3 * @copyright Copyright (C) 2010-2021, the Friendica project
5 * @license GNU AGPL version 3 or any later version
7 * This program is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU Affero General Public License as
9 * published by the Free Software Foundation, either version 3 of the
10 * License, or (at your option) any later version.
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU Affero General Public License for more details.
17 * You should have received a copy of the GNU Affero General Public License
18 * along with this program. If not, see <https://www.gnu.org/licenses/>.
22 namespace Friendica\Content\Text;
27 use Friendica\Content\ContactSelector;
28 use Friendica\Content\Item;
29 use Friendica\Content\OEmbed;
30 use Friendica\Content\PageInfo;
31 use Friendica\Content\Smilies;
32 use Friendica\Core\Hook;
33 use Friendica\Core\Logger;
34 use Friendica\Core\Protocol;
35 use Friendica\Core\Renderer;
37 use Friendica\Model\Contact;
38 use Friendica\Model\Event;
39 use Friendica\Model\Photo;
40 use Friendica\Model\Tag;
41 use Friendica\Object\Image;
42 use Friendica\Protocol\Activity;
43 use Friendica\Util\Images;
44 use Friendica\Util\Map;
45 use Friendica\Util\ParseUrl;
46 use Friendica\Util\Proxy as ProxyUtils;
47 use Friendica\Util\Strings;
48 use Friendica\Util\XML;
52 // Update this value to the current date whenever changes are made to BBCode::convert
53 const VERSION = '2021-05-21';
63 const ACTIVITYPUB = 9;
65 const TOP_ANCHOR = '<br class="top-anchor">';
66 const BOTTOM_ANCHOR = '<br class="button-anchor">';
68 * Fetches attachment data that were generated the old way
70 * @param string $body Message body
72 * 'type' -> Message type ('link', 'video', 'photo')
73 * 'text' -> Text before the shared message
74 * 'after' -> Text after the shared message
75 * 'image' -> Preview image of the message
76 * 'url' -> Url to the attached message
77 * 'title' -> Title of the attachment
78 * 'description' -> Description of the attachment
79 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
81 private static function getOldAttachmentData($body)
85 // Simplify image codes
86 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
88 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
89 foreach ($attached as $data) {
90 if (!in_array($data[1], ['type-link', 'type-video', 'type-photo'])) {
94 $post['type'] = substr($data[1], 5);
96 $pos = strpos($body, $data[0]);
98 $post['text'] = trim(substr($body, 0, $pos));
99 $post['after'] = trim(substr($body, $pos + strlen($data[0])));
101 $post['text'] = trim(str_replace($data[0], '', $body));
105 $attacheddata = $data[2];
107 if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
109 $picturedata = Images::getInfoFromURLCached($matches[1]);
112 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
113 $post['image'] = $matches[1];
115 $post['preview'] = $matches[1];
120 if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
121 $post['url'] = $matches[1];
122 $post['title'] = $matches[2];
124 if (!empty($post['url']) && (in_array($post['type'], ['link', 'video']))
125 && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
126 $post['url'] = $matches[1];
129 // Search for description
130 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
131 $post['description'] = $matches[1];
139 * Fetches attachment data that were generated with the "attachment" element
141 * @param string $body Message body
143 * 'type' -> Message type ('link', 'video', 'photo')
144 * 'text' -> Text before the shared message
145 * 'after' -> Text after the shared message
146 * 'image' -> Preview image of the message
147 * 'url' -> Url to the attached message
148 * 'title' -> Title of the attachment
149 * 'description' -> Description of the attachment
150 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
152 public static function getAttachmentData($body)
162 'provider_name' => '',
163 'provider_url' => '',
168 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
169 return self::getOldAttachmentData($body);
172 $attributes = $match[2];
174 $data['text'] = trim($match[1]);
176 foreach (['type', 'url', 'title', 'image', 'preview', 'publisher_name', 'publisher_url', 'author_name', 'author_url'] as $field) {
177 preg_match('/' . preg_quote($field, '/') . '=("|\')(.*?)\1/ism', $attributes, $matches);
178 $value = $matches[2] ?? '';
182 case 'publisher_name':
183 $data['provider_name'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
185 case 'publisher_url':
186 $data['provider_url'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
189 $data['author_name'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
190 if ($data['provider_name'] == $data['author_name']) {
191 $data['author_name'] = '';
195 $data['author_url'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
196 if ($data['provider_url'] == $data['author_url']) {
197 $data['author_url'] = '';
201 $value = self::convert(html_entity_decode($value, ENT_QUOTES, 'UTF-8'), false, true);
202 $value = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
203 $value = str_replace(['[', ']'], ['[', ']'], $value);
204 $data['title'] = $value;
206 $data[$field] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
212 if (!in_array($data['type'], ['link', 'audio', 'photo', 'video'])) {
216 $data['description'] = trim($match[3]);
218 $data['after'] = trim($match[4]);
220 $parts = parse_url($data['url']);
221 if (!empty($parts['scheme']) && !empty($parts['host'])) {
222 if (empty($data['provider_name'])) {
223 $data['provider_name'] = $parts['host'];
225 if (empty($data['provider_url']) || empty(parse_url($data['provider_url'], PHP_URL_SCHEME))) {
226 $data['provider_url'] = $parts['scheme'] . '://' . $parts['host'];
228 if (!empty($parts['port'])) {
229 $data['provider_url'] .= ':' . $parts['port'];
237 public static function getAttachedData($body, $item = [])
241 - type: link, video, photo
249 $has_title = !empty($item['title']);
250 $plink = $item['plink'] ?? '';
251 $post = self::getAttachmentData($body);
253 // Get all linked images with alternative image description
254 if (preg_match_all("/\[img=([^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
255 foreach ($pictures as $picture) {
256 if (Photo::isLocal($picture[1])) {
257 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2]];
259 $post['remote_images'][] = ['url' => $picture[1], 'description' => $picture[2]];
262 if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
263 $post['image_description'] = $post['images'][0]['description'];
267 if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
268 foreach ($pictures as $picture) {
269 if (Photo::isLocal($picture[1])) {
270 $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => ''];
272 $post['remote_images'][] = ['url' => $picture[1], 'description' => ''];
277 // if nothing is found, it maybe having an image.
278 if (!isset($post['type'])) {
279 // Simplify image codes
280 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
281 $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
282 $post['text'] = $body;
284 if (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $body, $pictures, PREG_SET_ORDER)) {
285 if ((count($pictures) == 1) && !$has_title) {
286 if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
287 // Replace the preview picture with the real picture
288 $url = str_replace('-1.', '-0.', $pictures[0][2]);
289 $data = ['url' => $url, 'type' => 'photo'];
291 // Checking, if the link goes to a picture
292 $data = ParseUrl::getSiteinfoCached($pictures[0][1]);
296 // Sometimes photo posts to the own album are not detected at the start.
297 // So we seem to cannot use the cache for these cases. That's strange.
298 if (($data['type'] != 'photo') && strstr($pictures[0][1], "/photos/")) {
299 $data = ParseUrl::getSiteinfo($pictures[0][1]);
302 if ($data['type'] == 'photo') {
303 $post['type'] = 'photo';
304 if (isset($data['images'][0])) {
305 $post['image'] = $data['images'][0]['src'];
306 $post['url'] = $data['url'];
308 $post['image'] = $data['url'];
311 $post['preview'] = $pictures[0][2];
312 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
314 $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
315 if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
316 $post['type'] = 'photo';
317 $post['image'] = $pictures[0][1];
318 $post['preview'] = $pictures[0][2];
319 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
322 } elseif (count($pictures) > 0) {
323 $post['type'] = 'link';
324 $post['url'] = $plink;
325 $post['image'] = $pictures[0][2];
326 $post['text'] = $body;
328 foreach ($pictures as $picture) {
329 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
332 } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
334 $post['type'] = 'link';
335 $post['url'] = $plink;
337 $post['type'] = 'photo';
340 $post['image'] = $pictures[0][1];
341 $post['text'] = $body;
342 foreach ($pictures as $picture) {
343 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
347 // Test for the external links
348 preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER);
349 preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $links2, PREG_SET_ORDER);
351 $links = array_merge($links1, $links2);
353 // If there is only a single one, then use it.
354 // This should cover link posts via API.
355 if ((count($links) == 1) && !isset($post['preview']) && !$has_title) {
356 $post['type'] = 'link';
357 $post['url'] = $links[0][1];
360 // Simplify "video" element
361 $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
363 // Now count the number of external media links
364 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
365 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
366 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER);
367 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $links4, PREG_SET_ORDER);
369 // Add them to the other external links
370 $links = array_merge($links, $links1, $links2, $links3, $links4);
372 // Are there more than one?
373 if (count($links) > 1) {
374 // The post will be the type "text", which means a blog post
375 unset($post['type']);
376 $post['url'] = $plink;
379 if (!isset($post['type'])) {
380 $post['type'] = "text";
381 $post['text'] = trim($body);
384 if (($post['type'] == 'photo') && empty($post['images']) && !empty($post['remote_images'])) {
385 $post['images'] = $post['remote_images'];
386 $post['image'] = $post['images'][0]['url'];
387 if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
388 $post['image_description'] = $post['images'][0]['description'];
391 unset($post['remote_images']);
392 } elseif (isset($post['url']) && ($post['type'] == 'video')) {
393 $data = ParseUrl::getSiteinfoCached($post['url']);
395 if (isset($data['images'][0])) {
396 $post['image'] = $data['images'][0]['src'];
404 * Remove [attachment] BBCode and replaces it with a regular [url]
406 * @param string $body
407 * @param boolean $no_link_desc No link description
409 * @return string with replaced body
411 public static function removeAttachment($body, $no_link_desc = false)
413 return preg_replace_callback("/\s*\[attachment (.*?)\](.*?)\[\/attachment\]\s*/ism",
414 function ($match) use ($no_link_desc) {
415 $attach_data = self::getAttachmentData($match[0]);
416 if (empty($attach_data['url'])) {
418 } elseif (empty($attach_data['title']) || $no_link_desc) {
419 return "\n[url]" . $attach_data['url'] . "[/url]\n";
421 return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
427 * Converts a BBCode text into plaintext
430 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
434 public static function toPlaintext($text, $keep_urls = true)
436 // Remove pictures in advance to avoid unneeded proxy calls
437 $text = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $2 ', $text);
438 $text = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $text);
441 $text = self::removeAttachment($text);
443 $naked_text = HTML::toPlaintext(self::convert($text, false, 0, true), 0, !$keep_urls);
448 private static function proxyUrl($image, $simplehtml = self::INTERNAL)
450 // Only send proxied pictures to API and for internal display
451 if (in_array($simplehtml, [self::INTERNAL, self::API])) {
452 return ProxyUtils::proxifyUrl($image);
459 * This function changing the visual size (not the real size) of images.
460 * The function does not work for pictures with an alternate text description.
461 * This could only be changed by using some new "img" BBCode format.
463 * @param string $srctext The body with images
464 * @return string The body with possibly scaled images
466 public static function scaleExternalImages(string $srctext)
470 // Simplify image links
471 $s = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $s);
474 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
476 foreach ($matches as $mtch) {
477 Logger::info('scale_external_image', ['image' => $mtch[1]]);
479 $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
480 if (stristr($mtch[1], $hostname)) {
484 $curlResult = DI::httpRequest()->get($mtch[1]);
485 if (!$curlResult->isSuccess()) {
489 $i = $curlResult->getBody();
490 $type = $curlResult->getContentType();
491 $type = Images::getMimeTypeByData($i, $mtch[1], $type);
494 $Image = new Image($i, $type);
495 if ($Image->isValid()) {
496 $orig_width = $Image->getWidth();
497 $orig_height = $Image->getHeight();
499 if ($orig_width > 640 || $orig_height > 640) {
500 $Image->scaleDown(640);
501 $new_width = $Image->getWidth();
502 $new_height = $Image->getHeight();
503 Logger::info('External images scaled', ['orig_width' => $orig_width, 'new_width' => $new_width, 'orig_height' => $orig_height, 'new_height' => $new_height, 'match' => $mtch[0]]);
506 '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
510 Logger::info('New string', ['image' => $s]);
521 * Truncates imported message body string length to max_import_size
523 * The purpose of this function is to apply system message length limits to
524 * imported messages without including any embedded photos in the length
526 * @param string $body
528 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
530 public static function limitBodySize($body)
532 $maxlen = DI::config()->get('config', 'max_import_size', 0);
534 // If the length of the body, including the embedded images, is smaller
535 // than the maximum, then don't waste time looking for the images
536 if ($maxlen && (strlen($body) > $maxlen)) {
538 Logger::info('the total body length exceeds the limit', ['maxlen' => $maxlen, 'body_len' => strlen($body)]);
544 $img_start = strpos($orig_body, '[img');
545 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
546 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
547 while (($img_st_close !== false) && ($img_end !== false)) {
549 $img_st_close++; // make it point to AFTER the closing bracket
550 $img_end += $img_start;
551 $img_end += strlen('[/img]');
553 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
554 // This is an embedded image
556 if (($textlen + $img_start) > $maxlen) {
557 if ($textlen < $maxlen) {
558 Logger::info('the limit happens before an embedded image');
559 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
563 $new_body = $new_body . substr($orig_body, 0, $img_start);
564 $textlen += $img_start;
567 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
570 if (($textlen + $img_end) > $maxlen) {
571 if ($textlen < $maxlen) {
572 Logger::info('the limit happens before the end of a non-embedded image');
573 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
577 $new_body = $new_body . substr($orig_body, 0, $img_end);
578 $textlen += $img_end;
581 $orig_body = substr($orig_body, $img_end);
583 if ($orig_body === false) {
584 // in case the body ends on a closing image tag
588 $img_start = strpos($orig_body, '[img');
589 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
590 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
593 if (($textlen + strlen($orig_body)) > $maxlen) {
594 if ($textlen < $maxlen) {
595 Logger::info('the limit happens after the end of the last image');
596 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
599 Logger::info('the text size with embedded images extracted did not violate the limit');
600 $new_body = $new_body . $orig_body;
610 * Processes [attachment] tags
612 * Note: Can produce a [bookmark] tag in the returned string
614 * @param string $text
615 * @param integer $simplehtml
616 * @param bool $tryoembed
619 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
621 public static function convertAttachment($text, $simplehtml = self::INTERNAL, $tryoembed = true, array $data = [])
623 $data = $data ?: self::getAttachmentData($text);
624 if (empty($data) || empty($data['url'])) {
628 $stamp1 = microtime(true);
630 if (isset($data['title'])) {
631 $data['title'] = strip_tags($data['title']);
632 $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
634 $data['title'] = null;
637 if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
638 $data['preview'] = $data['image'];
644 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
645 $return = OEmbed::getHTML($data['url'], $data['title']);
647 throw new Exception('OEmbed is disabled for this attachment.');
649 } catch (Exception $e) {
650 $data['title'] = ($data['title'] ?? '') ?: $data['url'];
652 if ($simplehtml != self::CONNECTORS) {
653 $return = sprintf('<div class="type-%s">', $data['type']);
656 if (!empty($data['title']) && !empty($data['url'])) {
657 if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
658 $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), $data['title']);
660 if (!empty($data['image'])) {
661 $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), $data['title']);
662 } elseif (!empty($data['preview'])) {
663 $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']);
665 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
669 if (!empty($data['description']) && $data['description'] != $data['title']) {
671 $return .= sprintf('<blockquote>%s</blockquote>', trim(HTML::purify($data['description'])));
674 if (!empty($data['provider_url']) && !empty($data['provider_name'])) {
675 if (!empty($data['author_name'])) {
676 $return .= sprintf('<sup><a href="%s">%s (%s)</a></sup>', $data['provider_url'], $data['author_name'], $data['provider_name']);
678 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['provider_url'], $data['provider_name']);
682 if ($simplehtml != self::CONNECTORS) {
687 DI::profiler()->saveTimestamp($stamp1, 'rendering');
688 return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
691 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
693 $data = self::getAttachmentData($Text);
698 return $data['text'] . ($data['after'] ?? '');
701 $title = htmlentities($data['title'] ?? '', ENT_QUOTES, 'UTF-8', false);
702 $text = htmlentities($data['text'], ENT_QUOTES, 'UTF-8', false);
703 if ($plaintext || (($title != '') && strstr($text, $title))) {
704 $data['title'] = $data['url'];
705 } elseif (($text != '') && strstr($title, $text)) {
706 $data['text'] = $data['title'];
707 $data['title'] = $data['url'];
710 if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
711 return $data['title'] . $data['after'];
714 // If the link already is included in the post, don't add it again
715 if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
716 return $data['text'] . $data['after'];
719 $text = $data['text'];
721 if (!empty($data['url']) && !empty($data['title'])) {
722 $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
723 } elseif (!empty($data['url'])) {
724 $text .= "\n[url]" . $data['url'] . '[/url]';
727 return $text . "\n" . $data['after'];
731 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
733 * @param array $match Array with the matching values
734 * @return string reformatted link including HTML codes
736 private static function convertUrlForActivityPubCallback($match)
740 if (isset($match[2]) && ($match[1] != $match[2])) {
744 $parts = parse_url($url);
745 if (!isset($parts['scheme'])) {
749 return self::convertUrlForActivityPub($url);
753 * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
755 * @param string $url URL that is about to be reformatted
756 * @return string reformatted link including HTML codes
758 private static function convertUrlForActivityPub($url)
760 $html = '<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>';
761 return sprintf($html, $url, self::getStyledURL($url));
765 * Converts an URL in a nicer format (without the scheme and possibly shortened)
767 * @param string $url URL that is about to be reformatted
768 * @return string reformatted link
770 private static function getStyledURL($url)
772 $parts = parse_url($url);
773 $scheme = $parts['scheme'] . '://';
774 $styled_url = str_replace($scheme, '', $url);
776 if (strlen($styled_url) > 30) {
777 $styled_url = substr($styled_url, 0, 30) . "…";
784 * [noparse][i]italic[/i][/noparse] turns into
785 * [noparse][ i ]italic[ /i ][/noparse],
786 * to hide them from parser.
788 private static function escapeNoparseCallback($match)
790 $whole_match = $match[0];
791 $captured = $match[1];
792 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
793 $new_str = str_replace($captured, $spacefied, $whole_match);
798 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
799 * now turns back and the [noparse] tags are trimed
800 * returning [i]italic[/i]
802 private static function unescapeNoparseCallback($match)
804 $captured = $match[1];
805 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
810 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
813 * @param string $text Text to search
814 * @param string $name Tag name
815 * @param int $occurrences Number of first occurrences to skip
816 * @return boolean|array
818 public static function getTagPosition($text, $name, $occurrences = 0)
820 if ($occurrences < 0) {
825 for ($i = 0; $i <= $occurrences; $i++) {
826 if ($start_open !== false) {
827 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
831 if ($start_open === false) {
835 $start_equal = strpos($text, '=', $start_open);
836 $start_close = strpos($text, ']', $start_open);
838 if ($start_close === false) {
844 $end_open = strpos($text, '[/' . $name . ']', $start_close);
846 if ($end_open === false) {
852 'open' => $start_open,
853 'close' => $start_close
857 'close' => $end_open + strlen('[/' . $name . ']')
861 if ($start_equal !== false) {
862 $res['start']['equal'] = $start_equal + 1;
869 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
871 * @param string $pattern Preg pattern string
872 * @param string $replace Preg replace string
873 * @param string $name BBCode tag name
874 * @param string $text Text to search
877 public static function pregReplaceInTag($pattern, $replace, $name, $text)
880 $pos = self::getTagPosition($text, $name, $occurrences);
881 while ($pos !== false && $occurrences++ < 1000) {
882 $start = substr($text, 0, $pos['start']['open']);
883 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
884 $end = substr($text, $pos['end']['close']);
885 if ($end === false) {
889 $subject = preg_replace($pattern, $replace, $subject);
890 $text = $start . $subject . $end;
892 $pos = self::getTagPosition($text, $name, $occurrences);
898 private static function extractImagesFromItemBody($body)
905 $img_start = strpos($orig_body, '[img');
906 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
907 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
908 while (($img_st_close !== false) && ($img_end !== false)) {
909 $img_st_close++; // make it point to AFTER the closing bracket
910 $img_end += $img_start;
912 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
913 // This is an embedded image
914 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
915 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
919 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
922 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
924 if ($orig_body === false) {
925 // in case the body ends on a closing image tag
929 $img_start = strpos($orig_body, '[img');
930 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
931 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
934 $new_body = $new_body . $orig_body;
936 return ['body' => $new_body, 'images' => $saved_image];
939 private static function interpolateSavedImagesIntoItemBody($body, array $images)
944 foreach ($images as $image) {
945 // We're depending on the property of 'foreach' (specified on the PHP website) that
946 // it loops over the array starting from the first element and going sequentially
947 // to the last element
948 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
949 '<img src="' . self::proxyUrl($image) . '" alt="' . DI::l10n()->t('Image/photo') . '" />', $newbody);
958 * @param string $text A BBCode string
959 * @return array share attributes
961 public static function fetchShareAttributes($text)
963 // See Issue https://github.com/friendica/friendica/issues/10454
964 // Hashtags in usernames are expanded to links. This here is a quick fix.
965 $text = preg_replace('/([@!#])\[url\=.*?\](.*?)\[\/url\]/ism', '$1$2', $text);
968 if (!preg_match("/(.*?)\[share(.*?)\](.*)\[\/share\]/ism", $text, $matches)) {
972 $attribute_string = $matches[2];
973 foreach (['author', 'profile', 'avatar', 'link', 'posted', 'guid'] as $field) {
974 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
975 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
981 * This function converts a [share] block to text according to a provided callback function whose signature is:
983 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
986 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
988 * - $author_contact is a contact record array
989 * - $content is the inner content of the [share] block
990 * - $is_quote_share indicates whether there's any content before the [share] block
991 * - Return value is the string that should replace the [share] block in the provided text
993 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
995 * @param string $text A BBCode string
996 * @param callable $callback
997 * @return string The BBCode string with all [share] blocks replaced
999 public static function convertShare($text, callable $callback)
1001 $return = preg_replace_callback(
1002 "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
1003 function ($match) use ($callback) {
1004 $attribute_string = $match[2];
1006 foreach (['author', 'profile', 'avatar', 'link', 'posted', 'guid'] as $field) {
1007 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
1008 $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
1011 $author_contact = Contact::getByURL($attributes['profile'], false, ['id', 'url', 'addr', 'name', 'micro']);
1012 $author_contact['url'] = ($author_contact['url'] ?? $attributes['profile']);
1013 $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']);
1015 $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author'];
1016 $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
1017 $attributes['profile'] = ($author_contact['url'] ?? '') ?: $attributes['profile'];
1019 if (!empty($author_contact['id'])) {
1020 $attributes['avatar'] = Contact::getAvatarUrlForId($author_contact['id'], ProxyUtils::SIZE_THUMB);
1021 } elseif ($attributes['avatar']) {
1022 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], ProxyUtils::SIZE_THUMB);
1025 $content = preg_replace(Strings::autoLinkRegEx(), '<a href="$1">$1</a>', $match[3]);
1027 return $match[1] . $callback($attributes, $author_contact, $content, trim($match[1]) != '');
1036 * Default [share] tag conversion callback
1038 * Note: Can produce a [bookmark] tag in the output
1040 * @see BBCode::convertShare()
1041 * @param array $attributes [share] block attribute values
1042 * @param array $author_contact Contact row of the shared author
1043 * @param string $content Inner content of the [share] block
1044 * @param boolean $is_quote_share Whether there is content before the [share] block
1045 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
1047 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1049 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
1051 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
1053 switch ($simplehtml) {
1055 $text = ($is_quote_share? '<br>' : '') .
1056 '<p><b><a href="' . $attributes['link'] . '">' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . "</a>:</b> </p>\n" .
1057 '<blockquote class="shared_content">' . $content . '</blockquote>';
1059 case self::DIASPORA:
1060 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
1061 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
1063 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
1065 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
1066 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
1069 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
1071 if (empty($attributes['posted']) && !empty($attributes['link'])) {
1072 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
1077 case self::CONNECTORS:
1078 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
1079 $headline .= DI::l10n()->t('<a href="%1$s" target="_blank" rel="noopener noreferrer">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
1080 $headline .= ':</b></p>' . "\n";
1082 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
1086 $text = ($is_quote_share? '<br>' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
1088 case self::ACTIVITYPUB:
1089 $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>';
1090 $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
1093 $text = ($is_quote_share? "\n" : '');
1095 $contact = Contact::getByURL($attributes['profile'], false, ['network']);
1096 $network = $contact['network'] ?? Protocol::PHANTOM;
1098 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1099 $text .= Renderer::replaceMacros($tpl, [
1100 '$profile' => $attributes['profile'],
1101 '$avatar' => $attributes['avatar'],
1102 '$author' => $attributes['author'],
1103 '$link' => $attributes['link'],
1104 '$link_title' => DI::l10n()->t('Link to source'),
1105 '$posted' => $attributes['posted'],
1106 '$guid' => $attributes['guid'],
1107 '$network_name' => ContactSelector::networkToName($network, $attributes['profile']),
1108 '$network_icon' => ContactSelector::networkToIcon($network, $attributes['profile']),
1109 '$content' => self::TOP_ANCHOR . self::setMentions(trim($content), 0, $network) . self::BOTTOM_ANCHOR,
1117 private static function removePictureLinksCallback($match)
1119 $cache_key = 'remove:' . $match[1];
1120 $text = DI::cache()->get($cache_key);
1122 if (is_null($text)) {
1123 $curlResult = DI::httpRequest()->head($match[1], ['timeout' => DI::config()->get('system', 'xrd_timeout')]);
1124 if ($curlResult->isSuccess()) {
1125 $mimetype = $curlResult->getHeader('Content-Type');
1130 if (substr($mimetype, 0, 6) == 'image/') {
1131 $text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
1133 $text = "[url=" . $match[2] . ']' . $match[2] . "[/url]";
1135 // if its not a picture then look if its a page that contains a picture link
1136 $body = DI::httpRequest()->fetch($match[1]);
1138 DI::cache()->set($cache_key, $text);
1142 $doc = new DOMDocument();
1143 @$doc->loadHTML($body);
1144 $xpath = new DOMXPath($doc);
1145 $list = $xpath->query("//meta[@name]");
1146 foreach ($list as $node) {
1149 if ($node->attributes->length) {
1150 foreach ($node->attributes as $attribute) {
1151 $attr[$attribute->name] = $attribute->value;
1155 if (strtolower($attr['name']) == 'twitter:image') {
1156 $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
1160 DI::cache()->set($cache_key, $text);
1166 private static function expandLinksCallback($match)
1168 if (($match[3] == '') || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1169 return ($match[1] . "[url]" . $match[2] . "[/url]");
1171 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1175 private static function cleanPictureLinksCallback($match)
1177 // When the picture link is the own photo path then we can avoid fetching the link
1178 $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
1179 if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
1180 if (!empty($match[3])) {
1181 $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
1183 $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
1188 $cache_key = 'clean:' . $match[1];
1189 $text = DI::cache()->get($cache_key);
1190 if (!is_null($text)) {
1194 $curlResult = DI::httpRequest()->head($match[1], ['timeout' => DI::config()->get('system', 'xrd_timeout')]);
1195 if ($curlResult->isSuccess()) {
1196 $mimetype = $curlResult->getHeader('Content-Type');
1201 // if its a link to a picture then embed this picture
1202 if (substr($mimetype, 0, 6) == 'image/') {
1203 $text = '[img]' . $match[1] . '[/img]';
1205 if (!empty($match[3])) {
1206 $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
1208 $text = '[img]' . $match[2] . '[/img]';
1211 // if its not a picture then look if its a page that contains a picture link
1212 $body = DI::httpRequest()->fetch($match[1]);
1214 DI::cache()->set($cache_key, $text);
1218 $doc = new DOMDocument();
1219 @$doc->loadHTML($body);
1220 $xpath = new DOMXPath($doc);
1221 $list = $xpath->query("//meta[@name]");
1222 foreach ($list as $node) {
1224 if ($node->attributes->length) {
1225 foreach ($node->attributes as $attribute) {
1226 $attr[$attribute->name] = $attribute->value;
1230 if (strtolower($attr['name']) == "twitter:image") {
1231 if (!empty($match[3])) {
1232 $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
1234 $text = "[img]" . $attr['content'] . "[/img]";
1239 DI::cache()->set($cache_key, $text);
1244 public static function cleanPictureLinks($text)
1246 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1247 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $return);
1251 public static function removeLinks(string $bbcode)
1253 $bbcode = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $1 ', $bbcode);
1254 $bbcode = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $bbcode);
1256 $bbcode = preg_replace('/[@!#]\[url\=.*?\].*?\[\/url\]/ism', '', $bbcode);
1257 $bbcode = preg_replace("/\[url=[^\[\]]*\](.*)\[\/url\]/Usi", ' $1 ', $bbcode);
1258 $bbcode = preg_replace('/[@!#]?\[url.*?\[\/url\]/ism', '', $bbcode);
1263 * Converts a BBCode message to HTML message
1265 * BBcode 2 HTML was written by WAY2WEB.net
1266 * extended to work with Mistpark/Friendica - Mike Macgirvin
1268 * Simple HTML values meaning:
1269 * - 0: Friendica display
1271 * - 2: Used for Windows Phone push, Friendica API
1272 * - 3: Used before converting to Markdown in bb2diaspora.php
1273 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1276 * - 7: Used for dfrn, OStatus
1277 * - 8: Used for Twitter, WP backlink text setting
1280 * @param string $text
1281 * @param bool $try_oembed
1282 * @param int $simple_html
1283 * @param bool $for_plaintext
1285 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1287 public static function convert(string $text = null, $try_oembed = true, $simple_html = self::INTERNAL, $for_plaintext = false)
1289 // Accounting for null default column values
1290 if (is_null($text) || $text === '') {
1294 Hook::callAll('bbcode', $text);
1298 $text = self::performWithEscapedTags($text, ['code'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
1299 $text = self::performWithEscapedTags($text, ['noparse', 'nobb', 'pre'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
1301 * preg_match_callback function to replace potential Oembed tags with Oembed content
1303 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1305 * $match[2] = $title or absent
1307 $try_oembed_callback = function ($match)
1310 $title = $match[2] ?? null;
1313 $return = OEmbed::getHTML($url, $title);
1314 } catch (Exception $ex) {
1315 $return = $match[0];
1321 // Remove the abstract element. It is a non visible element.
1322 $text = self::stripAbstract($text);
1324 // Line ending normalisation
1325 $text = str_replace("\r\n", "\n", $text);
1327 // Move new lines outside of tags
1328 $text = preg_replace("#\[(\w*)](\n*)#ism", '$2[$1]', $text);
1329 $text = preg_replace("#(\n*)\[/(\w*)]#ism", '[/$2]$1', $text);
1331 // Extract the private images which use data urls since preg has issues with
1332 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1333 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1335 $extracted = self::extractImagesFromItemBody($text);
1336 $text = $extracted['body'];
1337 $saved_image = $extracted['images'];
1339 // If we find any event code, turn it into an event.
1340 // After we're finished processing the bbcode we'll
1341 // replace all of the event code with a reformatted version.
1343 $ev = Event::fromBBCode($text);
1345 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1346 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1348 $text = str_replace("<", "<", $text);
1349 $text = str_replace(">", ">", $text);
1351 // remove some newlines before the general conversion
1352 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1353 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1355 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1357 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1360 // Remove linefeeds inside of the table elements. See issue #6799
1361 $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
1362 "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
1363 "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
1364 "\n[hr]", "[hr]\n", " [hr]", "[hr] ",
1365 "\n[attachment ", " [attachment ", "\n[/attachment]", "[/attachment]\n", " [/attachment]", "[/attachment] ",
1366 "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "];
1367 $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
1368 "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
1369 "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
1370 "[hr]", "[hr]", "[hr]", "[hr]",
1371 "[attachment ", "[attachment ", "[/attachment]", "[/attachment]", "[/attachment]", "[/attachment]",
1372 "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"];
1375 $text = str_replace($search, $replace, $text);
1376 } while ($oldtext != $text);
1378 // Replace these here only once
1379 $search = ["\n[table]", "[/table]\n"];
1380 $replace = ["[table]", "[/table]"];
1381 $text = str_replace($search, $replace, $text);
1383 // Trim new lines regardless of the system.remove_multiplicated_lines config value
1384 $text = trim($text, "\n");
1386 // removing multiplicated newlines
1387 if (DI::config()->get('system', 'remove_multiplicated_lines')) {
1388 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[*]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1389 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1390 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[*]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1391 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1394 $text = str_replace($search, $replace, $text);
1395 } while ($oldtext != $text);
1398 // Add HTML new lines
1399 $text = str_replace("\n", '<br>', $text);
1401 /// @todo Have a closer look at the different html modes
1402 // Handle attached links or videos
1403 if ($simple_html == self::ACTIVITYPUB) {
1404 $text = self::removeAttachment($text);
1405 } elseif (!in_array($simple_html, [self::INTERNAL, self::EXTERNAL, self::CONNECTORS])) {
1406 $text = self::removeAttachment($text, true);
1408 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1411 $nosmile = strpos($text, '[nosmile]') !== false;
1412 $text = str_replace('[nosmile]', '', $text);
1414 // Replace non graphical smilies for external posts
1416 $text = self::performWithEscapedTags($text, ['img'], function ($text) use ($simple_html, $for_plaintext) {
1417 return Smilies::replace($text, ($simple_html != self::INTERNAL) || $for_plaintext);
1421 // leave open the posibility of [map=something]
1422 // this is replaced in Item::prepareBody() which has knowledge of the item location
1423 if (strpos($text, '[/map]') !== false) {
1424 $text = preg_replace_callback(
1425 "/\[map\](.*?)\[\/map\]/ism",
1426 function ($match) use ($simple_html) {
1427 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1433 if (strpos($text, '[map=') !== false) {
1434 $text = preg_replace_callback(
1435 "/\[map=(.*?)\]/ism",
1436 function ($match) use ($simple_html) {
1437 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1443 if (strpos($text, '[map]') !== false) {
1444 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1447 // Check for headers
1448 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1449 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1450 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1451 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1452 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1453 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1455 // Check for paragraph
1456 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1458 // Check for bold text
1459 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1461 // Check for Italics text
1462 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1464 // Check for Underline text
1465 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1467 // Check for strike-through text
1468 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1470 // Check for over-line text
1471 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1473 // Check for colored text
1474 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1476 // Check for sized text
1477 // [size=50] --> font-size: 50px (with the unit).
1478 if ($simple_html != self::DIASPORA) {
1479 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", '<span style="font-size:$1px;line-height:normal;">$2</span>', $text);
1480 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", '<span style="font-size:$1;line-height:normal;">$2</span>', $text);
1482 // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
1483 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text);
1487 // Check for centered text
1488 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", '<div style="text-align:center;">$1</div>', $text);
1490 // Check for list text
1491 $text = str_replace("[*]", "<li>", $text);
1493 // Check for style sheet commands
1494 $text = preg_replace("(\[style=(.*?)\](.*?)\[\/style\])ism", '<span style="$1">$2</span>', $text);
1496 // Check for CSS classes
1497 $text = preg_replace("(\[class=(.*?)\](.*?)\[\/class\])ism", '<span class="$1">$2</span>', $text);
1499 // handle nested lists
1502 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1503 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1504 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1505 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1506 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1507 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1508 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1509 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1510 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1511 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1512 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1513 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1514 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1515 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1518 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1519 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1520 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1521 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1523 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1524 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1526 $text = str_replace('[hr]', '<hr />', $text);
1528 if (!$for_plaintext) {
1529 $text = self::performWithEscapedTags($text, ['url', 'img', 'audio', 'video', 'youtube', 'vimeo', 'share', 'attachment', 'iframe', 'bookmark'], function ($text) {
1530 return preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1534 // Check for font change text
1535 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1537 // Declare the format for [spoiler] layout
1538 $SpoilerLayout = '<details class="spoiler"><summary>' . DI::l10n()->t('Click to open/close') . '</summary>$1</details>';
1540 // Check for [spoiler] text
1541 // handle nested quotes
1543 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1544 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", $SpoilerLayout, $text);
1547 // Check for [spoiler=Title] text
1549 // handle nested quotes
1551 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1552 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1553 '<details class="spoiler"><summary>$1</summary>$2</details>',
1557 // Declare the format for [quote] layout
1558 $QuoteLayout = '<blockquote>$1</blockquote>';
1560 // Check for [quote] text
1561 // handle nested quotes
1563 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1564 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1567 // Check for [quote=Author] text
1569 $t_wrote = DI::l10n()->t('$1 wrote:');
1571 // handle nested quotes
1573 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1574 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1575 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1580 // [img=widthxheight]image source[/img]
1581 $text = preg_replace_callback(
1582 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1583 function ($matches) use ($simple_html) {
1584 if (strpos($matches[3], "data:image/") === 0) {
1588 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1589 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1594 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1595 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1597 $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/ism",
1598 function ($matches) use ($simple_html) {
1599 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1600 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1601 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '" title="' . $matches[2] . '">';
1606 // [img]pathtoimage[/img]
1607 $text = preg_replace_callback(
1608 "/\[img\](.*?)\[\/img\]/ism",
1609 function ($matches) use ($simple_html) {
1610 if (strpos($matches[1], "data:image/") === 0) {
1614 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1615 return "[img]" . $matches[1] . "[/img]";
1620 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1621 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1623 $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);
1624 $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>', $text);
1625 //$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);
1627 // Simplify "video" element
1628 $text = preg_replace('(\[video[^\]]*?\ssrc\s?=\s?([^\s\]]+)[^\]]*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
1631 // html5 video and audio
1632 $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism",
1633 '<video src="$1" controls width="100%" height="auto"><a href="$1">$1</a></video>', $text);
1635 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1636 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1638 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1639 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1640 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", '<audio src="$1" controls><a href="$1">$1</a></audio>', $text);
1642 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1643 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1644 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1645 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1648 // Backward compatibility, [iframe] support has been removed in version 2020.12
1649 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1651 // Youtube extensions
1653 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1654 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1655 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1658 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1659 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1660 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1663 $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);
1665 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1666 '<a href="https://www.youtube.com/watch?v=$1" target="_blank" rel="noopener noreferrer">https://www.youtube.com/watch?v=$1</a>', $text);
1670 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1671 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1674 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1675 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1678 $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);
1680 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1681 '<a href="https://vimeo.com/$1" target="_blank" rel="noopener noreferrer">https://vimeo.com/$1</a>', $text);
1685 $text = OEmbed::BBCode2HTML($text);
1687 // Avoid triple linefeeds through oembed
1688 $text = str_replace("<br style='clear:left'></span><br><br>", "<br style='clear:left'></span><br>", $text);
1690 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1691 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1692 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1693 // start which is always required). Allow desc with a missing summary for compatibility.
1695 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1696 $sub = Event::getHTML($ev, $simple_html);
1698 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1699 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1700 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1701 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1702 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1703 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1704 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1707 if (!$for_plaintext && DI::config()->get('system', 'big_emojis') && ($simple_html != self::DIASPORA)) {
1708 $conv = html_entity_decode(str_replace([' ', "\n", "\r"], '', $text));
1709 // Emojis are always 4 byte Unicode characters
1710 if (!empty($conv) && (strlen($conv) / mb_strlen($conv) == 4)) {
1711 $text = '<span style="font-size: xx-large; line-height: normal;">' . $text . '</span>';
1715 if (!$for_plaintext) {
1716 if (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1717 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1718 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1721 $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1722 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1725 // Remove all hashtag addresses
1726 if ($simple_html && !in_array($simple_html, [self::DIASPORA, self::OSTATUS, self::ACTIVITYPUB])) {
1727 $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1728 } elseif ($simple_html == self::DIASPORA) {
1729 // The ! is converted to @ since Diaspora only understands the @
1730 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1731 '@<a href="$2">$3</a>',
1733 } elseif (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1734 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1735 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1737 } elseif (!$simple_html) {
1738 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1739 '<bdi>$1<a href="$2" class="userinfo mention" title="$3">$3</a></bdi>',
1743 // Bookmarks in red - will be converted to bookmarks in friendica
1744 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1745 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1746 $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1747 "[bookmark=$1]$2[/bookmark]", $text);
1749 if (in_array($simple_html, [self::API, self::OSTATUS, self::TWITTER])) {
1750 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1751 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1752 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1755 // Perform URL Search
1757 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1760 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1762 // Handle Diaspora posts
1763 $text = preg_replace_callback(
1764 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1766 return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1770 $text = preg_replace_callback(
1771 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1773 return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1777 // Server independent link to posts and comments
1778 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1779 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1780 $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
1784 * - #[url=<anything>]<term>[/url]
1785 * - [url=<anything>]#<term>[/url]
1787 $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) use ($simple_html) {
1788 if ($simple_html == BBCode::ACTIVITYPUB) {
1789 return '<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1790 . '" data-tag="' . XML::escape($matches[1]) . '" rel="tag ugc">#'
1791 . XML::escape($matches[1]) . '</a>';
1793 return '#<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1794 . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
1795 . XML::escape($matches[1]) . '</a>';
1799 // We need no target="_blank" rel="noopener noreferrer" for local links
1800 // convert links start with DI::baseUrl() as local link without the target="_blank" rel="noopener noreferrer" attribute
1801 $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
1802 $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1803 $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1805 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1806 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1808 // Red compatibility, though the link can't be authenticated on Friendica
1809 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1812 // we may need to restrict this further if it picks up too many strays
1813 // link acct:user@host to a webfinger profile redirector
1815 $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', '<a href="' . DI::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
1817 // Perform MAIL Search
1818 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1819 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1821 /// @todo What is the meaning of these lines?
1822 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1823 $text = preg_replace('/\&\#039\;/', '\'', $text);
1825 // Currently deactivated, it made problems with " inside of alt texts.
1826 //$text = preg_replace('/\"\;/', '"', $text);
1828 // fix any escaped ampersands that may have been converted into links
1829 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1831 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1832 $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:'];
1834 array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
1836 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1837 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $text);
1839 // sanitize href attributes (only allowlisted protocols URLs)
1840 // default value for backward compatibility
1841 $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
1843 // Always allowed protocol even if config isn't set or not including it
1844 $allowed_link_protocols[] = '//';
1845 $allowed_link_protocols[] = 'http://';
1846 $allowed_link_protocols[] = 'https://';
1847 $allowed_link_protocols[] = 'redir/';
1849 array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
1851 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1852 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
1855 $text = self::convertShare(
1857 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1858 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1862 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1865 }); // Escaped noparse, nobb, pre
1867 // Remove escaping tags and replace new lines that remain
1868 $text = preg_replace_callback('/\[(noparse|nobb)](.*?)\[\/\1]/ism', function ($match) {
1869 return str_replace("\n", "<br>", $match[2]);
1872 // Additionally, [pre] tags preserve spaces
1873 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", function ($match) {
1874 return str_replace([' ', "\n"], [' ', "<br>"], htmlentities($match[1], ENT_NOQUOTES,'UTF-8'));
1880 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1881 function ($matches) {
1882 if (strpos($matches[2], "\n") !== false) {
1883 $return = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlentities(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
1885 $return = '<code>' . htmlentities($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
1893 // Default iframe allowed domains/path
1894 $allowedIframeDomains = [
1895 DI::baseUrl()->getHostname()
1896 . (DI::baseUrl()->getUrlPath() ? '/' . DI::baseUrl()->getUrlPath() : '')
1897 . '/oembed/', # The path part has to change with the source in Content\Oembed::iframe
1898 'www.youtube.com/embed/',
1899 'player.vimeo.com/video/',
1902 $allowedIframeDomains = array_merge(
1903 $allowedIframeDomains,
1904 DI::config()->get('system', 'allowed_oembed') ?
1905 explode(',', DI::config()->get('system', 'allowed_oembed'))
1909 $text = HTML::purify($text, $allowedIframeDomains);
1915 * Strips the "abstract" tag from the provided text
1917 * @param string $text The text with BBCode
1918 * @return string The same text - but without "abstract" element
1920 public static function stripAbstract($text)
1922 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1923 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1929 * Returns the value of the "abstract" element
1931 * @param string $text The text that maybe contains the element
1932 * @param string $addon The addon for which the abstract is meant for
1933 * @return string The abstract
1935 public static function getAbstract($text, $addon = '')
1939 $addon = strtolower($addon);
1941 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1942 foreach ($results AS $result) {
1943 $abstracts[strtolower($result[1])] = $result[2];
1947 if (isset($abstracts[$addon])) {
1948 $abstract = $abstracts[$addon];
1951 if ($abstract == '' && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1952 $abstract = $result[1];
1959 * Callback function to replace a Friendica style mention in a mention for Diaspora
1961 * @param array $match Matching values for the callback
1962 * [1] = Mention type (! or @)
1965 * @return string Replaced mention
1966 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1967 * @throws \ImagickException
1969 private static function bbCodeMention2DiasporaCallback($match)
1971 $contact = Contact::getByURL($match[3], false, ['addr']);
1972 if (empty($contact['addr'])) {
1976 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
1981 * Converts a BBCode text into Markdown
1983 * This function converts a BBCode item body to be sent to Markdown-enabled
1984 * systems like Diaspora and Libertree
1986 * @param string $text
1987 * @param bool $for_diaspora Diaspora requires more changes than Libertree
1989 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1991 public static function toMarkdown($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, self::DIASPORA);
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, self::CONNECTORS);
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");
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 BBCode::performWithEscapedTags($string, ['noparse', 'pre', 'code', 'img'], function ($string) use (&$ret) {
2083 // Convert hashtag links to hashtags
2084 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2 ', $string);
2086 // Force line feeds at bbtags
2087 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
2089 // ignore anything in a bbtag
2090 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
2092 // Match full names against @tags including the space between first and last
2093 // We will look these up afterward to see if they are full names or not recognisable.
2095 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
2096 foreach ($matches[1] as $match) {
2097 if (strstr($match, ']')) {
2098 // we might be inside a bbcode color tag - leave it alone
2102 if (substr($match, -1, 1) === '.') {
2103 $ret[] = substr($match, 0, -1);
2110 // Otherwise pull out single word tags. These can be @nickname, @first_last
2113 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?\']*[^\^ \x0D\x0A,;:?!\'.])/', $string, $matches)) {
2114 foreach ($matches[1] as $match) {
2115 if (strstr($match, ']')) {
2116 // we might be inside a bbcode color tag - leave it alone
2120 // try not to catch url fragments
2121 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
2130 return array_unique($ret);
2134 * Expand tags to URLs
2136 * @param string $body
2137 * @return string body with expanded tags
2139 public static function expandTags(string $body)
2141 return preg_replace_callback("/([!#@])([^\^ \x0D\x0A,;:?\']*[^\^ \x0D\x0A,;:?!\'.])/",
2143 switch ($match[1]) {
2146 $contact = Contact::getByURL($match[2]);
2147 if (!empty($contact)) {
2148 return $match[1] . '[url=' . $contact['url'] . ']' . $contact['name'] . '[/url]';
2150 return $match[1] . $match[2];
2154 return $match[1] . '[url=' . 'https://' . DI::baseUrl() . '/search?tag=' . $match[2] . ']' . $match[2] . '[/url]';
2160 * Perform a custom function on a text after having escaped blocks enclosed in the provided tag list.
2162 * @param string $text
2163 * @param array $tagList A list of tag names, e.g ['noparse', 'nobb', 'pre']
2164 * @param callable $callback
2167 *@see Strings::performWithEscapedBlocks
2170 public static function performWithEscapedTags(string $text, array $tagList, callable $callback)
2172 $tagList = array_map('preg_quote', $tagList);
2174 return Strings::performWithEscapedBlocks($text, '#\[(?:' . implode('|', $tagList) . ').*?\[/(?:' . implode('|', $tagList) . ')]#ism', $callback);
2178 * Replaces mentions in the provided message body for the provided user and network if any
2181 * @param $profile_uid
2184 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2185 * @throws \ImagickException
2187 public static function setMentions($body, $profile_uid = 0, $network = '')
2189 BBCode::performWithEscapedTags($body, ['noparse', 'pre', 'code', 'img'], function ($body) use ($profile_uid, $network) {
2190 $tags = BBCode::getTags($body);
2195 foreach ($tags as $tag) {
2196 $tag_type = substr($tag, 0, 1);
2198 if ($tag_type == Tag::TAG_CHARACTER[Tag::HASHTAG]) {
2203 * If we already tagged 'Robert Johnson', don't try and tag 'Robert'.
2204 * Robert Johnson should be first in the $tags array
2206 foreach ($tagged as $nextTag) {
2207 if (stristr($nextTag, $tag . ' ')) {
2212 if (($success = Item::replaceTag($body, $inform, $profile_uid, $tag, $network)) && $success['replaced']) {
2224 * @param string $author Author display name
2225 * @param string $profile Author profile URL
2226 * @param string $avatar Author profile picture URL
2227 * @param string $link Post source URL
2228 * @param string $posted Post created date
2229 * @param string|null $guid Post guid (if any)
2231 * @TODO Rewrite to handle over whole record array
2233 public static function getShareOpeningTag(string $author, string $profile, string $avatar, string $link, string $posted, string $guid = null)
2235 $header = "[share author='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $author) .
2236 "' profile='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $profile) .
2237 "' avatar='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $avatar) .
2238 "' link='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $link) .
2239 "' posted='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $posted);
2242 $header .= "' guid='" . str_replace(["'", "[", "]"], ["'", "[", "]"], $guid);
2251 * Returns the BBCode relevant to embed the provided URL in a post body.
2252 * For media type, it will return [img], [video] and [audio] tags.
2253 * For regular web pages, it will either output a [bookmark] tag if title and description were provided,
2254 * an [attachment] tag or a simple [url] tag depending on $tryAttachment.
2256 * @param string $url
2257 * @param bool $tryAttachment
2258 * @param string|null $title
2259 * @param string|null $description
2260 * @param string|null $tags
2262 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2263 *@see ParseUrl::getSiteinfoCached
2266 public static function embedURL(string $url, bool $tryAttachment = true, string $title = null, string $description = null, string $tags = null): string
2268 DI::logger()->info($url);
2270 // If there is already some content information submitted we don't
2271 // need to parse the url for content.
2272 if (!empty($title) && !empty($description)) {
2273 $title = str_replace(["\r", "\n"], ['', ''], $title);
2275 $description = '[quote]' . trim($description) . '[/quote]' . "\n";
2278 if (!empty($tags)) {
2279 $arr_tags = ParseUrl::convertTagsToArray($tags);
2280 if (count($arr_tags)) {
2281 $str_tags = "\n" . implode(' ', $arr_tags) . "\n";
2285 $result = sprintf('[bookmark=%s]%s[/bookmark]%s', $url, ($title) ? $title : $url, $description) . $str_tags;
2287 DI::logger()->info('(unparsed): returns: ' . $result);
2292 $siteinfo = ParseUrl::getSiteinfoCached($url);
2294 if (in_array($siteinfo['type'], ['image', 'video', 'audio'])) {
2295 switch ($siteinfo['type']) {
2297 $bbcode = "\n" . '[video]' . $url . '[/video]' . "\n";
2300 $bbcode = "\n" . '[audio]' . $url . '[/audio]' . "\n";
2303 $bbcode = "\n" . '[img]' . $url . '[/img]' . "\n";
2310 unset($siteinfo['keywords']);
2312 // Bypass attachment if parse url for a comment
2313 if (!$tryAttachment) {
2314 return "\n" . '[url=' . $url . ']' . $siteinfo['title'] . '[/url]';
2317 // Format it as BBCode attachment
2318 return "\n" . PageInfo::getFooterFromData($siteinfo);