3 * @copyright Copyright (C) 2020, Friendica
5 * @license GNU AGPL version 3 or any later version
7 * This program is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU Affero General Public License as
9 * published by the Free Software Foundation, either version 3 of the
10 * License, or (at your option) any later version.
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU Affero General Public License for more details.
17 * You should have received a copy of the GNU Affero General Public License
18 * along with this program. If not, see <https://www.gnu.org/licenses/>.
22 namespace Friendica\Content\Text;
27 use Friendica\Content\OEmbed;
28 use Friendica\Content\Smilies;
29 use Friendica\Core\Hook;
30 use Friendica\Core\Logger;
31 use Friendica\Core\Protocol;
32 use Friendica\Core\Renderer;
33 use Friendica\Core\System;
35 use Friendica\Model\Contact;
36 use Friendica\Model\Event;
37 use Friendica\Model\Photo;
38 use Friendica\Network\Probe;
39 use Friendica\Object\Image;
40 use Friendica\Protocol\Activity;
41 use Friendica\Util\Images;
42 use Friendica\Util\Map;
43 use Friendica\Util\Network;
44 use Friendica\Util\ParseUrl;
45 use Friendica\Util\Proxy as ProxyUtils;
46 use Friendica\Util\Strings;
47 use Friendica\Util\XML;
52 * Fetches attachment data that were generated the old way
54 * @param string $body Message body
56 * 'type' -> Message type ('link', 'video', 'photo')
57 * 'text' -> Text before the shared message
58 * 'after' -> Text after the shared message
59 * 'image' -> Preview image of the message
60 * 'url' -> Url to the attached message
61 * 'title' -> Title of the attachment
62 * 'description' -> Description of the attachment
63 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
65 private static function getOldAttachmentData($body)
69 // Simplify image codes
70 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
72 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
73 foreach ($attached as $data) {
74 if (!in_array($data[1], ['type-link', 'type-video', 'type-photo'])) {
78 $post['type'] = substr($data[1], 5);
80 $pos = strpos($body, $data[0]);
82 $post['text'] = trim(substr($body, 0, $pos));
83 $post['after'] = trim(substr($body, $pos + strlen($data[0])));
85 $post['text'] = trim(str_replace($data[0], '', $body));
89 $attacheddata = $data[2];
91 if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
93 $picturedata = Images::getInfoFromURLCached($matches[1]);
96 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
97 $post['image'] = $matches[1];
99 $post['preview'] = $matches[1];
104 if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
105 $post['url'] = $matches[1];
106 $post['title'] = $matches[2];
108 if (!empty($post['url']) && (in_array($post['type'], ['link', 'video']))
109 && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
110 $post['url'] = $matches[1];
113 // Search for description
114 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
115 $post['description'] = $matches[1];
123 * Fetches attachment data that were generated with the "attachment" element
125 * @param string $body Message body
127 * 'type' -> Message type ('link', 'video', 'photo')
128 * 'text' -> Text before the shared message
129 * 'after' -> Text after the shared message
130 * 'image' -> Preview image of the message
131 * 'url' -> Url to the attached message
132 * 'title' -> Title of the attachment
133 * 'description' -> Description of the attachment
134 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
136 public static function getAttachmentData($body)
148 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
149 return self::getOldAttachmentData($body);
152 $attributes = $match[2];
154 $data['text'] = trim($match[1]);
157 preg_match("/type='(.*?)'/ism", $attributes, $matches);
158 if (!empty($matches[1])) {
159 $type = strtolower($matches[1]);
162 preg_match('/type="(.*?)"/ism', $attributes, $matches);
163 if (!empty($matches[1])) {
164 $type = strtolower($matches[1]);
171 if (!in_array($type, ['link', 'audio', 'photo', 'video'])) {
176 $data['type'] = $type;
180 preg_match("/url='(.*?)'/ism", $attributes, $matches);
181 if (!empty($matches[1])) {
185 preg_match('/url="(.*?)"/ism', $attributes, $matches);
186 if (!empty($matches[1])) {
191 $data['url'] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
195 preg_match("/title='(.*?)'/ism", $attributes, $matches);
196 if (!empty($matches[1])) {
197 $title = $matches[1];
200 preg_match('/title="(.*?)"/ism', $attributes, $matches);
201 if (!empty($matches[1])) {
202 $title = $matches[1];
206 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
207 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
208 $title = str_replace(['[', ']'], ['[', ']'], $title);
209 $data['title'] = $title;
213 preg_match("/image='(.*?)'/ism", $attributes, $matches);
214 if (!empty($matches[1])) {
215 $image = $matches[1];
218 preg_match('/image="(.*?)"/ism', $attributes, $matches);
219 if (!empty($matches[1])) {
220 $image = $matches[1];
224 $data['image'] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
228 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
229 if (!empty($matches[1])) {
230 $preview = $matches[1];
233 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
234 if (!empty($matches[1])) {
235 $preview = $matches[1];
238 if ($preview != '') {
239 $data['preview'] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
242 $data['description'] = trim($match[3]);
244 $data['after'] = trim($match[4]);
249 public static function getAttachedData($body, $item = [])
253 - type: link, video, photo
261 $has_title = !empty($item['title']);
262 $plink = $item['plink'] ?? '';
263 $post = self::getAttachmentData($body);
265 // Get all linked images with alternative image description
266 if (preg_match_all("/\[img=([^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
267 foreach ($pictures as $picture) {
268 if (Photo::isLocal($picture[1])) {
269 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2]];
272 if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
273 $post['image_description'] = $post['images'][0]['description'];
277 if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
278 foreach ($pictures as $picture) {
279 if (Photo::isLocal($picture[1])) {
280 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => ''];
285 // if nothing is found, it maybe having an image.
286 if (!isset($post['type'])) {
287 // Simplify image codes
288 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
289 $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
290 $post['text'] = $body;
292 if (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $body, $pictures, PREG_SET_ORDER)) {
293 if ((count($pictures) == 1) && !$has_title) {
294 if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
295 // Replace the preview picture with the real picture
296 $url = str_replace('-1.', '-0.', $pictures[0][2]);
297 $data = ['url' => $url, 'type' => 'photo'];
299 // Checking, if the link goes to a picture
300 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
304 // Sometimes photo posts to the own album are not detected at the start.
305 // So we seem to cannot use the cache for these cases. That's strange.
306 if (($data['type'] != 'photo') && strstr($pictures[0][1], "/photos/")) {
307 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
310 if ($data['type'] == 'photo') {
311 $post['type'] = 'photo';
312 if (isset($data['images'][0])) {
313 $post['image'] = $data['images'][0]['src'];
314 $post['url'] = $data['url'];
316 $post['image'] = $data['url'];
319 $post['preview'] = $pictures[0][2];
320 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
322 $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
323 if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
324 $post['type'] = 'photo';
325 $post['image'] = $pictures[0][1];
326 $post['preview'] = $pictures[0][2];
327 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
330 } elseif (count($pictures) > 0) {
331 $post['type'] = 'link';
332 $post['url'] = $plink;
333 $post['image'] = $pictures[0][2];
334 $post['text'] = $body;
336 foreach ($pictures as $picture) {
337 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
340 } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
341 if ((count($pictures) == 1) && !$has_title) {
342 $post['type'] = 'photo';
343 $post['image'] = $pictures[0][1];
344 $post['text'] = str_replace($pictures[0][0], '', $body);
345 } elseif (count($pictures) > 0) {
346 $post['type'] = 'link';
347 $post['url'] = $plink;
348 $post['image'] = $pictures[0][1];
349 $post['text'] = $body;
351 foreach ($pictures as $picture) {
352 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
357 // Test for the external links
358 preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER);
359 preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $links2, PREG_SET_ORDER);
361 $links = array_merge($links1, $links2);
363 // If there is only a single one, then use it.
364 // This should cover link posts via API.
365 if ((count($links) == 1) && !isset($post['preview']) && !$has_title) {
366 $post['type'] = 'link';
367 $post['url'] = $links[0][1];
370 // Simplify "video" element
371 $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
373 // Now count the number of external media links
374 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
375 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
376 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER);
377 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $links4, PREG_SET_ORDER);
379 // Add them to the other external links
380 $links = array_merge($links, $links1, $links2, $links3, $links4);
382 // Are there more than one?
383 if (count($links) > 1) {
384 // The post will be the type "text", which means a blog post
385 unset($post['type']);
386 $post['url'] = $plink;
389 if (!isset($post['type'])) {
390 $post['type'] = "text";
391 $post['text'] = trim($body);
393 } elseif (isset($post['url']) && ($post['type'] == 'video')) {
394 $data = ParseUrl::getSiteinfoCached($post['url'], true);
396 if (isset($data['images'][0])) {
397 $post['image'] = $data['images'][0]['src'];
405 * Remove [attachment] BBCode and replaces it with a regular [url]
407 * @param string $body
408 * @param boolean $no_link_desc No link description
410 * @return string with replaced body
412 public static function removeAttachment($body, $no_link_desc = false)
414 return preg_replace_callback("/\s*\[attachment (.*)\](.*?)\[\/attachment\]\s*/ism",
415 function ($match) use ($no_link_desc) {
416 $attach_data = self::getAttachmentData($match[0]);
417 if (empty($attach_data['url'])) {
419 } elseif (empty($attach_data['title']) || $no_link_desc) {
420 return "\n[url]" . $attach_data['url'] . "[/url]\n";
422 return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
428 * Converts a BBCode text into plaintext
431 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
435 public static function toPlaintext($text, $keep_urls = true)
437 $naked_text = HTML::toPlaintext(BBCode::convert($text, false, 0, true), 0, !$keep_urls);
442 private static function proxyUrl($image, $simplehtml = false)
444 // Only send proxied pictures to API and for internal display
445 if (in_array($simplehtml, [false, 2])) {
446 return ProxyUtils::proxifyUrl($image);
453 * This function changing the visual size (not the real size) of images.
454 * The function does not work for pictures with an alternate text description.
455 * This could only be changed by using some new "img" BBCode format.
457 * @param string $srctext The body with images
458 * @return string The body with possibly scaled images
460 public static function scaleExternalImages(string $srctext)
464 // Simplify image links
465 $s = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $s);
468 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
470 foreach ($matches as $mtch) {
471 Logger::log('scale_external_image: ' . $mtch[1]);
473 $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
474 if (stristr($mtch[1], $hostname)) {
478 $curlResult = Network::curl($mtch[1], true);
479 if (!$curlResult->isSuccess()) {
483 $i = $curlResult->getBody();
484 $type = $curlResult->getContentType();
485 $type = Images::getMimeTypeByData($i, $mtch[1], $type);
488 $Image = new Image($i, $type);
489 if ($Image->isValid()) {
490 $orig_width = $Image->getWidth();
491 $orig_height = $Image->getHeight();
493 if ($orig_width > 640 || $orig_height > 640) {
494 $Image->scaleDown(640);
495 $new_width = $Image->getWidth();
496 $new_height = $Image->getHeight();
497 Logger::log('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], Logger::DEBUG);
500 '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
504 Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG);
515 * Truncates imported message body string length to max_import_size
517 * The purpose of this function is to apply system message length limits to
518 * imported messages without including any embedded photos in the length
520 * @param string $body
522 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
524 public static function limitBodySize($body)
526 $maxlen = DI::config()->get('config', 'max_import_size', 0);
528 // If the length of the body, including the embedded images, is smaller
529 // than the maximum, then don't waste time looking for the images
530 if ($maxlen && (strlen($body) > $maxlen)) {
532 Logger::log('the total body length exceeds the limit', Logger::DEBUG);
538 $img_start = strpos($orig_body, '[img');
539 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
540 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
541 while (($img_st_close !== false) && ($img_end !== false)) {
543 $img_st_close++; // make it point to AFTER the closing bracket
544 $img_end += $img_start;
545 $img_end += strlen('[/img]');
547 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
548 // This is an embedded image
550 if (($textlen + $img_start) > $maxlen) {
551 if ($textlen < $maxlen) {
552 Logger::log('the limit happens before an embedded image', Logger::DEBUG);
553 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
557 $new_body = $new_body . substr($orig_body, 0, $img_start);
558 $textlen += $img_start;
561 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
564 if (($textlen + $img_end) > $maxlen) {
565 if ($textlen < $maxlen) {
566 Logger::log('the limit happens before the end of a non-embedded image', Logger::DEBUG);
567 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
571 $new_body = $new_body . substr($orig_body, 0, $img_end);
572 $textlen += $img_end;
575 $orig_body = substr($orig_body, $img_end);
577 if ($orig_body === false) {
578 // in case the body ends on a closing image tag
582 $img_start = strpos($orig_body, '[img');
583 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
584 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
587 if (($textlen + strlen($orig_body)) > $maxlen) {
588 if ($textlen < $maxlen) {
589 Logger::log('the limit happens after the end of the last image', Logger::DEBUG);
590 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
593 Logger::log('the text size with embedded images extracted did not violate the limit', Logger::DEBUG);
594 $new_body = $new_body . $orig_body;
604 * Processes [attachment] tags
606 * Note: Can produce a [bookmark] tag in the returned string
608 * @param string $text
609 * @param bool|int $simplehtml
610 * @param bool $tryoembed
612 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
614 private static function convertAttachment($text, $simplehtml = false, $tryoembed = true)
616 $data = self::getAttachmentData($text);
617 if (empty($data) || empty($data['url'])) {
621 if (isset($data['title'])) {
622 $data['title'] = strip_tags($data['title']);
623 $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
625 $data['title'] = null;
628 if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
629 $data['preview'] = $data['image'];
635 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
636 $return = OEmbed::getHTML($data['url'], $data['title']);
638 throw new Exception('OEmbed is disabled for this attachment.');
640 } catch (Exception $e) {
641 $data['title'] = ($data['title'] ?? '') ?: $data['url'];
643 if ($simplehtml != 4) {
644 $return = sprintf('<div class="type-%s">', $data['type']);
647 if (!empty($data['title']) && !empty($data['url'])) {
648 if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
649 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
651 if (!empty($data['image'])) {
652 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
653 } elseif (!empty($data['preview'])) {
654 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data['url'], self::proxyUrl($data['preview'], $simplehtml), $data['title']);
656 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
660 if (!empty($data['description']) && $data['description'] != $data['title']) {
661 // Sanitize the HTML by converting it to BBCode
662 $bbcode = HTML::toBBCode($data['description']);
663 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
666 if (!empty($data['url'])) {
667 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
670 if ($simplehtml != 4) {
675 return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
678 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
680 $data = self::getAttachmentData($Text);
685 return $data['text'] . ($data['after'] ?? '');
688 $title = htmlentities($data['title'] ?? '', ENT_QUOTES, 'UTF-8', false);
689 $text = htmlentities($data['text'], ENT_QUOTES, 'UTF-8', false);
690 if ($plaintext || (($title != '') && strstr($text, $title))) {
691 $data['title'] = $data['url'];
692 } elseif (($text != '') && strstr($title, $text)) {
693 $data['text'] = $data['title'];
694 $data['title'] = $data['url'];
697 if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
698 return $data['title'] . $data['after'];
701 // If the link already is included in the post, don't add it again
702 if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
703 return $data['text'] . $data['after'];
706 $text = $data['text'];
708 if (!empty($data['url']) && !empty($data['title'])) {
709 $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
710 } elseif (!empty($data['url'])) {
711 $text .= "\n[url]" . $data['url'] . '[/url]';
714 return $text . "\n" . $data['after'];
718 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
720 * @param array $match Array with the matching values
721 * @return string reformatted link including HTML codes
723 private static function convertUrlForActivityPubCallback($match)
727 if (isset($match[2]) && ($match[1] != $match[2])) {
731 $parts = parse_url($url);
732 if (!isset($parts['scheme'])) {
736 return self::convertUrlForActivityPub($url);
740 * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
742 * @param string $url URL that is about to be reformatted
743 * @return string reformatted link including HTML codes
745 private static function convertUrlForActivityPub($url)
747 $html = '<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>';
748 return sprintf($html, $url, self::getStyledURL($url));
752 * Converts an URL in a nicer format (without the scheme and possibly shortened)
754 * @param string $url URL that is about to be reformatted
755 * @return string reformatted link
757 private static function getStyledURL($url)
759 $parts = parse_url($url);
760 $scheme = $parts['scheme'] . '://';
761 $styled_url = str_replace($scheme, '', $url);
763 if (strlen($styled_url) > 30) {
764 $styled_url = substr($styled_url, 0, 30) . "…";
771 * [noparse][i]italic[/i][/noparse] turns into
772 * [noparse][ i ]italic[ /i ][/noparse],
773 * to hide them from parser.
775 private static function escapeNoparseCallback($match)
777 $whole_match = $match[0];
778 $captured = $match[1];
779 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
780 $new_str = str_replace($captured, $spacefied, $whole_match);
785 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
786 * now turns back and the [noparse] tags are trimed
787 * returning [i]italic[/i]
789 private static function unescapeNoparseCallback($match)
791 $captured = $match[1];
792 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
797 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
800 * @param string $text Text to search
801 * @param string $name Tag name
802 * @param int $occurrences Number of first occurrences to skip
803 * @return boolean|array
805 public static function getTagPosition($text, $name, $occurrences = 0)
807 if ($occurrences < 0) {
812 for ($i = 0; $i <= $occurrences; $i++) {
813 if ($start_open !== false) {
814 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
818 if ($start_open === false) {
822 $start_equal = strpos($text, '=', $start_open);
823 $start_close = strpos($text, ']', $start_open);
825 if ($start_close === false) {
831 $end_open = strpos($text, '[/' . $name . ']', $start_close);
833 if ($end_open === false) {
839 'open' => $start_open,
840 'close' => $start_close
844 'close' => $end_open + strlen('[/' . $name . ']')
848 if ($start_equal !== false) {
849 $res['start']['equal'] = $start_equal + 1;
856 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
858 * @param string $pattern Preg pattern string
859 * @param string $replace Preg replace string
860 * @param string $name BBCode tag name
861 * @param string $text Text to search
864 public static function pregReplaceInTag($pattern, $replace, $name, $text)
867 $pos = self::getTagPosition($text, $name, $occurrences);
868 while ($pos !== false && $occurrences++ < 1000) {
869 $start = substr($text, 0, $pos['start']['open']);
870 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
871 $end = substr($text, $pos['end']['close']);
872 if ($end === false) {
876 $subject = preg_replace($pattern, $replace, $subject);
877 $text = $start . $subject . $end;
879 $pos = self::getTagPosition($text, $name, $occurrences);
885 private static function extractImagesFromItemBody($body)
892 $img_start = strpos($orig_body, '[img');
893 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
894 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
895 while (($img_st_close !== false) && ($img_end !== false)) {
896 $img_st_close++; // make it point to AFTER the closing bracket
897 $img_end += $img_start;
899 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
900 // This is an embedded image
901 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
902 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
906 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
909 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
911 if ($orig_body === false) {
912 // in case the body ends on a closing image tag
916 $img_start = strpos($orig_body, '[img');
917 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
918 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
921 $new_body = $new_body . $orig_body;
923 return ['body' => $new_body, 'images' => $saved_image];
926 private static function interpolateSavedImagesIntoItemBody($body, array $images)
931 foreach ($images as $image) {
932 // We're depending on the property of 'foreach' (specified on the PHP website) that
933 // it loops over the array starting from the first element and going sequentially
934 // to the last element
935 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
936 '<img src="' . self::proxyUrl($image) . '" alt="' . DI::l10n()->t('Image/photo') . '" />', $newbody);
944 * This function converts a [share] block to text according to a provided callback function whose signature is:
946 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
949 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
951 * - $author_contact is a contact record array
952 * - $content is the inner content of the [share] block
953 * - $is_quote_share indicates whether there's any content before the [share] block
954 * - Return value is the string that should replace the [share] block in the provided text
956 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
958 * @param string $text A BBCode string
959 * @param callable $callback
960 * @return string The BBCode string with all [share] blocks replaced
962 public static function convertShare($text, callable $callback)
964 $return = preg_replace_callback(
965 "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
966 function ($match) use ($callback) {
967 $attribute_string = $match[2];
969 foreach (['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
970 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
971 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
974 // We only call this so that a previously unknown contact can be added.
975 // This is important for the function "Model\Contact::getDetailsByURL()".
976 // This function then can fetch an entry from the contact table.
977 $default['url'] = $attributes['profile'];
979 if (!empty($attributes['author'])) {
980 $default['name'] = $attributes['author'];
983 if (!empty($attributes['avatar'])) {
984 $default['photo'] = $attributes['avatar'];
987 Contact::getIdForURL($attributes['profile'], 0, true, $default);
989 $author_contact = Contact::getDetailsByURL($attributes['profile']);
990 $author_contact['url'] = ($author_contact['url'] ?? $attributes['profile']);
991 $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']);
993 $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author'];
994 $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
995 $attributes['profile'] = ($author_contact['url'] ?? '') ?: $attributes['profile'];
997 if ($attributes['avatar']) {
998 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
1001 return $match[1] . $callback($attributes, $author_contact, $match[3], trim($match[1]) != '');
1010 * Default [share] tag conversion callback
1012 * Note: Can produce a [bookmark] tag in the output
1014 * @see BBCode::convertShare()
1015 * @param array $attributes [share] block attribute values
1016 * @param array $author_contact Contact row of the shared author
1017 * @param string $content Inner content of the [share] block
1018 * @param boolean $is_quote_share Whether there is content before the [share] block
1019 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
1021 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1023 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
1025 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
1027 switch ($simplehtml) {
1029 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' <a href="' . $attributes['profile'] . '">' . $mention . '</a>: </p>' . "\n" . '«' . $content . '»';
1032 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
1035 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
1036 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
1038 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
1040 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
1041 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
1044 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
1046 if (empty($attributes['posted']) && !empty($attributes['link'])) {
1047 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
1053 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
1054 $headline .= DI::l10n()->t('<a href="%1$s" target="_blank" rel="noopener noreferrer">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
1055 $headline .= ':</b></p>' . "\n";
1057 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
1061 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
1063 case 7: // statusnet/GNU Social
1064 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
1066 case 9: // ActivityPub
1067 $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>';
1068 $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
1071 // Transforms quoted tweets in rich attachments to avoid nested tweets
1072 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) {
1074 $text = ($is_quote_share? '<br />' : '') . OEmbed::getHTML($attributes['link']);
1075 } catch (Exception $e) {
1076 $text = ($is_quote_share? '<br />' : '') . sprintf('[bookmark=%s]%s[/bookmark]', $attributes['link'], $content);
1079 $text = ($is_quote_share? "\n" : '');
1081 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1082 $text .= Renderer::replaceMacros($tpl, [
1083 '$profile' => $attributes['profile'],
1084 '$avatar' => $attributes['avatar'],
1085 '$author' => $attributes['author'],
1086 '$link' => $attributes['link'],
1087 '$posted' => $attributes['posted'],
1088 '$content' => trim($content)
1097 private static function removePictureLinksCallback($match)
1099 $cache_key = 'remove:' . $match[1];
1100 $text = DI::cache()->get($cache_key);
1102 if (is_null($text)) {
1105 $stamp1 = microtime(true);
1107 $ch = @curl_init($match[1]);
1108 @curl_setopt($ch, CURLOPT_NOBODY, true);
1109 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1110 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1112 $curl_info = @curl_getinfo($ch);
1114 DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
1116 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1117 $text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
1119 $text = "[url=" . $match[2] . ']' . $match[2] . "[/url]";
1121 // if its not a picture then look if its a page that contains a picture link
1122 $body = Network::fetchUrl($match[1]);
1124 $doc = new DOMDocument();
1125 @$doc->loadHTML($body);
1126 $xpath = new DOMXPath($doc);
1127 $list = $xpath->query("//meta[@name]");
1128 foreach ($list as $node) {
1131 if ($node->attributes->length) {
1132 foreach ($node->attributes as $attribute) {
1133 $attr[$attribute->name] = $attribute->value;
1137 if (strtolower($attr['name']) == 'twitter:image') {
1138 $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
1142 DI::cache()->set($cache_key, $text);
1148 private static function expandLinksCallback($match)
1150 if (($match[3] == '') || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1151 return ($match[1] . "[url]" . $match[2] . "[/url]");
1153 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1157 private static function cleanPictureLinksCallback($match)
1161 // When the picture link is the own photo path then we can avoid fetching the link
1162 $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
1163 if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
1164 if (!empty($match[3])) {
1165 $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
1167 $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
1172 $cache_key = 'clean:' . $match[1];
1173 $text = DI::cache()->get($cache_key);
1174 if (!is_null($text)) {
1178 // Only fetch the header, not the content
1179 $stamp1 = microtime(true);
1181 $ch = @curl_init($match[1]);
1182 @curl_setopt($ch, CURLOPT_NOBODY, true);
1183 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1184 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1186 $curl_info = @curl_getinfo($ch);
1188 DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
1190 // if its a link to a picture then embed this picture
1191 if (substr($curl_info['content_type'], 0, 6) == 'image/') {
1192 $text = '[img]' . $match[1] . '[/img]';
1194 if (!empty($match[3])) {
1195 $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
1197 $text = '[img]' . $match[2] . '[/img]';
1200 // if its not a picture then look if its a page that contains a picture link
1201 $body = Network::fetchUrl($match[1]);
1203 $doc = new DOMDocument();
1204 @$doc->loadHTML($body);
1205 $xpath = new DOMXPath($doc);
1206 $list = $xpath->query("//meta[@name]");
1207 foreach ($list as $node) {
1209 if ($node->attributes->length) {
1210 foreach ($node->attributes as $attribute) {
1211 $attr[$attribute->name] = $attribute->value;
1215 if (strtolower($attr['name']) == "twitter:image") {
1216 if (!empty($match[3])) {
1217 $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
1219 $text = "[img]" . $attr['content'] . "[/img]";
1224 DI::cache()->set($cache_key, $text);
1229 public static function cleanPictureLinks($text)
1231 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1232 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $return);
1237 * Converts a BBCode message to HTML message
1239 * BBcode 2 HTML was written by WAY2WEB.net
1240 * extended to work with Mistpark/Friendica - Mike Macgirvin
1242 * Simple HTML values meaning:
1243 * - 0: Friendica display
1245 * - 2: Used for Windows Phone push, Friendica API
1246 * - 3: Used before converting to Markdown in bb2diaspora.php
1247 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1250 * - 7: Used for dfrn, OStatus
1251 * - 8: Used for Twitter, WP backlink text setting
1254 * @param string $text
1255 * @param bool $try_oembed
1256 * @param int $simple_html
1257 * @param bool $for_plaintext
1259 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1261 public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false)
1266 * preg_match_callback function to replace potential Oembed tags with Oembed content
1268 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1270 * $match[2] = $title or absent
1272 $try_oembed_callback = function ($match)
1275 $title = $match[2] ?? null;
1278 $return = OEmbed::getHTML($url, $title);
1279 } catch (Exception $ex) {
1280 $return = $match[0];
1286 // Extracting code blocks before the whitespace processing and the autolinker
1289 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1290 function ($matches) use (&$codeblocks) {
1291 $return = '#codeblock-' . count($codeblocks) . '#';
1292 if (strpos($matches[2], "\n") !== false) {
1293 $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
1295 $codeblocks[] = '<code>' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
1303 // Hide all [noparse] contained bbtags by spacefying them
1304 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1306 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1307 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1308 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1310 // Remove the abstract element. It is a non visible element.
1311 $text = self::stripAbstract($text);
1313 // Move all spaces out of the tags
1314 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1315 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1317 // Extract the private images which use data urls since preg has issues with
1318 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1319 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1321 $extracted = self::extractImagesFromItemBody($text);
1322 $text = $extracted['body'];
1323 $saved_image = $extracted['images'];
1325 // If we find any event code, turn it into an event.
1326 // After we're finished processing the bbcode we'll
1327 // replace all of the event code with a reformatted version.
1329 $ev = Event::fromBBCode($text);
1331 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1332 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1334 $text = str_replace("<", "<", $text);
1335 $text = str_replace(">", ">", $text);
1337 // remove some newlines before the general conversion
1338 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1339 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1341 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1343 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1346 // Convert new line chars to html <br /> tags
1348 // nlbr seems to be hopelessly messed up
1349 // $Text = nl2br($Text);
1351 // We'll emulate it.
1353 $text = trim($text);
1354 $text = str_replace("\r\n", "\n", $text);
1356 // Remove linefeeds inside of the table elements. See issue #6799
1357 $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
1358 "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
1359 "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
1360 "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "];
1361 $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
1362 "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
1363 "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
1364 "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"];
1367 $text = str_replace($search, $replace, $text);
1368 } while ($oldtext != $text);
1370 // Replace these here only once
1371 $search = ["\n[table]", "[/table]\n"];
1372 $replace = ["[table]", "[/table]"];
1373 $text = str_replace($search, $replace, $text);
1375 // removing multiplicated newlines
1376 if (DI::config()->get('system', 'remove_multiplicated_lines')) {
1377 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1378 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1379 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1380 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1383 $text = str_replace($search, $replace, $text);
1384 } while ($oldtext != $text);
1387 /// @todo Have a closer look at the different html modes
1388 // Handle attached links or videos
1389 if (in_array($simple_html, [9])) {
1390 $text = self::removeAttachment($text);
1391 } elseif (!in_array($simple_html, [0, 4])) {
1392 $text = self::removeAttachment($text, true);
1394 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1397 // leave open the posibility of [map=something]
1398 // this is replaced in Item::prepareBody() which has knowledge of the item location
1399 if (strpos($text, '[/map]') !== false) {
1400 $text = preg_replace_callback(
1401 "/\[map\](.*?)\[\/map\]/ism",
1402 function ($match) use ($simple_html) {
1403 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1409 if (strpos($text, '[map=') !== false) {
1410 $text = preg_replace_callback(
1411 "/\[map=(.*?)\]/ism",
1412 function ($match) use ($simple_html) {
1413 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1419 if (strpos($text, '[map]') !== false) {
1420 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1423 // Check for headers
1424 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1425 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1426 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1427 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1428 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1429 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1431 // Check for paragraph
1432 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1434 // Check for bold text
1435 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1437 // Check for Italics text
1438 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1440 // Check for Underline text
1441 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1443 // Check for strike-through text
1444 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1446 // Check for over-line text
1447 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1449 // Check for colored text
1450 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1452 // Check for sized text
1453 // [size=50] --> font-size: 50px (with the unit).
1454 if ($simple_html != 3) {
1455 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1456 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1458 // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
1459 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text);
1463 // Check for centered text
1464 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1466 // Check for list text
1467 $text = str_replace("[*]", "<li>", $text);
1469 // Check for style sheet commands
1470 $text = preg_replace_callback(
1471 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1473 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1478 // Check for CSS classes
1479 $text = preg_replace_callback(
1480 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1482 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1487 // handle nested lists
1490 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1491 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1492 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1493 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1494 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1495 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1496 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1497 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1498 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1499 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1500 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1501 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1502 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1503 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1506 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1507 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1508 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1509 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1511 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1512 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1514 $text = str_replace('[hr]', '<hr />', $text);
1516 if (!$for_plaintext) {
1519 // Escaping BBCodes susceptible to contain rogue URL we don'' want the autolinker to catch
1520 $text = preg_replace_callback('#\[(url|img|audio|video|youtube|vimeo|share|attachment|iframe|bookmark).+?\[/\1\]#ism',
1521 function ($matches) use (&$escaped) {
1522 $return = '{escaped-' . count($escaped) . '}';
1523 $escaped[] = $matches[0];
1530 // Autolinker for isolated URLs
1531 $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1533 // Restoring escaped blocks
1534 $text = preg_replace_callback('/{escaped-([0-9]+)}/iU',
1535 function ($matches) use ($escaped) {
1536 return $escaped[intval($matches[1])] ?? $matches[0];
1542 // This is actually executed in Item::prepareBody()
1544 $nosmile = strpos($text, '[nosmile]') !== false;
1545 $text = str_replace('[nosmile]', '', $text);
1547 // Check for font change text
1548 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1550 // Declare the format for [spoiler] layout
1551 $SpoilerLayout = '<details class="spoiler"><summary>' . DI::l10n()->t('Click to open/close') . '</summary>$1</details>';
1553 // Check for [spoiler] text
1554 // handle nested quotes
1556 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1557 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", $SpoilerLayout, $text);
1560 // Check for [spoiler=Title] text
1562 // handle nested quotes
1564 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1565 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1566 '<details class="spoiler"><summary>$1</summary>$2</details>',
1570 // Declare the format for [quote] layout
1571 $QuoteLayout = '<blockquote>$1</blockquote>';
1573 // Check for [quote] text
1574 // handle nested quotes
1576 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1577 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1580 // Check for [quote=Author] text
1582 $t_wrote = DI::l10n()->t('$1 wrote:');
1584 // handle nested quotes
1586 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1587 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1588 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1593 // [img=widthxheight]image source[/img]
1594 $text = preg_replace_callback(
1595 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1596 function ($matches) use ($simple_html) {
1597 if (strpos($matches[3], "data:image/") === 0) {
1601 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1602 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1607 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1608 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1610 $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/ism",
1611 function ($matches) use ($simple_html) {
1612 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1613 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1614 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '" title="' . $matches[2] . '">';
1619 // [img]pathtoimage[/img]
1620 $text = preg_replace_callback(
1621 "/\[img\](.*?)\[\/img\]/ism",
1622 function ($matches) use ($simple_html) {
1623 if (strpos($matches[1], "data:image/") === 0) {
1627 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1628 return "[img]" . $matches[1] . "[/img]";
1633 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1634 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1636 $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);
1637 $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 />', $text);
1638 //$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);
1640 // Simplify "video" element
1641 $text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
1645 $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);
1646 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
1648 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1649 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1651 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1652 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1653 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1654 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1657 // html5 video and audio
1661 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1663 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1666 // Youtube extensions
1668 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1669 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1670 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1673 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1674 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1675 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1678 $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);
1680 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1681 '<a href="https://www.youtube.com/watch?v=$1" target="_blank" rel="noopener noreferrer">https://www.youtube.com/watch?v=$1</a>', $text);
1685 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1686 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1689 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1690 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1693 $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);
1695 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1696 '<a href="https://vimeo.com/$1" target="_blank" rel="noopener noreferrer">https://vimeo.com/$1</a>', $text);
1700 $text = OEmbed::BBCode2HTML($text);
1702 // Avoid triple linefeeds through oembed
1703 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1705 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1706 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1707 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1708 // start which is always required). Allow desc with a missing summary for compatibility.
1710 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1711 $sub = Event::getHTML($ev, $simple_html);
1713 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1714 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1715 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1716 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1717 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1718 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1719 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1722 // Replace non graphical smilies for external posts
1723 if (!$nosmile && !$for_plaintext) {
1724 $text = Smilies::replace($text);
1727 if (!$for_plaintext) {
1728 if (in_array($simple_html, [7, 9])) {
1729 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1730 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1733 $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1734 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1737 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1739 // Remove all hashtag addresses
1740 if ($simple_html && !in_array($simple_html, [3, 7, 9])) {
1741 $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1742 } elseif ($simple_html == 3) {
1743 // The ! is converted to @ since Diaspora only understands the @
1744 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1745 '@<a href="$2">$3</a>',
1747 } elseif (in_array($simple_html, [7, 9])) {
1748 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1749 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1751 } elseif (!$simple_html) {
1752 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1753 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1757 // Bookmarks in red - will be converted to bookmarks in friendica
1758 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1759 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1760 $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1761 "[bookmark=$1]$2[/bookmark]", $text);
1763 if (in_array($simple_html, [2, 6, 7, 8])) {
1764 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1765 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1766 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1769 if ($simple_html == 5) {
1770 $text = preg_replace("/[^#@!]\[url\=(.*?)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1773 // Perform URL Search
1775 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1778 if ($simple_html == 5) {
1779 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1781 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1784 // Handle Diaspora posts
1785 $text = preg_replace_callback(
1786 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1788 return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1792 $text = preg_replace_callback(
1793 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1795 return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1799 // Server independent link to posts and comments
1800 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1801 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1802 $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
1806 * - #[url=<anything>]<term>[/url]
1807 * - [url=<anything>]#<term>[/url]
1809 $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) {
1811 . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1812 . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
1813 . XML::escape($matches[1])
1817 // We need no target="_blank" rel="noopener noreferrer" for local links
1818 // convert links start with DI::baseUrl() as local link without the target="_blank" rel="noopener noreferrer" attribute
1819 $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
1820 $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1821 $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1823 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1824 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1826 // Red compatibility, though the link can't be authenticated on Friendica
1827 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1830 // we may need to restrict this further if it picks up too many strays
1831 // link acct:user@host to a webfinger profile redirector
1833 $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);
1835 // Perform MAIL Search
1836 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1837 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1839 // Unhide all [noparse] contained bbtags unspacefying them
1840 // and triming the [noparse] tag.
1842 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1843 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1844 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1846 /// @todo What is the meaning of these lines?
1847 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1848 $text = preg_replace('/\&\#039\;/', '\'', $text);
1850 // Currently deactivated, it made problems with " inside of alt texts.
1851 //$text = preg_replace('/\"\;/', '"', $text);
1853 // fix any escaped ampersands that may have been converted into links
1854 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1856 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1857 $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:'];
1859 array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
1861 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1862 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $text);
1864 // sanitize href attributes (only whitelisted protocols URLs)
1865 // default value for backward compatibility
1866 $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
1868 // Always allowed protocol even if config isn't set or not including it
1869 $allowed_link_protocols[] = '//';
1870 $allowed_link_protocols[] = 'http://';
1871 $allowed_link_protocols[] = 'https://';
1872 $allowed_link_protocols[] = 'redir/';
1874 array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
1876 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1877 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
1880 $text = self::convertShare(
1882 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1883 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1888 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1891 // Restore code blocks
1892 $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
1893 function ($matches) use ($codeblocks) {
1894 $return = $matches[0];
1895 if (isset($codeblocks[intval($matches[1])])) {
1896 $return = $codeblocks[$matches[1]];
1903 // Clean up the HTML by loading and saving the HTML with the DOM.
1904 // Bad structured html can break a whole page.
1905 // For performance reasons do it only with activated item cache or at export.
1906 if (!$try_oembed || (get_itemcachepath() != '')) {
1907 $doc = new DOMDocument();
1908 $doc->preserveWhiteSpace = false;
1910 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1912 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1913 $encoding = '<?xml encoding="UTF-8">';
1914 @$doc->loadHTML($encoding . $doctype . '<html><body>' . $text . '</body></html>');
1915 $doc->encoding = 'UTF-8';
1916 $text = $doc->saveHTML();
1917 $text = str_replace(['<html><body>', '</body></html>', $doctype, $encoding], ['', '', '', ''], $text);
1919 $text = str_replace('<br></li>', '</li>', $text);
1921 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1924 // Clean up some useless linebreaks in lists
1925 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1926 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1927 //$Text = str_replace('</li><br />', '</li>', $Text);
1928 //$Text = str_replace('<br /><li>', '<li>', $Text);
1929 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1931 Hook::callAll('bbcode', $text);
1937 * Strips the "abstract" tag from the provided text
1939 * @param string $text The text with BBCode
1940 * @return string The same text - but without "abstract" element
1942 public static function stripAbstract($text)
1944 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1945 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1951 * Returns the value of the "abstract" element
1953 * @param string $text The text that maybe contains the element
1954 * @param string $addon The addon for which the abstract is meant for
1955 * @return string The abstract
1957 public static function getAbstract($text, $addon = '')
1961 $addon = strtolower($addon);
1963 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1964 foreach ($results AS $result) {
1965 $abstracts[strtolower($result[1])] = $result[2];
1969 if (isset($abstracts[$addon])) {
1970 $abstract = $abstracts[$addon];
1973 if ($abstract == '' && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1974 $abstract = $result[1];
1981 * Callback function to replace a Friendica style mention in a mention for Diaspora
1983 * @param array $match Matching values for the callback
1984 * [1] = Mention type (! or @)
1987 * @return string Replaced mention
1988 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1989 * @throws \ImagickException
1991 private static function bbCodeMention2DiasporaCallback($match)
1993 $contact = Contact::getDetailsByURL($match[3]);
1995 if (empty($contact['addr'])) {
1996 $contact = Probe::uri($match[3]);
1999 if (empty($contact['addr'])) {
2003 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
2008 * Converts a BBCode text into Markdown
2010 * This function converts a BBCode item body to be sent to Markdown-enabled
2011 * systems like Diaspora and Libertree
2013 * @param string $text
2014 * @param bool $for_diaspora Diaspora requires more changes than Libertree
2016 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2018 public static function toMarkdown($text, $for_diaspora = true)
2020 $original_text = $text;
2022 // Since Diaspora is creating a summary for links, this function removes them before posting
2023 if ($for_diaspora) {
2024 $text = self::removeShareInformation($text);
2028 * Transform #tags, strip off the [url] and replace spaces with underscore
2030 $url_search_string = "^\[\]";
2031 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
2032 function ($matches) {
2033 return '#' . str_replace(' ', '_', $matches[2]);
2038 // Converting images with size parameters to simple images. Markdown doesn't know it.
2039 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2041 // Convert it to HTML - don't try oembed
2042 if ($for_diaspora) {
2043 $text = self::convert($text, false, 3);
2045 // Add all tags that maybe were removed
2046 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2048 foreach ($tags[2] as $tag) {
2049 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2050 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2051 $tagline .= '#' . $tag . ' ';
2054 $text = $text . " " . $tagline;
2057 $text = self::convert($text, false, 4);
2060 // If a link is followed by a quote then there should be a newline before it
2061 // Maybe we should make this newline at every time before a quote.
2062 $text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
2064 $stamp1 = microtime(true);
2066 // Now convert HTML to Markdown
2067 $text = HTML::toMarkdown($text);
2069 DI::profiler()->saveTimestamp($stamp1, "parser", System::callstack());
2071 // Libertree has a problem with escaped hashtags.
2072 $text = str_replace(['\#'], ['#'], $text);
2074 // Remove any leading or trailing whitespace, as this will mess up
2075 // the Diaspora signature verification and cause the item to disappear
2076 $text = trim($text);
2078 if ($for_diaspora) {
2079 $url_search_string = "^\[\]";
2080 $text = preg_replace_callback(
2081 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
2082 ['self', 'bbCodeMention2DiasporaCallback'],
2087 Hook::callAll('bb2diaspora', $text);
2093 * Pull out all #hashtags and @person tags from $string.
2095 * We also get @person@domain.com - which would make
2096 * the regex quite complicated as tags can also
2097 * end a sentence. So we'll run through our results
2098 * and strip the period from any tags which end with one.
2099 * Returns array of tags found, or empty array.
2101 * @param string $string Post content
2103 * @return array List of tag and person names
2105 public static function getTags($string)
2109 // Convert hashtag links to hashtags
2110 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2 ', $string);
2112 // ignore anything in a code block
2113 $string = preg_replace('/\[code.*?\].*?\[\/code\]/sm', '', $string);
2115 // Force line feeds at bbtags
2116 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
2118 // ignore anything in a bbtag
2119 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
2121 // Match full names against @tags including the space between first and last
2122 // We will look these up afterward to see if they are full names or not recognisable.
2124 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
2125 foreach ($matches[1] as $match) {
2126 if (strstr($match, ']')) {
2127 // we might be inside a bbcode color tag - leave it alone
2131 if (substr($match, -1, 1) === '.') {
2132 $ret[] = substr($match, 0, -1);
2139 // Otherwise pull out single word tags. These can be @nickname, @first_last
2142 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
2143 foreach ($matches[1] as $match) {
2144 if (strstr($match, ']')) {
2145 // we might be inside a bbcode color tag - leave it alone
2149 if (substr($match, -1, 1) === '.') {
2150 $match = substr($match,0,-1);
2153 // ignore strictly numeric tags like #1
2154 if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
2158 // try not to catch url fragments
2159 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {