[/img]');
520 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
521 // This is an embedded image
523 if (($textlen + $img_start) > $maxlen) {
524 if ($textlen < $maxlen) {
525 Logger::log('the limit happens before an embedded image', Logger::DEBUG);
526 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
530 $new_body = $new_body . substr($orig_body, 0, $img_start);
531 $textlen += $img_start;
534 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
537 if (($textlen + $img_end) > $maxlen) {
538 if ($textlen < $maxlen) {
539 Logger::log('the limit happens before the end of a non-embedded image', Logger::DEBUG);
540 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
544 $new_body = $new_body . substr($orig_body, 0, $img_end);
545 $textlen += $img_end;
548 $orig_body = substr($orig_body, $img_end);
550 if ($orig_body === false) {
551 // in case the body ends on a closing image tag
555 $img_start = strpos($orig_body, '[img');
556 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
557 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
560 if (($textlen + strlen($orig_body)) > $maxlen) {
561 if ($textlen < $maxlen) {
562 Logger::log('the limit happens after the end of the last image', Logger::DEBUG);
563 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
566 Logger::log('the text size with embedded images extracted did not violate the limit', Logger::DEBUG);
567 $new_body = $new_body . $orig_body;
577 * Processes [attachment] tags
579 * Note: Can produce a [bookmark] tag in the returned string
581 * @brief Processes [attachment] tags
582 * @param string $text
583 * @param bool|int $simplehtml
584 * @param bool $tryoembed
586 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
588 private static function convertAttachment($text, $simplehtml = false, $tryoembed = true)
590 $data = self::getAttachmentData($text);
591 if (empty($data) || empty($data['url'])) {
595 if (isset($data['title'])) {
596 $data['title'] = strip_tags($data['title']);
597 $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
599 $data['title'] = null;
602 if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data['image'])) {
603 $data['preview'] = $data['image'];
609 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
610 $return = OEmbed::getHTML($data['url'], $data['title']);
612 throw new Exception('OEmbed is disabled for this attachment.');
614 } catch (Exception $e) {
615 $data['title'] = ($data['title'] ?? '') ?: $data['url'];
617 if ($simplehtml != 4) {
618 $return = sprintf('<div class="type-%s">', $data['type']);
621 if (!empty($data['title']) && !empty($data['url'])) {
622 if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
623 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
625 if (!empty($data['image'])) {
626 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
627 } elseif (!empty($data['preview'])) {
628 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data['url'], self::proxyUrl($data['preview'], $simplehtml), $data['title']);
630 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
634 if (!empty($data['description']) && $data['description'] != $data['title']) {
635 // Sanitize the HTML by converting it to BBCode
636 $bbcode = HTML::toBBCode($data['description']);
637 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
640 if (!empty($data['url'])) {
641 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
644 if ($simplehtml != 4) {
649 return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
652 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
654 $data = self::getAttachmentData($Text);
659 return $data['text'] . ($data['after'] ?? '');
662 $title = htmlentities($data['title'] ?? '', ENT_QUOTES, 'UTF-8', false);
663 $text = htmlentities($data['text'], ENT_QUOTES, 'UTF-8', false);
664 if ($plaintext || (($title != '') && strstr($text, $title))) {
665 $data['title'] = $data['url'];
666 } elseif (($text != '') && strstr($title, $text)) {
667 $data['text'] = $data['title'];
668 $data['title'] = $data['url'];
671 if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
672 return $data['title'] . $data['after'];
675 // If the link already is included in the post, don't add it again
676 if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
677 return $data['text'] . $data['after'];
680 $text = $data['text'];
682 if (!empty($data['url']) && !empty($data['title'])) {
683 $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
684 } elseif (!empty($data['url'])) {
685 $text .= "\n[url]" . $data['url'] . '[/url]';
688 return $text . "\n" . $data['after'];
692 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
694 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
695 * @param array $match Array with the matching values
696 * @return string reformatted link including HTML codes
698 private static function convertUrlForActivityPubCallback($match)
702 if (isset($match[2]) && ($match[1] != $match[2])) {
706 $parts = parse_url($url);
707 if (!isset($parts['scheme'])) {
711 return self::convertUrlForActivityPub($url);
715 * @brief Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
716 * @param string $url URL that is about to be reformatted
717 * @return string reformatted link including HTML codes
719 private static function convertUrlForActivityPub($url)
721 $html = '<a href="%s" target="_blank">%s</a>';
722 return sprintf($html, $url, self::getStyledURL($url));
726 * Converts an URL in a nicer format (without the scheme and possibly shortened)
727 * @param string $url URL that is about to be reformatted
728 * @return string reformatted link
730 private static function getStyledURL($url)
732 $parts = parse_url($url);
733 $scheme = $parts['scheme'] . '://';
734 $styled_url = str_replace($scheme, '', $url);
736 if (strlen($styled_url) > 30) {
737 $styled_url = substr($styled_url, 0, 30) . "…";
744 * [noparse][i]italic[/i][/noparse] turns into
745 * [noparse][ i ]italic[ /i ][/noparse],
746 * to hide them from parser.
748 private static function escapeNoparseCallback($match)
750 $whole_match = $match[0];
751 $captured = $match[1];
752 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
753 $new_str = str_replace($captured, $spacefied, $whole_match);
758 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
759 * now turns back and the [noparse] tags are trimed
760 * returning [i]italic[/i]
762 private static function unescapeNoparseCallback($match)
764 $captured = $match[1];
765 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
770 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
773 * @param string $text Text to search
774 * @param string $name Tag name
775 * @param int $occurrences Number of first occurrences to skip
776 * @return boolean|array
778 public static function getTagPosition($text, $name, $occurrences = 0)
780 if ($occurrences < 0) {
785 for ($i = 0; $i <= $occurrences; $i++) {
786 if ($start_open !== false) {
787 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
791 if ($start_open === false) {
795 $start_equal = strpos($text, '=', $start_open);
796 $start_close = strpos($text, ']', $start_open);
798 if ($start_close === false) {
804 $end_open = strpos($text, '[/' . $name . ']', $start_close);
806 if ($end_open === false) {
812 'open' => $start_open,
813 'close' => $start_close
817 'close' => $end_open + strlen('[/' . $name . ']')
821 if ($start_equal !== false) {
822 $res['start']['equal'] = $start_equal + 1;
829 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
831 * @param string $pattern Preg pattern string
832 * @param string $replace Preg replace string
833 * @param string $name BBCode tag name
834 * @param string $text Text to search
837 public static function pregReplaceInTag($pattern, $replace, $name, $text)
840 $pos = self::getTagPosition($text, $name, $occurrences);
841 while ($pos !== false && $occurrences++ < 1000) {
842 $start = substr($text, 0, $pos['start']['open']);
843 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
844 $end = substr($text, $pos['end']['close']);
845 if ($end === false) {
849 $subject = preg_replace($pattern, $replace, $subject);
850 $text = $start . $subject . $end;
852 $pos = self::getTagPosition($text, $name, $occurrences);
858 private static function extractImagesFromItemBody($body)
865 $img_start = strpos($orig_body, '[img');
866 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
867 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
868 while (($img_st_close !== false) && ($img_end !== false)) {
869 $img_st_close++; // make it point to AFTER the closing bracket
870 $img_end += $img_start;
872 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
873 // This is an embedded image
874 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
875 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
879 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
882 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
884 if ($orig_body === false) {
885 // in case the body ends on a closing image tag
889 $img_start = strpos($orig_body, '[img');
890 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
891 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
894 $new_body = $new_body . $orig_body;
896 return ['body' => $new_body, 'images' => $saved_image];
899 private static function interpolateSavedImagesIntoItemBody($body, array $images)
904 foreach ($images as $image) {
905 // We're depending on the property of 'foreach' (specified on the PHP website) that
906 // it loops over the array starting from the first element and going sequentially
907 // to the last element
908 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
909 '<img src="' . self::proxyUrl($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
917 * This function converts a [share] block to text according to a provided callback function whose signature is:
919 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
922 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
924 * - $author_contact is a contact record array
925 * - $content is the inner content of the [share] block
926 * - $is_quote_share indicates whether there's any content before the [share] block
927 * - Return value is the string that should replace the [share] block in the provided text
929 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
931 * @param string $text A BBCode string
932 * @param callable $callback
933 * @return string The BBCode string with all [share] blocks replaced
935 public static function convertShare($text, callable $callback)
937 $return = preg_replace_callback(
938 "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
939 function ($match) use ($callback) {
940 $attribute_string = $match[2];
942 foreach (['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
943 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
944 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
947 // We only call this so that a previously unknown contact can be added.
948 // This is important for the function "Model\Contact::getDetailsByURL()".
949 // This function then can fetch an entry from the contact table.
950 $default['url'] = $attributes['profile'];
952 if (!empty($attributes['author'])) {
953 $default['name'] = $attributes['author'];
956 if (!empty($attributes['avatar'])) {
957 $default['photo'] = $attributes['avatar'];
960 Contact::getIdForURL($attributes['profile'], 0, true, $default);
962 $author_contact = Contact::getDetailsByURL($attributes['profile']);
963 $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']);
965 $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author'];
966 $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
967 $attributes['profile'] = ($author_contact['url'] ?? '') ?: $attributes['profile'];
969 if ($attributes['avatar']) {
970 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
973 return $match[1] . $callback($attributes, $author_contact, $match[3], trim($match[1]) != '');
982 * Default [share] tag conversion callback
984 * Note: Can produce a [bookmark] tag in the output
986 * @see BBCode::convertShare()
987 * @param array $attributes [share] block attribute values
988 * @param array $author_contact Contact row of the shared author
989 * @param string $content Inner content of the [share] block
990 * @param boolean $is_quote_share Whether there is content before the [share] block
991 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
993 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
995 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
997 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
999 switch ($simplehtml) {
1001 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' <a href="' . $attributes['profile'] . '">' . $mention . '</a>: </p>' . "\n" . '«' . $content . '»';
1004 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
1007 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
1008 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
1010 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
1012 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
1013 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
1016 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
1018 if (empty($attributes['posted']) && !empty($attributes['link'])) {
1019 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
1025 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
1026 $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
1027 $headline .= ':</b></p>' . "\n";
1029 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
1033 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
1035 case 7: // statusnet/GNU Social
1036 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
1038 case 9: // ActivityPub
1039 $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>';
1040 $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
1043 // Transforms quoted tweets in rich attachments to avoid nested tweets
1044 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) {
1046 $text = ($is_quote_share? '<br />' : '') . OEmbed::getHTML($attributes['link']);
1047 } catch (Exception $e) {
1048 $text = ($is_quote_share? '<br />' : '') . sprintf('[bookmark=%s]%s[/bookmark]', $attributes['link'], $content);
1051 $text = ($is_quote_share? "\n" : '');
1053 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1054 $text .= Renderer::replaceMacros($tpl, [
1055 '$profile' => $attributes['profile'],
1056 '$avatar' => $attributes['avatar'],
1057 '$author' => $attributes['author'],
1058 '$link' => $attributes['link'],
1059 '$posted' => $attributes['posted'],
1060 '$content' => trim($content)
1069 private static function removePictureLinksCallback($match)
1071 $cache_key = 'remove:' . $match[1];
1072 $text = DI::cache()->get($cache_key);
1074 if (is_null($text)) {
1077 $stamp1 = microtime(true);
1079 $ch = @curl_init($match[1]);
1080 @curl_setopt($ch, CURLOPT_NOBODY, true);
1081 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1082 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1084 $curl_info = @curl_getinfo($ch);
1086 DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
1088 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1089 $text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
1091 $text = "[url=" . $match[2] . ']' . $match[2] . "[/url]";
1093 // if its not a picture then look if its a page that contains a picture link
1094 $body = Network::fetchUrl($match[1]);
1096 $doc = new DOMDocument();
1097 @$doc->loadHTML($body);
1098 $xpath = new DOMXPath($doc);
1099 $list = $xpath->query("//meta[@name]");
1100 foreach ($list as $node) {
1103 if ($node->attributes->length) {
1104 foreach ($node->attributes as $attribute) {
1105 $attr[$attribute->name] = $attribute->value;
1109 if (strtolower($attr['name']) == 'twitter:image') {
1110 $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
1114 DI::cache()->set($cache_key, $text);
1120 private static function expandLinksCallback($match)
1122 if (($match[3] == '') || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1123 return ($match[1] . "[url]" . $match[2] . "[/url]");
1125 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1129 private static function cleanPictureLinksCallback($match)
1133 // When the picture link is the own photo path then we can avoid fetching the link
1134 $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
1135 if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
1136 if (!empty($match[3])) {
1137 $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
1139 $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
1144 $cache_key = 'clean:' . $match[1];
1145 $text = DI::cache()->get($cache_key);
1146 if (!is_null($text)) {
1150 // Only fetch the header, not the content
1151 $stamp1 = microtime(true);
1153 $ch = @curl_init($match[1]);
1154 @curl_setopt($ch, CURLOPT_NOBODY, true);
1155 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1156 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1158 $curl_info = @curl_getinfo($ch);
1160 DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
1162 // if its a link to a picture then embed this picture
1163 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1164 $text = '[img]' . $match[1] . '[/img]';
1166 if (!empty($match[3])) {
1167 $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
1169 $text = '[img]' . $match[2] . '[/img]';
1172 // if its not a picture then look if its a page that contains a picture link
1173 $body = Network::fetchUrl($match[1]);
1175 $doc = new DOMDocument();
1176 @$doc->loadHTML($body);
1177 $xpath = new DOMXPath($doc);
1178 $list = $xpath->query("//meta[@name]");
1179 foreach ($list as $node) {
1181 if ($node->attributes->length) {
1182 foreach ($node->attributes as $attribute) {
1183 $attr[$attribute->name] = $attribute->value;
1187 if (strtolower($attr['name']) == "twitter:image") {
1188 if (!empty($match[3])) {
1189 $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
1191 $text = "[img]" . $attr['content'] . "[/img]";
1196 DI::cache()->set($cache_key, $text);
1201 public static function cleanPictureLinks($text)
1203 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1204 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $return);
1209 * @brief Converts a BBCode message to HTML message
1211 * BBcode 2 HTML was written by WAY2WEB.net
1212 * extended to work with Mistpark/Friendica - Mike Macgirvin
1214 * Simple HTML values meaning:
1215 * - 0: Friendica display
1217 * - 2: Used for Windows Phone push, Friendica API
1218 * - 3: Used before converting to Markdown in bb2diaspora.php
1219 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1222 * - 7: Used for dfrn, OStatus
1223 * - 8: Used for Twitter, WP backlink text setting
1226 * @param string $text
1227 * @param bool $try_oembed
1228 * @param int $simple_html
1229 * @param bool $for_plaintext
1231 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1233 public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false)
1238 * preg_match_callback function to replace potential Oembed tags with Oembed content
1240 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1242 * $match[2] = $title or absent
1244 $try_oembed_callback = function ($match)
1247 $title = $match[2] ?? null;
1250 $return = OEmbed::getHTML($url, $title);
1251 } catch (Exception $ex) {
1252 $return = $match[0];
1258 // Extracting code blocks before the whitespace processing and the autolinker
1261 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1262 function ($matches) use (&$codeblocks) {
1263 $return = '#codeblock-' . count($codeblocks) . '#';
1264 if (strpos($matches[2], "\n") !== false) {
1265 $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
1267 $codeblocks[] = '<code>' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
1275 // Hide all [noparse] contained bbtags by spacefying them
1276 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1278 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1279 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1280 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1282 // Remove the abstract element. It is a non visible element.
1283 $text = self::stripAbstract($text);
1285 // Move all spaces out of the tags
1286 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1287 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1289 // Extract the private images which use data urls since preg has issues with
1290 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1291 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1293 $extracted = self::extractImagesFromItemBody($text);
1294 $text = $extracted['body'];
1295 $saved_image = $extracted['images'];
1297 // If we find any event code, turn it into an event.
1298 // After we're finished processing the bbcode we'll
1299 // replace all of the event code with a reformatted version.
1301 $ev = Event::fromBBCode($text);
1303 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1304 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1306 $text = str_replace("<", "<", $text);
1307 $text = str_replace(">", ">", $text);
1309 // remove some newlines before the general conversion
1310 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1311 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1313 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1315 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1318 // Convert new line chars to html <br /> tags
1320 // nlbr seems to be hopelessly messed up
1321 // $Text = nl2br($Text);
1323 // We'll emulate it.
1325 $text = trim($text);
1326 $text = str_replace("\r\n", "\n", $text);
1328 // Remove linefeeds inside of the table elements. See issue #6799
1329 $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
1330 "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
1331 "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
1332 "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "];
1333 $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
1334 "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
1335 "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
1336 "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"];
1339 $text = str_replace($search, $replace, $text);
1340 } while ($oldtext != $text);
1342 // Replace these here only once
1343 $search = ["\n[table]", "[/table]\n"];
1344 $replace = ["[table]", "[/table]"];
1345 $text = str_replace($search, $replace, $text);
1347 // removing multiplicated newlines
1348 if (Config::get('system', 'remove_multiplicated_lines')) {
1349 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1350 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1351 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1352 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1355 $text = str_replace($search, $replace, $text);
1356 } while ($oldtext != $text);
1359 /// @todo Have a closer look at the different html modes
1360 // Handle attached links or videos
1361 if (in_array($simple_html, [9])) {
1362 $text = self::removeAttachment($text);
1363 } elseif (!in_array($simple_html, [0, 4])) {
1364 $text = self::removeAttachment($text, true);
1366 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1369 // leave open the posibility of [map=something]
1370 // this is replaced in Item::prepareBody() which has knowledge of the item location
1371 if (strpos($text, '[/map]') !== false) {
1372 $text = preg_replace_callback(
1373 "/\[map\](.*?)\[\/map\]/ism",
1374 function ($match) use ($simple_html) {
1375 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1381 if (strpos($text, '[map=') !== false) {
1382 $text = preg_replace_callback(
1383 "/\[map=(.*?)\]/ism",
1384 function ($match) use ($simple_html) {
1385 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1391 if (strpos($text, '[map]') !== false) {
1392 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1395 // Check for headers
1396 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1397 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1398 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1399 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1400 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1401 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1403 // Check for paragraph
1404 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1406 // Check for bold text
1407 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1409 // Check for Italics text
1410 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1412 // Check for Underline text
1413 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1415 // Check for strike-through text
1416 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1418 // Check for over-line text
1419 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1421 // Check for colored text
1422 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1424 // Check for sized text
1425 // [size=50] --> font-size: 50px (with the unit).
1426 if ($simple_html != 3) {
1427 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1428 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1430 // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
1431 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text);
1435 // Check for centered text
1436 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1438 // Check for list text
1439 $text = str_replace("[*]", "<li>", $text);
1441 // Check for style sheet commands
1442 $text = preg_replace_callback(
1443 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1445 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1450 // Check for CSS classes
1451 $text = preg_replace_callback(
1452 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1454 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1459 // handle nested lists
1462 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1463 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1464 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1465 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1466 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1467 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1468 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1469 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1470 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1471 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1472 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1473 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1474 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1475 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1478 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1479 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1480 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1481 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1483 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1484 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1486 $text = str_replace('[hr]', '<hr />', $text);
1488 if (!$for_plaintext) {
1491 // Escaping BBCodes susceptible to contain rogue URL we don'' want the autolinker to catch
1492 $text = preg_replace_callback('#\[(url|img|audio|video|youtube|vimeo|share|attachment|iframe|bookmark).+?\[/\1\]#ism',
1493 function ($matches) use (&$escaped) {
1494 $return = '{escaped-' . count($escaped) . '}';
1495 $escaped[] = $matches[0];
1502 // Autolinker for isolated URLs
1503 $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1505 // Restoring escaped blocks
1506 $text = preg_replace_callback('/{escaped-([0-9]+)}/iU',
1507 function ($matches) use ($escaped) {
1508 return $escaped[intval($matches[1])] ?? $matches[0];
1514 // This is actually executed in Item::prepareBody()
1516 $nosmile = strpos($text, '[nosmile]') !== false;
1517 $text = str_replace('[nosmile]', '', $text);
1519 // Check for font change text
1520 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1522 // Declare the format for [spoiler] layout
1523 $SpoilerLayout = '<details class="spoiler"><summary>' . L10n::t('Click to open/close') . '</summary>$1</details>';
1525 // Check for [spoiler] text
1526 // handle nested quotes
1528 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1529 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", $SpoilerLayout, $text);
1532 // Check for [spoiler=Title] text
1534 // handle nested quotes
1536 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1537 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1538 '<details class="spoiler"><summary>$1</summary>$2</details>',
1542 // Declare the format for [quote] layout
1543 $QuoteLayout = '<blockquote>$1</blockquote>';
1545 // Check for [quote] text
1546 // handle nested quotes
1548 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1549 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1552 // Check for [quote=Author] text
1554 $t_wrote = L10n::t('$1 wrote:');
1556 // handle nested quotes
1558 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1559 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1560 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1565 // [img=widthxheight]image source[/img]
1566 $text = preg_replace_callback(
1567 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1568 function ($matches) use ($simple_html) {
1569 if (strpos($matches[3], "data:image/") === 0) {
1573 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1574 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1579 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1580 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1582 $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/ism",
1583 function ($matches) use ($simple_html) {
1584 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1585 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1586 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '" title="' . $matches[2] . '">';
1591 // [img]pathtoimage[/img]
1592 $text = preg_replace_callback(
1593 "/\[img\](.*?)\[\/img\]/ism",
1594 function ($matches) use ($simple_html) {
1595 if (strpos($matches[1], "data:image/") === 0) {
1599 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1600 return "[img]" . $matches[1] . "[/img]";
1605 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1606 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1608 $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br/><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . L10n::t('Encrypted content') . '" /><br />', $text);
1609 $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $text);
1610 //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $Text);
1612 // Simplify "video" element
1613 $text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
1617 $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);
1618 $text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3).*?)\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
1620 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1621 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1623 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1624 '<a href="$1" target="_blank">$1</a>', $text);
1625 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1626 '<a href="$1" target="_blank">$1</a>', $text);
1629 // html5 video and audio
1633 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1635 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1638 // Youtube extensions
1640 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1641 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1642 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1645 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1646 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1647 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1650 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '<iframe width="' . $a->videowidth . '" height="' . $a->videoheight . '" src="https://www.youtube.com/embed/$1" frameborder="0" ></iframe>', $text);
1652 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1653 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1657 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1658 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1661 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1662 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1665 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '<iframe width="' . $a->videowidth . '" height="' . $a->videoheight . '" src="https://player.vimeo.com/video/$1" frameborder="0" ></iframe>', $text);
1667 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1668 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1672 $text = OEmbed::BBCode2HTML($text);
1674 // Avoid triple linefeeds through oembed
1675 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1677 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1678 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1679 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1680 // start which is always required). Allow desc with a missing summary for compatibility.
1682 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1683 $sub = Event::getHTML($ev, $simple_html);
1685 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1686 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1687 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1688 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1689 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1690 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1691 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1694 // Replace non graphical smilies for external posts
1695 if (!$nosmile && !$for_plaintext) {
1696 $text = Smilies::replace($text);
1699 if (!$for_plaintext) {
1700 if (in_array($simple_html, [7, 9])) {
1701 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1702 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1705 $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1706 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1709 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1711 // Remove all hashtag addresses
1712 if ($simple_html && !in_array($simple_html, [3, 7, 9])) {
1713 $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1714 } elseif ($simple_html == 3) {
1715 // The ! is converted to @ since Diaspora only understands the @
1716 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1717 '@<a href="$2">$3</a>',
1719 } elseif (in_array($simple_html, [7, 9])) {
1720 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1721 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1723 } elseif (!$simple_html) {
1724 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1725 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1729 // Bookmarks in red - will be converted to bookmarks in friendica
1730 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1731 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1732 $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1733 "[bookmark=$1]$2[/bookmark]", $text);
1735 if (in_array($simple_html, [2, 6, 7, 8])) {
1736 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1737 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1738 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1741 if ($simple_html == 5) {
1742 $text = preg_replace("/[^#@!]\[url\=(.*?)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1745 // Perform URL Search
1747 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1750 if ($simple_html == 5) {
1751 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1753 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1756 // Handle Diaspora posts
1757 $text = preg_replace_callback(
1758 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1760 return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1764 $text = preg_replace_callback(
1765 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1767 return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1771 // Server independent link to posts and comments
1772 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1773 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1774 $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
1778 * - #[url=<anything>]<term>[/url]
1779 * - [url=<anything>]#<term>[/url]
1781 $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) {
1783 . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1784 . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
1785 . XML::escape($matches[1])
1789 // We need no target="_blank" for local links
1790 // convert links start with DI::baseUrl() as local link without the target="_blank" attribute
1791 $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
1792 $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1793 $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1795 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1796 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1798 // Red compatibility, though the link can't be authenticated on Friendica
1799 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1802 // we may need to restrict this further if it picks up too many strays
1803 // link acct:user@host to a webfinger profile redirector
1805 $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);
1807 // Perform MAIL Search
1808 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1809 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1811 // Unhide all [noparse] contained bbtags unspacefying them
1812 // and triming the [noparse] tag.
1814 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1815 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1816 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1818 /// @todo What is the meaning of these lines?
1819 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1820 $text = preg_replace('/\&\#039\;/', '\'', $text);
1822 // Currently deactivated, it made problems with " inside of alt texts.
1823 //$text = preg_replace('/\"\;/', '"', $text);
1825 // fix any escaped ampersands that may have been converted into links
1826 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1828 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1829 $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:'];
1831 array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
1833 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1834 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
1836 // sanitize href attributes (only whitelisted protocols URLs)
1837 // default value for backward compatibility
1838 $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', []);
1840 // Always allowed protocol even if config isn't set or not including it
1841 $allowed_link_protocols[] = '//';
1842 $allowed_link_protocols[] = 'http://';
1843 $allowed_link_protocols[] = 'https://';
1844 $allowed_link_protocols[] = 'redir/';
1846 array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
1848 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1849 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
1852 $text = self::convertShare(
1854 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1855 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1860 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1863 // Restore code blocks
1864 $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
1865 function ($matches) use ($codeblocks) {
1866 $return = $matches[0];
1867 if (isset($codeblocks[intval($matches[1])])) {
1868 $return = $codeblocks[$matches[1]];
1875 // Clean up the HTML by loading and saving the HTML with the DOM.
1876 // Bad structured html can break a whole page.
1877 // For performance reasons do it only with activated item cache or at export.
1878 if (!$try_oembed || (get_itemcachepath() != '')) {
1879 $doc = new DOMDocument();
1880 $doc->preserveWhiteSpace = false;
1882 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1884 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1885 $encoding = '<?xml encoding="UTF-8">';
1886 @$doc->loadHTML($encoding . $doctype . '<html><body>' . $text . '</body></html>');
1887 $doc->encoding = 'UTF-8';
1888 $text = $doc->saveHTML();
1889 $text = str_replace(['<html><body>', '</body></html>', $doctype, $encoding], ['', '', '', ''], $text);
1891 $text = str_replace('<br></li>', '</li>', $text);
1893 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1896 // Clean up some useless linebreaks in lists
1897 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1898 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1899 //$Text = str_replace('</li><br />', '</li>', $Text);
1900 //$Text = str_replace('<br /><li>', '<li>', $Text);
1901 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1903 Hook::callAll('bbcode', $text);
1909 * @brief Strips the "abstract" tag from the provided text
1911 * @param string $text The text with BBCode
1912 * @return string The same text - but without "abstract" element
1914 public static function stripAbstract($text)
1916 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1917 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1923 * @brief Returns the value of the "abstract" element
1925 * @param string $text The text that maybe contains the element
1926 * @param string $addon The addon for which the abstract is meant for
1927 * @return string The abstract
1929 public static function getAbstract($text, $addon = '')
1933 $addon = strtolower($addon);
1935 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1936 foreach ($results AS $result) {
1937 $abstracts[strtolower($result[1])] = $result[2];
1941 if (isset($abstracts[$addon])) {
1942 $abstract = $abstracts[$addon];
1945 if ($abstract == '' && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1946 $abstract = $result[1];
1953 * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
1955 * @param array $match Matching values for the callback
1956 * [1] = Mention type (! or @)
1959 * @return string Replaced mention
1960 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1961 * @throws \ImagickException
1963 private static function bbCodeMention2DiasporaCallback($match)
1965 $contact = Contact::getDetailsByURL($match[3]);
1967 if (empty($contact['addr'])) {
1968 $contact = Probe::uri($match[3]);
1971 if (empty($contact['addr'])) {
1975 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
1980 * @brief Converts a BBCode text into Markdown
1982 * This function converts a BBCode item body to be sent to Markdown-enabled
1983 * systems like Diaspora and Libertree
1985 * @param string $text
1986 * @param bool $for_diaspora Diaspora requires more changes than Libertree
1988 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1990 public static function toMarkdown($text, $for_diaspora = true)
1994 $original_text = $text;
1996 // Since Diaspora is creating a summary for links, this function removes them before posting
1997 if ($for_diaspora) {
1998 $text = self::removeShareInformation($text);
2002 * Transform #tags, strip off the [url] and replace spaces with underscore
2004 $url_search_string = "^\[\]";
2005 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
2006 function ($matches) {
2007 return '#' . str_replace(' ', '_', $matches[2]);
2012 // Converting images with size parameters to simple images. Markdown doesn't know it.
2013 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2015 // Convert it to HTML - don't try oembed
2016 if ($for_diaspora) {
2017 $text = self::convert($text, false, 3);
2019 // Add all tags that maybe were removed
2020 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2022 foreach ($tags[2] as $tag) {
2023 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2024 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2025 $tagline .= '#' . $tag . ' ';
2028 $text = $text . " " . $tagline;
2031 $text = self::convert($text, false, 4);
2034 // If a link is followed by a quote then there should be a newline before it
2035 // Maybe we should make this newline at every time before a quote.
2036 $text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
2038 $stamp1 = microtime(true);
2040 // Now convert HTML to Markdown
2041 $text = HTML::toMarkdown($text);
2043 DI::profiler()->saveTimestamp($stamp1, "parser", System::callstack());
2045 // Libertree has a problem with escaped hashtags.
2046 $text = str_replace(['\#'], ['#'], $text);
2048 // Remove any leading or trailing whitespace, as this will mess up
2049 // the Diaspora signature verification and cause the item to disappear
2050 $text = trim($text);
2052 if ($for_diaspora) {
2053 $url_search_string = "^\[\]";
2054 $text = preg_replace_callback(
2055 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
2056 ['self', 'bbCodeMention2DiasporaCallback'],
2061 Hook::callAll('bb2diaspora', $text);
2067 * @brief Pull out all #hashtags and @person tags from $string.
2069 * We also get @person@domain.com - which would make
2070 * the regex quite complicated as tags can also
2071 * end a sentence. So we'll run through our results
2072 * and strip the period from any tags which end with one.
2073 * Returns array of tags found, or empty array.
2075 * @param string $string Post content
2077 * @return array List of tag and person names
2079 public static function getTags($string)
2083 // Convert hashtag links to hashtags
2084 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2', $string);
2086 // ignore anything in a code block
2087 $string = preg_replace('/\[code.*?\].*?\[\/code\]/sm', '', $string);
2089 // Force line feeds at bbtags
2090 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
2092 // ignore anything in a bbtag
2093 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
2095 // Match full names against @tags including the space between first and last
2096 // We will look these up afterward to see if they are full names or not recognisable.
2098 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
2099 foreach ($matches[1] as $match) {
2100 if (strstr($match, ']')) {
2101 // we might be inside a bbcode color tag - leave it alone
2105 if (substr($match, -1, 1) === '.') {
2106 $ret[] = substr($match, 0, -1);
2113 // Otherwise pull out single word tags. These can be @nickname, @first_last
2116 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
2117 foreach ($matches[1] as $match) {
2118 if (strstr($match, ']')) {
2119 // we might be inside a bbcode color tag - leave it alone
2122 if (substr($match, -1, 1) === '.') {
2123 $match = substr($match,0,-1);
2125 // ignore strictly numeric tags like #1
2126 if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
2129 // try not to catch url fragments
2130 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {