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\Hook;
14 use Friendica\Core\Logger;
15 use Friendica\Core\Protocol;
16 use Friendica\Core\Renderer;
17 use Friendica\Core\System;
19 use Friendica\Model\Contact;
20 use Friendica\Model\Event;
21 use Friendica\Model\Photo;
22 use Friendica\Network\Probe;
23 use Friendica\Object\Image;
24 use Friendica\Protocol\Activity;
25 use Friendica\Util\Images;
26 use Friendica\Util\Map;
27 use Friendica\Util\Network;
28 use Friendica\Util\ParseUrl;
29 use Friendica\Util\Proxy as ProxyUtils;
30 use Friendica\Util\Strings;
31 use Friendica\Util\XML;
36 * Fetches attachment data that were generated the old way
38 * @param string $body Message body
40 * 'type' -> Message type ('link', 'video', 'photo')
41 * 'text' -> Text before the shared message
42 * 'after' -> Text after the shared message
43 * 'image' -> Preview image of the message
44 * 'url' -> Url to the attached message
45 * 'title' -> Title of the attachment
46 * 'description' -> Description of the attachment
47 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
49 private static function getOldAttachmentData($body)
53 // Simplify image codes
54 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
56 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
57 foreach ($attached as $data) {
58 if (!in_array($data[1], ['type-link', 'type-video', 'type-photo'])) {
62 $post['type'] = substr($data[1], 5);
64 $pos = strpos($body, $data[0]);
66 $post['text'] = trim(substr($body, 0, $pos));
67 $post['after'] = trim(substr($body, $pos + strlen($data[0])));
69 $post['text'] = trim(str_replace($data[0], '', $body));
73 $attacheddata = $data[2];
75 if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
77 $picturedata = Images::getInfoFromURLCached($matches[1]);
80 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
81 $post['image'] = $matches[1];
83 $post['preview'] = $matches[1];
88 if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
89 $post['url'] = $matches[1];
90 $post['title'] = $matches[2];
92 if (!empty($post['url']) && (in_array($post['type'], ['link', 'video']))
93 && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
94 $post['url'] = $matches[1];
97 // Search for description
98 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
99 $post['description'] = $matches[1];
107 * Fetches attachment data that were generated with the "attachment" element
109 * @param string $body Message body
111 * 'type' -> Message type ('link', 'video', 'photo')
112 * 'text' -> Text before the shared message
113 * 'after' -> Text after the shared message
114 * 'image' -> Preview image of the message
115 * 'url' -> Url to the attached message
116 * 'title' -> Title of the attachment
117 * 'description' -> Description of the attachment
118 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
120 public static function getAttachmentData($body)
132 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
133 return self::getOldAttachmentData($body);
136 $attributes = $match[2];
138 $data['text'] = trim($match[1]);
141 preg_match("/type='(.*?)'/ism", $attributes, $matches);
142 if (!empty($matches[1])) {
143 $type = strtolower($matches[1]);
146 preg_match('/type="(.*?)"/ism', $attributes, $matches);
147 if (!empty($matches[1])) {
148 $type = strtolower($matches[1]);
155 if (!in_array($type, ['link', 'audio', 'photo', 'video'])) {
160 $data['type'] = $type;
164 preg_match("/url='(.*?)'/ism", $attributes, $matches);
165 if (!empty($matches[1])) {
169 preg_match('/url="(.*?)"/ism', $attributes, $matches);
170 if (!empty($matches[1])) {
175 $data['url'] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
179 preg_match("/title='(.*?)'/ism", $attributes, $matches);
180 if (!empty($matches[1])) {
181 $title = $matches[1];
184 preg_match('/title="(.*?)"/ism', $attributes, $matches);
185 if (!empty($matches[1])) {
186 $title = $matches[1];
190 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
191 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
192 $title = str_replace(['[', ']'], ['[', ']'], $title);
193 $data['title'] = $title;
197 preg_match("/image='(.*?)'/ism", $attributes, $matches);
198 if (!empty($matches[1])) {
199 $image = $matches[1];
202 preg_match('/image="(.*?)"/ism', $attributes, $matches);
203 if (!empty($matches[1])) {
204 $image = $matches[1];
208 $data['image'] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
212 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
213 if (!empty($matches[1])) {
214 $preview = $matches[1];
217 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
218 if (!empty($matches[1])) {
219 $preview = $matches[1];
222 if ($preview != '') {
223 $data['preview'] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
226 $data['description'] = trim($match[3]);
228 $data['after'] = trim($match[4]);
233 public static function getAttachedData($body, $item = [])
237 - type: link, video, photo
245 $has_title = !empty($item['title']);
246 $plink = $item['plink'] ?? '';
247 $post = self::getAttachmentData($body);
249 // Get all linked images with alternative image description
250 if (preg_match_all("/\[img=([^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
251 foreach ($pictures as $picture) {
252 if (Photo::isLocal($picture[1])) {
253 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2]];
256 if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
257 $post['image_description'] = $post['images'][0]['description'];
261 if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
262 foreach ($pictures as $picture) {
263 if (Photo::isLocal($picture[1])) {
264 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => ''];
269 // if nothing is found, it maybe having an image.
270 if (!isset($post['type'])) {
271 // Simplify image codes
272 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
273 $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
274 $post['text'] = $body;
276 if (preg_match_all("(\[url=(.*?)\]\s*\[img\](.*?)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
277 if ((count($pictures) == 1) && !$has_title) {
278 if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
279 // Replace the preview picture with the real picture
280 $url = str_replace('-1.', '-0.', $pictures[0][2]);
281 $data = ['url' => $url, 'type' => 'photo'];
283 // Checking, if the link goes to a picture
284 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
288 // Sometimes photo posts to the own album are not detected at the start.
289 // So we seem to cannot use the cache for these cases. That's strange.
290 if (($data['type'] != 'photo') && strstr($pictures[0][1], "/photos/")) {
291 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
294 if ($data['type'] == 'photo') {
295 $post['type'] = 'photo';
296 if (isset($data['images'][0])) {
297 $post['image'] = $data['images'][0]['src'];
298 $post['url'] = $data['url'];
300 $post['image'] = $data['url'];
303 $post['preview'] = $pictures[0][2];
304 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
306 $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
307 if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
308 $post['type'] = 'photo';
309 $post['image'] = $pictures[0][1];
310 $post['preview'] = $pictures[0][2];
311 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
314 } elseif (count($pictures) > 0) {
315 $post['type'] = 'link';
316 $post['url'] = $plink;
317 $post['image'] = $pictures[0][2];
318 $post['text'] = $body;
320 foreach ($pictures as $picture) {
321 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
324 } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
325 if ((count($pictures) == 1) && !$has_title) {
326 $post['type'] = 'photo';
327 $post['image'] = $pictures[0][1];
328 $post['text'] = str_replace($pictures[0][0], '', $body);
329 } elseif (count($pictures) > 0) {
330 $post['type'] = 'link';
331 $post['url'] = $plink;
332 $post['image'] = $pictures[0][1];
333 $post['text'] = $body;
335 foreach ($pictures as $picture) {
336 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
341 // Test for the external links
342 preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER);
343 preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $links2, PREG_SET_ORDER);
345 $links = array_merge($links1, $links2);
347 // If there is only a single one, then use it.
348 // This should cover link posts via API.
349 if ((count($links) == 1) && !isset($post['preview']) && !$has_title) {
350 $post['type'] = 'link';
351 $post['url'] = $links[0][1];
354 // Simplify "video" element
355 $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
357 // Now count the number of external media links
358 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
359 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
360 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER);
361 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $links4, PREG_SET_ORDER);
363 // Add them to the other external links
364 $links = array_merge($links, $links1, $links2, $links3, $links4);
366 // Are there more than one?
367 if (count($links) > 1) {
368 // The post will be the type "text", which means a blog post
369 unset($post['type']);
370 $post['url'] = $plink;
373 if (!isset($post['type'])) {
374 $post['type'] = "text";
375 $post['text'] = trim($body);
377 } elseif (isset($post['url']) && ($post['type'] == 'video')) {
378 $data = ParseUrl::getSiteinfoCached($post['url'], true);
380 if (isset($data['images'][0])) {
381 $post['image'] = $data['images'][0]['src'];
389 * Remove [attachment] BBCode and replaces it with a regular [url]
391 * @param string $body
392 * @param boolean $no_link_desc No link description
394 * @return string with replaced body
396 public static function removeAttachment($body, $no_link_desc = false)
398 return preg_replace_callback("/\s*\[attachment (.*)\](.*?)\[\/attachment\]\s*/ism",
399 function ($match) use ($no_link_desc) {
400 $attach_data = self::getAttachmentData($match[0]);
401 if (empty($attach_data['url'])) {
403 } elseif (empty($attach_data['title']) || $no_link_desc) {
404 return "\n[url]" . $attach_data['url'] . "[/url]\n";
406 return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
412 * Converts a BBCode text into plaintext
415 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
419 public static function toPlaintext($text, $keep_urls = true)
421 $naked_text = HTML::toPlaintext(BBCode::convert($text, false, 0, true), 0, !$keep_urls);
426 private static function proxyUrl($image, $simplehtml = false)
428 // Only send proxied pictures to API and for internal display
429 if (in_array($simplehtml, [false, 2])) {
430 return ProxyUtils::proxifyUrl($image);
436 public static function scaleExternalImages($srctext)
441 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
443 foreach ($matches as $mtch) {
444 Logger::log('scale_external_image: ' . $mtch[1]);
446 $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
447 if (stristr($mtch[1], $hostname)) {
451 $i = Network::fetchUrl($mtch[1]);
456 // guess mimetype from headers or filename
457 $type = Images::guessType($mtch[1], true);
460 $Image = new Image($i, $type);
461 if ($Image->isValid()) {
462 $orig_width = $Image->getWidth();
463 $orig_height = $Image->getHeight();
465 if ($orig_width > 640 || $orig_height > 640) {
466 $Image->scaleDown(640);
467 $new_width = $Image->getWidth();
468 $new_height = $Image->getHeight();
469 Logger::log('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], Logger::DEBUG);
472 '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
476 Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG);
487 * Truncates imported message body string length to max_import_size
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 * @param string $body
494 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
496 public static function limitBodySize($body)
498 $maxlen = DI::config()->get('config', 'max_import_size', 0);
500 // If the length of the body, including the embedded images, is smaller
501 // than the maximum, then don't waste time looking for the images
502 if ($maxlen && (strlen($body) > $maxlen)) {
504 Logger::log('the total body length exceeds the limit', Logger::DEBUG);
510 $img_start = strpos($orig_body, '[img');
511 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
512 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
513 while (($img_st_close !== false) && ($img_end !== false)) {
515 $img_st_close++; // make it point to AFTER the closing bracket
516 $img_end += $img_start;
517 $img_end += strlen('[/img]');
519 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
520 // This is an embedded image
522 if (($textlen + $img_start) > $maxlen) {
523 if ($textlen < $maxlen) {
524 Logger::log('the limit happens before an embedded image', Logger::DEBUG);
525 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
529 $new_body = $new_body . substr($orig_body, 0, $img_start);
530 $textlen += $img_start;
533 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
536 if (($textlen + $img_end) > $maxlen) {
537 if ($textlen < $maxlen) {
538 Logger::log('the limit happens before the end of a non-embedded image', Logger::DEBUG);
539 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
543 $new_body = $new_body . substr($orig_body, 0, $img_end);
544 $textlen += $img_end;
547 $orig_body = substr($orig_body, $img_end);
549 if ($orig_body === false) {
550 // in case the body ends on a closing image tag
554 $img_start = strpos($orig_body, '[img');
555 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
556 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
559 if (($textlen + strlen($orig_body)) > $maxlen) {
560 if ($textlen < $maxlen) {
561 Logger::log('the limit happens after the end of the last image', Logger::DEBUG);
562 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
565 Logger::log('the text size with embedded images extracted did not violate the limit', Logger::DEBUG);
566 $new_body = $new_body . $orig_body;
576 * Processes [attachment] tags
578 * Note: Can produce a [bookmark] tag in the returned string
580 * @param string $text
581 * @param bool|int $simplehtml
582 * @param bool $tryoembed
584 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
586 private static function convertAttachment($text, $simplehtml = false, $tryoembed = true)
588 $data = self::getAttachmentData($text);
589 if (empty($data) || empty($data['url'])) {
593 if (isset($data['title'])) {
594 $data['title'] = strip_tags($data['title']);
595 $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
597 $data['title'] = null;
600 if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
601 $data['preview'] = $data['image'];
607 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
608 $return = OEmbed::getHTML($data['url'], $data['title']);
610 throw new Exception('OEmbed is disabled for this attachment.');
612 } catch (Exception $e) {
613 $data['title'] = ($data['title'] ?? '') ?: $data['url'];
615 if ($simplehtml != 4) {
616 $return = sprintf('<div class="type-%s">', $data['type']);
619 if (!empty($data['title']) && !empty($data['url'])) {
620 if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
621 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
623 if (!empty($data['image'])) {
624 $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), $data['title']);
625 } elseif (!empty($data['preview'])) {
626 $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']);
628 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
632 if (!empty($data['description']) && $data['description'] != $data['title']) {
633 // Sanitize the HTML by converting it to BBCode
634 $bbcode = HTML::toBBCode($data['description']);
635 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
638 if (!empty($data['url'])) {
639 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
642 if ($simplehtml != 4) {
647 return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
650 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
652 $data = self::getAttachmentData($Text);
657 return $data['text'] . ($data['after'] ?? '');
660 $title = htmlentities($data['title'] ?? '', ENT_QUOTES, 'UTF-8', false);
661 $text = htmlentities($data['text'], ENT_QUOTES, 'UTF-8', false);
662 if ($plaintext || (($title != '') && strstr($text, $title))) {
663 $data['title'] = $data['url'];
664 } elseif (($text != '') && strstr($title, $text)) {
665 $data['text'] = $data['title'];
666 $data['title'] = $data['url'];
669 if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
670 return $data['title'] . $data['after'];
673 // If the link already is included in the post, don't add it again
674 if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
675 return $data['text'] . $data['after'];
678 $text = $data['text'];
680 if (!empty($data['url']) && !empty($data['title'])) {
681 $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
682 } elseif (!empty($data['url'])) {
683 $text .= "\n[url]" . $data['url'] . '[/url]';
686 return $text . "\n" . $data['after'];
690 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
692 * @param array $match Array with the matching values
693 * @return string reformatted link including HTML codes
695 private static function convertUrlForActivityPubCallback($match)
699 if (isset($match[2]) && ($match[1] != $match[2])) {
703 $parts = parse_url($url);
704 if (!isset($parts['scheme'])) {
708 return self::convertUrlForActivityPub($url);
712 * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
714 * @param string $url URL that is about to be reformatted
715 * @return string reformatted link including HTML codes
717 private static function convertUrlForActivityPub($url)
719 $html = '<a href="%s" target="_blank">%s</a>';
720 return sprintf($html, $url, self::getStyledURL($url));
724 * Converts an URL in a nicer format (without the scheme and possibly shortened)
726 * @param string $url URL that is about to be reformatted
727 * @return string reformatted link
729 private static function getStyledURL($url)
731 $parts = parse_url($url);
732 $scheme = $parts['scheme'] . '://';
733 $styled_url = str_replace($scheme, '', $url);
735 if (strlen($styled_url) > 30) {
736 $styled_url = substr($styled_url, 0, 30) . "…";
743 * [noparse][i]italic[/i][/noparse] turns into
744 * [noparse][ i ]italic[ /i ][/noparse],
745 * to hide them from parser.
747 private static function escapeNoparseCallback($match)
749 $whole_match = $match[0];
750 $captured = $match[1];
751 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
752 $new_str = str_replace($captured, $spacefied, $whole_match);
757 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
758 * now turns back and the [noparse] tags are trimed
759 * returning [i]italic[/i]
761 private static function unescapeNoparseCallback($match)
763 $captured = $match[1];
764 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
769 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
772 * @param string $text Text to search
773 * @param string $name Tag name
774 * @param int $occurrences Number of first occurrences to skip
775 * @return boolean|array
777 public static function getTagPosition($text, $name, $occurrences = 0)
779 if ($occurrences < 0) {
784 for ($i = 0; $i <= $occurrences; $i++) {
785 if ($start_open !== false) {
786 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
790 if ($start_open === false) {
794 $start_equal = strpos($text, '=', $start_open);
795 $start_close = strpos($text, ']', $start_open);
797 if ($start_close === false) {
803 $end_open = strpos($text, '[/' . $name . ']', $start_close);
805 if ($end_open === false) {
811 'open' => $start_open,
812 'close' => $start_close
816 'close' => $end_open + strlen('[/' . $name . ']')
820 if ($start_equal !== false) {
821 $res['start']['equal'] = $start_equal + 1;
828 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
830 * @param string $pattern Preg pattern string
831 * @param string $replace Preg replace string
832 * @param string $name BBCode tag name
833 * @param string $text Text to search
836 public static function pregReplaceInTag($pattern, $replace, $name, $text)
839 $pos = self::getTagPosition($text, $name, $occurrences);
840 while ($pos !== false && $occurrences++ < 1000) {
841 $start = substr($text, 0, $pos['start']['open']);
842 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
843 $end = substr($text, $pos['end']['close']);
844 if ($end === false) {
848 $subject = preg_replace($pattern, $replace, $subject);
849 $text = $start . $subject . $end;
851 $pos = self::getTagPosition($text, $name, $occurrences);
857 private static function extractImagesFromItemBody($body)
864 $img_start = strpos($orig_body, '[img');
865 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
866 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
867 while (($img_st_close !== false) && ($img_end !== false)) {
868 $img_st_close++; // make it point to AFTER the closing bracket
869 $img_end += $img_start;
871 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
872 // This is an embedded image
873 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
874 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
878 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
881 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
883 if ($orig_body === false) {
884 // in case the body ends on a closing image tag
888 $img_start = strpos($orig_body, '[img');
889 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
890 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
893 $new_body = $new_body . $orig_body;
895 return ['body' => $new_body, 'images' => $saved_image];
898 private static function interpolateSavedImagesIntoItemBody($body, array $images)
903 foreach ($images as $image) {
904 // We're depending on the property of 'foreach' (specified on the PHP website) that
905 // it loops over the array starting from the first element and going sequentially
906 // to the last element
907 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
908 '<img src="' . self::proxyUrl($image) . '" alt="' . DI::l10n()->t('Image/photo') . '" />', $newbody);
916 * This function converts a [share] block to text according to a provided callback function whose signature is:
918 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
921 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
923 * - $author_contact is a contact record array
924 * - $content is the inner content of the [share] block
925 * - $is_quote_share indicates whether there's any content before the [share] block
926 * - Return value is the string that should replace the [share] block in the provided text
928 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
930 * @param string $text A BBCode string
931 * @param callable $callback
932 * @return string The BBCode string with all [share] blocks replaced
934 public static function convertShare($text, callable $callback)
936 $return = preg_replace_callback(
937 "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
938 function ($match) use ($callback) {
939 $attribute_string = $match[2];
941 foreach (['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
942 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
943 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
946 // We only call this so that a previously unknown contact can be added.
947 // This is important for the function "Model\Contact::getDetailsByURL()".
948 // This function then can fetch an entry from the contact table.
949 $default['url'] = $attributes['profile'];
951 if (!empty($attributes['author'])) {
952 $default['name'] = $attributes['author'];
955 if (!empty($attributes['avatar'])) {
956 $default['photo'] = $attributes['avatar'];
959 Contact::getIdForURL($attributes['profile'], 0, true, $default);
961 $author_contact = Contact::getDetailsByURL($attributes['profile']);
962 $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']);
964 $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author'];
965 $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
966 $attributes['profile'] = ($author_contact['url'] ?? '') ?: $attributes['profile'];
968 if ($attributes['avatar']) {
969 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
972 return $match[1] . $callback($attributes, $author_contact, $match[3], trim($match[1]) != '');
981 * Default [share] tag conversion callback
983 * Note: Can produce a [bookmark] tag in the output
985 * @see BBCode::convertShare()
986 * @param array $attributes [share] block attribute values
987 * @param array $author_contact Contact row of the shared author
988 * @param string $content Inner content of the [share] block
989 * @param boolean $is_quote_share Whether there is content before the [share] block
990 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
992 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
994 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
996 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
998 switch ($simplehtml) {
1000 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' <a href="' . $attributes['profile'] . '">' . $mention . '</a>: </p>' . "\n" . '«' . $content . '»';
1003 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
1006 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
1007 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
1009 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
1011 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
1012 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
1015 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
1017 if (empty($attributes['posted']) && !empty($attributes['link'])) {
1018 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
1024 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
1025 $headline .= DI::l10n()->t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
1026 $headline .= ':</b></p>' . "\n";
1028 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
1032 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
1034 case 7: // statusnet/GNU Social
1035 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
1037 case 9: // ActivityPub
1038 $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>';
1039 $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
1042 // Transforms quoted tweets in rich attachments to avoid nested tweets
1043 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) {
1045 $text = ($is_quote_share? '<br />' : '') . OEmbed::getHTML($attributes['link']);
1046 } catch (Exception $e) {
1047 $text = ($is_quote_share? '<br />' : '') . sprintf('[bookmark=%s]%s[/bookmark]', $attributes['link'], $content);
1050 $text = ($is_quote_share? "\n" : '');
1052 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1053 $text .= Renderer::replaceMacros($tpl, [
1054 '$profile' => $attributes['profile'],
1055 '$avatar' => $attributes['avatar'],
1056 '$author' => $attributes['author'],
1057 '$link' => $attributes['link'],
1058 '$posted' => $attributes['posted'],
1059 '$content' => trim($content)
1068 private static function removePictureLinksCallback($match)
1070 $cache_key = 'remove:' . $match[1];
1071 $text = DI::cache()->get($cache_key);
1073 if (is_null($text)) {
1076 $stamp1 = microtime(true);
1078 $ch = @curl_init($match[1]);
1079 @curl_setopt($ch, CURLOPT_NOBODY, true);
1080 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1081 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1083 $curl_info = @curl_getinfo($ch);
1085 DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
1087 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1088 $text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
1090 $text = "[url=" . $match[2] . ']' . $match[2] . "[/url]";
1092 // if its not a picture then look if its a page that contains a picture link
1093 $body = Network::fetchUrl($match[1]);
1095 $doc = new DOMDocument();
1096 @$doc->loadHTML($body);
1097 $xpath = new DOMXPath($doc);
1098 $list = $xpath->query("//meta[@name]");
1099 foreach ($list as $node) {
1102 if ($node->attributes->length) {
1103 foreach ($node->attributes as $attribute) {
1104 $attr[$attribute->name] = $attribute->value;
1108 if (strtolower($attr['name']) == 'twitter:image') {
1109 $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
1113 DI::cache()->set($cache_key, $text);
1119 private static function expandLinksCallback($match)
1121 if (($match[3] == '') || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1122 return ($match[1] . "[url]" . $match[2] . "[/url]");
1124 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1128 private static function cleanPictureLinksCallback($match)
1132 // When the picture link is the own photo path then we can avoid fetching the link
1133 $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
1134 if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
1135 if (!empty($match[3])) {
1136 $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
1138 $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
1143 $cache_key = 'clean:' . $match[1];
1144 $text = DI::cache()->get($cache_key);
1145 if (!is_null($text)) {
1149 // Only fetch the header, not the content
1150 $stamp1 = microtime(true);
1152 $ch = @curl_init($match[1]);
1153 @curl_setopt($ch, CURLOPT_NOBODY, true);
1154 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1155 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1157 $curl_info = @curl_getinfo($ch);
1159 DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
1161 // if its a link to a picture then embed this picture
1162 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1163 $text = '[img]' . $match[1] . '[/img]';
1165 if (!empty($match[3])) {
1166 $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
1168 $text = '[img]' . $match[2] . '[/img]';
1171 // if its not a picture then look if its a page that contains a picture link
1172 $body = Network::fetchUrl($match[1]);
1174 $doc = new DOMDocument();
1175 @$doc->loadHTML($body);
1176 $xpath = new DOMXPath($doc);
1177 $list = $xpath->query("//meta[@name]");
1178 foreach ($list as $node) {
1180 if ($node->attributes->length) {
1181 foreach ($node->attributes as $attribute) {
1182 $attr[$attribute->name] = $attribute->value;
1186 if (strtolower($attr['name']) == "twitter:image") {
1187 if (!empty($match[3])) {
1188 $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
1190 $text = "[img]" . $attr['content'] . "[/img]";
1195 DI::cache()->set($cache_key, $text);
1200 public static function cleanPictureLinks($text)
1202 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1203 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $return);
1208 * Converts a BBCode message to HTML message
1210 * BBcode 2 HTML was written by WAY2WEB.net
1211 * extended to work with Mistpark/Friendica - Mike Macgirvin
1213 * Simple HTML values meaning:
1214 * - 0: Friendica display
1216 * - 2: Used for Windows Phone push, Friendica API
1217 * - 3: Used before converting to Markdown in bb2diaspora.php
1218 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1221 * - 7: Used for dfrn, OStatus
1222 * - 8: Used for Twitter, WP backlink text setting
1225 * @param string $text
1226 * @param bool $try_oembed
1227 * @param int $simple_html
1228 * @param bool $for_plaintext
1230 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1232 public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false)
1237 * preg_match_callback function to replace potential Oembed tags with Oembed content
1239 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1241 * $match[2] = $title or absent
1243 $try_oembed_callback = function ($match)
1246 $title = $match[2] ?? null;
1249 $return = OEmbed::getHTML($url, $title);
1250 } catch (Exception $ex) {
1251 $return = $match[0];
1257 // Extracting code blocks before the whitespace processing and the autolinker
1260 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1261 function ($matches) use (&$codeblocks) {
1262 $return = '#codeblock-' . count($codeblocks) . '#';
1263 if (strpos($matches[2], "\n") !== false) {
1264 $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
1266 $codeblocks[] = '<code>' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
1274 // Hide all [noparse] contained bbtags by spacefying them
1275 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1277 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1278 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1279 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1281 // Remove the abstract element. It is a non visible element.
1282 $text = self::stripAbstract($text);
1284 // Move all spaces out of the tags
1285 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1286 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1288 // Extract the private images which use data urls since preg has issues with
1289 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1290 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1292 $extracted = self::extractImagesFromItemBody($text);
1293 $text = $extracted['body'];
1294 $saved_image = $extracted['images'];
1296 // If we find any event code, turn it into an event.
1297 // After we're finished processing the bbcode we'll
1298 // replace all of the event code with a reformatted version.
1300 $ev = Event::fromBBCode($text);
1302 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1303 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1305 $text = str_replace("<", "<", $text);
1306 $text = str_replace(">", ">", $text);
1308 // remove some newlines before the general conversion
1309 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1310 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1312 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1314 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1317 // Convert new line chars to html <br /> tags
1319 // nlbr seems to be hopelessly messed up
1320 // $Text = nl2br($Text);
1322 // We'll emulate it.
1324 $text = trim($text);
1325 $text = str_replace("\r\n", "\n", $text);
1327 // Remove linefeeds inside of the table elements. See issue #6799
1328 $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
1329 "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
1330 "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
1331 "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "];
1332 $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
1333 "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
1334 "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
1335 "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"];
1338 $text = str_replace($search, $replace, $text);
1339 } while ($oldtext != $text);
1341 // Replace these here only once
1342 $search = ["\n[table]", "[/table]\n"];
1343 $replace = ["[table]", "[/table]"];
1344 $text = str_replace($search, $replace, $text);
1346 // removing multiplicated newlines
1347 if (DI::config()->get('system', 'remove_multiplicated_lines')) {
1348 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1349 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1350 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1351 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1354 $text = str_replace($search, $replace, $text);
1355 } while ($oldtext != $text);
1358 /// @todo Have a closer look at the different html modes
1359 // Handle attached links or videos
1360 if (in_array($simple_html, [9])) {
1361 $text = self::removeAttachment($text);
1362 } elseif (!in_array($simple_html, [0, 4])) {
1363 $text = self::removeAttachment($text, true);
1365 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1368 // leave open the posibility of [map=something]
1369 // this is replaced in Item::prepareBody() which has knowledge of the item location
1370 if (strpos($text, '[/map]') !== false) {
1371 $text = preg_replace_callback(
1372 "/\[map\](.*?)\[\/map\]/ism",
1373 function ($match) use ($simple_html) {
1374 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1380 if (strpos($text, '[map=') !== false) {
1381 $text = preg_replace_callback(
1382 "/\[map=(.*?)\]/ism",
1383 function ($match) use ($simple_html) {
1384 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1390 if (strpos($text, '[map]') !== false) {
1391 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1394 // Check for headers
1395 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1396 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1397 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1398 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1399 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1400 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1402 // Check for paragraph
1403 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1405 // Check for bold text
1406 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1408 // Check for Italics text
1409 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1411 // Check for Underline text
1412 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1414 // Check for strike-through text
1415 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1417 // Check for over-line text
1418 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1420 // Check for colored text
1421 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1423 // Check for sized text
1424 // [size=50] --> font-size: 50px (with the unit).
1425 if ($simple_html != 3) {
1426 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1427 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1429 // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
1430 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text);
1434 // Check for centered text
1435 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1437 // Check for list text
1438 $text = str_replace("[*]", "<li>", $text);
1440 // Check for style sheet commands
1441 $text = preg_replace_callback(
1442 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1444 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1449 // Check for CSS classes
1450 $text = preg_replace_callback(
1451 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1453 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1458 // handle nested lists
1461 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1462 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1463 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1464 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1465 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1466 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1467 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1468 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1469 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1470 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1471 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1472 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1473 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1474 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1477 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1478 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1479 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1480 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1482 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1483 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1485 $text = str_replace('[hr]', '<hr />', $text);
1487 if (!$for_plaintext) {
1490 // Escaping BBCodes susceptible to contain rogue URL we don'' want the autolinker to catch
1491 $text = preg_replace_callback('#\[(url|img|audio|video|youtube|vimeo|share|attachment|iframe|bookmark).+?\[/\1\]#ism',
1492 function ($matches) use (&$escaped) {
1493 $return = '{escaped-' . count($escaped) . '}';
1494 $escaped[] = $matches[0];
1501 // Autolinker for isolated URLs
1502 $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1504 // Restoring escaped blocks
1505 $text = preg_replace_callback('/{escaped-([0-9]+)}/iU',
1506 function ($matches) use ($escaped) {
1507 return $escaped[intval($matches[1])] ?? $matches[0];
1513 // This is actually executed in Item::prepareBody()
1515 $nosmile = strpos($text, '[nosmile]') !== false;
1516 $text = str_replace('[nosmile]', '', $text);
1518 // Check for font change text
1519 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1521 // Declare the format for [spoiler] layout
1522 $SpoilerLayout = '<details class="spoiler"><summary>' . DI::l10n()->t('Click to open/close') . '</summary>$1</details>';
1524 // Check for [spoiler] text
1525 // handle nested quotes
1527 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1528 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", $SpoilerLayout, $text);
1531 // Check for [spoiler=Title] text
1533 // handle nested quotes
1535 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1536 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1537 '<details class="spoiler"><summary>$1</summary>$2</details>',
1541 // Declare the format for [quote] layout
1542 $QuoteLayout = '<blockquote>$1</blockquote>';
1544 // Check for [quote] text
1545 // handle nested quotes
1547 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1548 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1551 // Check for [quote=Author] text
1553 $t_wrote = DI::l10n()->t('$1 wrote:');
1555 // handle nested quotes
1557 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1558 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1559 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1564 // [img=widthxheight]image source[/img]
1565 $text = preg_replace_callback(
1566 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1567 function ($matches) use ($simple_html) {
1568 if (strpos($matches[3], "data:image/") === 0) {
1572 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1573 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1578 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1579 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1581 $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/ism",
1582 function ($matches) use ($simple_html) {
1583 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1584 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1585 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '" title="' . $matches[2] . '">';
1590 // [img]pathtoimage[/img]
1591 $text = preg_replace_callback(
1592 "/\[img\](.*?)\[\/img\]/ism",
1593 function ($matches) use ($simple_html) {
1594 if (strpos($matches[1], "data:image/") === 0) {
1598 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1599 return "[img]" . $matches[1] . "[/img]";
1604 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1605 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1607 $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);
1608 $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 />', $text);
1609 //$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);
1611 // Simplify "video" element
1612 $text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
1616 $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);
1617 $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);
1619 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1620 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1622 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1623 '<a href="$1" target="_blank">$1</a>', $text);
1624 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1625 '<a href="$1" target="_blank">$1</a>', $text);
1628 // html5 video and audio
1632 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1634 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1637 // Youtube extensions
1639 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1640 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1641 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1644 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1645 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1646 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1649 $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);
1651 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1652 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1656 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1657 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1660 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1661 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1664 $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);
1666 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1667 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1671 $text = OEmbed::BBCode2HTML($text);
1673 // Avoid triple linefeeds through oembed
1674 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1676 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1677 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1678 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1679 // start which is always required). Allow desc with a missing summary for compatibility.
1681 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1682 $sub = Event::getHTML($ev, $simple_html);
1684 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1685 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1686 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1687 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1688 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1689 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1690 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1693 // Replace non graphical smilies for external posts
1694 if (!$nosmile && !$for_plaintext) {
1695 $text = Smilies::replace($text);
1698 if (!$for_plaintext) {
1699 if (in_array($simple_html, [7, 9])) {
1700 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1701 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1704 $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1705 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1708 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1710 // Remove all hashtag addresses
1711 if ($simple_html && !in_array($simple_html, [3, 7, 9])) {
1712 $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1713 } elseif ($simple_html == 3) {
1714 // The ! is converted to @ since Diaspora only understands the @
1715 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1716 '@<a href="$2">$3</a>',
1718 } elseif (in_array($simple_html, [7, 9])) {
1719 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1720 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1722 } elseif (!$simple_html) {
1723 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1724 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1728 // Bookmarks in red - will be converted to bookmarks in friendica
1729 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1730 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1731 $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1732 "[bookmark=$1]$2[/bookmark]", $text);
1734 if (in_array($simple_html, [2, 6, 7, 8])) {
1735 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1736 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1737 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1740 if ($simple_html == 5) {
1741 $text = preg_replace("/[^#@!]\[url\=(.*?)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1744 // Perform URL Search
1746 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1749 if ($simple_html == 5) {
1750 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1752 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1755 // Handle Diaspora posts
1756 $text = preg_replace_callback(
1757 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1759 return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1763 $text = preg_replace_callback(
1764 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1766 return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1770 // Server independent link to posts and comments
1771 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1772 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1773 $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
1777 * - #[url=<anything>]<term>[/url]
1778 * - [url=<anything>]#<term>[/url]
1780 $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) {
1782 . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1783 . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
1784 . XML::escape($matches[1])
1788 // We need no target="_blank" for local links
1789 // convert links start with DI::baseUrl() as local link without the target="_blank" attribute
1790 $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
1791 $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1792 $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1794 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1795 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1797 // Red compatibility, though the link can't be authenticated on Friendica
1798 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1801 // we may need to restrict this further if it picks up too many strays
1802 // link acct:user@host to a webfinger profile redirector
1804 $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);
1806 // Perform MAIL Search
1807 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1808 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1810 // Unhide all [noparse] contained bbtags unspacefying them
1811 // and triming the [noparse] tag.
1813 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1814 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1815 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1817 /// @todo What is the meaning of these lines?
1818 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1819 $text = preg_replace('/\&\#039\;/', '\'', $text);
1821 // Currently deactivated, it made problems with " inside of alt texts.
1822 //$text = preg_replace('/\"\;/', '"', $text);
1824 // fix any escaped ampersands that may have been converted into links
1825 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1827 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1828 $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:'];
1830 array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
1832 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1833 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $text);
1835 // sanitize href attributes (only whitelisted protocols URLs)
1836 // default value for backward compatibility
1837 $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
1839 // Always allowed protocol even if config isn't set or not including it
1840 $allowed_link_protocols[] = '//';
1841 $allowed_link_protocols[] = 'http://';
1842 $allowed_link_protocols[] = 'https://';
1843 $allowed_link_protocols[] = 'redir/';
1845 array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
1847 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1848 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
1851 $text = self::convertShare(
1853 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1854 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1859 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1862 // Restore code blocks
1863 $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
1864 function ($matches) use ($codeblocks) {
1865 $return = $matches[0];
1866 if (isset($codeblocks[intval($matches[1])])) {
1867 $return = $codeblocks[$matches[1]];
1874 // Clean up the HTML by loading and saving the HTML with the DOM.
1875 // Bad structured html can break a whole page.
1876 // For performance reasons do it only with activated item cache or at export.
1877 if (!$try_oembed || (get_itemcachepath() != '')) {
1878 $doc = new DOMDocument();
1879 $doc->preserveWhiteSpace = false;
1881 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1883 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1884 $encoding = '<?xml encoding="UTF-8">';
1885 @$doc->loadHTML($encoding . $doctype . '<html><body>' . $text . '</body></html>');
1886 $doc->encoding = 'UTF-8';
1887 $text = $doc->saveHTML();
1888 $text = str_replace(['<html><body>', '</body></html>', $doctype, $encoding], ['', '', '', ''], $text);
1890 $text = str_replace('<br></li>', '</li>', $text);
1892 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1895 // Clean up some useless linebreaks in lists
1896 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1897 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1898 //$Text = str_replace('</li><br />', '</li>', $Text);
1899 //$Text = str_replace('<br /><li>', '<li>', $Text);
1900 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1902 Hook::callAll('bbcode', $text);
1908 * Strips the "abstract" tag from the provided text
1910 * @param string $text The text with BBCode
1911 * @return string The same text - but without "abstract" element
1913 public static function stripAbstract($text)
1915 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1916 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1922 * Returns the value of the "abstract" element
1924 * @param string $text The text that maybe contains the element
1925 * @param string $addon The addon for which the abstract is meant for
1926 * @return string The abstract
1928 public static function getAbstract($text, $addon = '')
1932 $addon = strtolower($addon);
1934 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1935 foreach ($results AS $result) {
1936 $abstracts[strtolower($result[1])] = $result[2];
1940 if (isset($abstracts[$addon])) {
1941 $abstract = $abstracts[$addon];
1944 if ($abstract == '' && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1945 $abstract = $result[1];
1952 * Callback function to replace a Friendica style mention in a mention for Diaspora
1954 * @param array $match Matching values for the callback
1955 * [1] = Mention type (! or @)
1958 * @return string Replaced mention
1959 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1960 * @throws \ImagickException
1962 private static function bbCodeMention2DiasporaCallback($match)
1964 $contact = Contact::getDetailsByURL($match[3]);
1966 if (empty($contact['addr'])) {
1967 $contact = Probe::uri($match[3]);
1970 if (empty($contact['addr'])) {
1974 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
1979 * Converts a BBCode text into Markdown
1981 * This function converts a BBCode item body to be sent to Markdown-enabled
1982 * systems like Diaspora and Libertree
1984 * @param string $text
1985 * @param bool $for_diaspora Diaspora requires more changes than Libertree
1987 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1989 public static function toMarkdown($text, $for_diaspora = true)
1993 $original_text = $text;
1995 // Since Diaspora is creating a summary for links, this function removes them before posting
1996 if ($for_diaspora) {
1997 $text = self::removeShareInformation($text);
2001 * Transform #tags, strip off the [url] and replace spaces with underscore
2003 $url_search_string = "^\[\]";
2004 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
2005 function ($matches) {
2006 return '#' . str_replace(' ', '_', $matches[2]);
2011 // Converting images with size parameters to simple images. Markdown doesn't know it.
2012 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2014 // Convert it to HTML - don't try oembed
2015 if ($for_diaspora) {
2016 $text = self::convert($text, false, 3);
2018 // Add all tags that maybe were removed
2019 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2021 foreach ($tags[2] as $tag) {
2022 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2023 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2024 $tagline .= '#' . $tag . ' ';
2027 $text = $text . " " . $tagline;
2030 $text = self::convert($text, false, 4);
2033 // If a link is followed by a quote then there should be a newline before it
2034 // Maybe we should make this newline at every time before a quote.
2035 $text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
2037 $stamp1 = microtime(true);
2039 // Now convert HTML to Markdown
2040 $text = HTML::toMarkdown($text);
2042 DI::profiler()->saveTimestamp($stamp1, "parser", System::callstack());
2044 // Libertree has a problem with escaped hashtags.
2045 $text = str_replace(['\#'], ['#'], $text);
2047 // Remove any leading or trailing whitespace, as this will mess up
2048 // the Diaspora signature verification and cause the item to disappear
2049 $text = trim($text);
2051 if ($for_diaspora) {
2052 $url_search_string = "^\[\]";
2053 $text = preg_replace_callback(
2054 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
2055 ['self', 'bbCodeMention2DiasporaCallback'],
2060 Hook::callAll('bb2diaspora', $text);
2066 * Pull out all #hashtags and @person tags from $string.
2068 * We also get @person@domain.com - which would make
2069 * the regex quite complicated as tags can also
2070 * end a sentence. So we'll run through our results
2071 * and strip the period from any tags which end with one.
2072 * Returns array of tags found, or empty array.
2074 * @param string $string Post content
2076 * @return array List of tag and person names
2078 public static function getTags($string)
2082 // Convert hashtag links to hashtags
2083 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2', $string);
2085 // ignore anything in a code block
2086 $string = preg_replace('/\[code.*?\].*?\[\/code\]/sm', '', $string);
2088 // Force line feeds at bbtags
2089 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
2091 // ignore anything in a bbtag
2092 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
2094 // Match full names against @tags including the space between first and last
2095 // We will look these up afterward to see if they are full names or not recognisable.
2097 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
2098 foreach ($matches[1] as $match) {
2099 if (strstr($match, ']')) {
2100 // we might be inside a bbcode color tag - leave it alone
2104 if (substr($match, -1, 1) === '.') {
2105 $ret[] = substr($match, 0, -1);
2112 // Otherwise pull out single word tags. These can be @nickname, @first_last
2115 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
2116 foreach ($matches[1] as $match) {
2117 if (strstr($match, ']')) {
2118 // we might be inside a bbcode color tag - leave it alone
2122 if (substr($match, -1, 1) === '.') {
2123 $match = substr($match,0,-1);
2126 // ignore strictly numeric tags like #1
2127 if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
2131 // try not to catch url fragments
2132 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {