3 * @file src/Content/Text/BBCode.php
6 namespace Friendica\Content\Text;
11 use Friendica\BaseObject;
12 use Friendica\Content\OEmbed;
13 use Friendica\Content\Smilies;
14 use Friendica\Core\Cache;
15 use Friendica\Core\Config;
16 use Friendica\Core\Hook;
17 use Friendica\Core\L10n;
18 use Friendica\Core\Logger;
19 use Friendica\Core\Protocol;
20 use Friendica\Core\Renderer;
21 use Friendica\Core\System;
22 use Friendica\Model\Contact;
23 use Friendica\Model\Event;
24 use Friendica\Network\Probe;
25 use Friendica\Object\Image;
26 use Friendica\Util\Map;
27 use Friendica\Util\Network;
28 use Friendica\Util\ParseUrl;
29 use Friendica\Util\Proxy as ProxyUtils;
30 use Friendica\Util\Strings;
31 use Friendica\Util\XML;
33 class BBCode extends BaseObject
36 * @brief Fetches attachment data that were generated the old way
38 * @param string $body Message body
40 * 'type' -> Message type ("link", "video", "photo")
41 * 'text' -> Text before the shared message
42 * 'after' -> Text after the shared message
43 * 'image' -> Preview image of the message
44 * 'url' -> Url to the attached message
45 * 'title' -> Title of the attachment
46 * 'description' -> Description of the attachment
47 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
49 private static function getOldAttachmentData($body)
53 // Simplify image codes
54 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
56 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
57 foreach ($attached as $data) {
58 if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
62 $post["type"] = substr($data[1], 5);
64 $pos = strpos($body, $data[0]);
66 $post["text"] = trim(substr($body, 0, $pos));
67 $post["after"] = trim(substr($body, $pos + strlen($data[0])));
69 $post["text"] = trim(str_replace($data[0], "", $body));
73 $attacheddata = $data[2];
75 if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
77 $picturedata = Image::getInfoFromURL($matches[1]);
80 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
81 $post["image"] = $matches[1];
83 $post["preview"] = $matches[1];
88 if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
89 $post["url"] = $matches[1];
90 $post["title"] = $matches[2];
92 if (!empty($post["url"]) && (in_array($post["type"], ["link", "video"]))
93 && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
94 $post["url"] = $matches[1];
97 // Search for description
98 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
99 $post["description"] = $matches[1];
107 * @brief Fetches attachment data that were generated with the "attachment" element
109 * @param string $body Message body
111 * 'type' -> Message type ("link", "video", "photo")
112 * 'text' -> Text before the shared message
113 * 'after' -> Text after the shared message
114 * 'image' -> Preview image of the message
115 * 'url' -> Url to the attached message
116 * 'title' -> Title of the attachment
117 * 'description' -> Description of the attachment
118 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
120 public static function getAttachmentData($body)
124 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
125 return self::getOldAttachmentData($body);
128 $attributes = $match[2];
130 $data["text"] = trim($match[1]);
133 preg_match("/type='(.*?)'/ism", $attributes, $matches);
134 if (!empty($matches[1])) {
135 $type = strtolower($matches[1]);
138 preg_match('/type="(.*?)"/ism', $attributes, $matches);
139 if (!empty($matches[1])) {
140 $type = strtolower($matches[1]);
147 if (!in_array($type, ["link", "audio", "photo", "video"])) {
152 $data["type"] = $type;
156 preg_match("/url='(.*?)'/ism", $attributes, $matches);
157 if (!empty($matches[1])) {
161 preg_match('/url="(.*?)"/ism', $attributes, $matches);
162 if (!empty($matches[1])) {
167 $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
171 preg_match("/title='(.*?)'/ism", $attributes, $matches);
172 if (!empty($matches[1])) {
173 $title = $matches[1];
176 preg_match('/title="(.*?)"/ism', $attributes, $matches);
177 if (!empty($matches[1])) {
178 $title = $matches[1];
182 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
183 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
184 $title = str_replace(["[", "]"], ["[", "]"], $title);
185 $data["title"] = $title;
189 preg_match("/image='(.*?)'/ism", $attributes, $matches);
190 if (!empty($matches[1])) {
191 $image = $matches[1];
194 preg_match('/image="(.*?)"/ism', $attributes, $matches);
195 if (!empty($matches[1])) {
196 $image = $matches[1];
200 $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
204 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
205 if (!empty($matches[1])) {
206 $preview = $matches[1];
209 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
210 if (!empty($matches[1])) {
211 $preview = $matches[1];
214 if ($preview != "") {
215 $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
218 $data["description"] = trim($match[3]);
220 $data["after"] = trim($match[4]);
225 public static function getAttachedData($body, $item = [])
229 - type: link, video, photo
237 $has_title = !empty($item['title']);
238 $plink = defaults($item, 'plink', '');
239 $post = self::getAttachmentData($body);
241 // if nothing is found, it maybe having an image.
242 if (!isset($post["type"])) {
243 // Simplify image codes
244 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
246 $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
248 if (preg_match_all("(\[url=(.*?)\]\s*\[img\](.*?)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
249 if ((count($pictures) == 1) && !$has_title) {
250 if (!empty($item['object-type']) && ($item['object-type'] == ACTIVITY_OBJ_IMAGE)) {
251 // Replace the preview picture with the real picture
252 $url = str_replace('-1.', '-0.', $pictures[0][2]);
253 $data = ['url' => $url, 'type' => 'photo'];
255 // Checking, if the link goes to a picture
256 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
260 // Sometimes photo posts to the own album are not detected at the start.
261 // So we seem to cannot use the cache for these cases. That's strange.
262 if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
263 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
266 if ($data["type"] == "photo") {
267 $post["type"] = "photo";
268 if (isset($data["images"][0])) {
269 $post["image"] = $data["images"][0]["src"];
270 $post["url"] = $data["url"];
272 $post["image"] = $data["url"];
275 $post["preview"] = $pictures[0][2];
276 $post["text"] = trim(str_replace($pictures[0][0], "", $body));
278 $imgdata = Image::getInfoFromURL($pictures[0][1]);
279 if ($imgdata && substr($imgdata["mime"], 0, 6) == "image/") {
280 $post["type"] = "photo";
281 $post["image"] = $pictures[0][1];
282 $post["preview"] = $pictures[0][2];
283 $post["text"] = trim(str_replace($pictures[0][0], "", $body));
286 } elseif (count($pictures) > 0) {
287 $post["type"] = "link";
288 $post["url"] = $plink;
289 $post["image"] = $pictures[0][2];
290 $post["text"] = $body;
292 } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
293 if ((count($pictures) == 1) && !$has_title) {
294 $post["type"] = "photo";
295 $post["image"] = $pictures[0][1];
296 $post["text"] = str_replace($pictures[0][0], "", $body);
297 } elseif (count($pictures) > 0) {
298 $post["type"] = "link";
299 $post["url"] = $plink;
300 $post["image"] = $pictures[0][1];
301 $post["text"] = $body;
305 // Test for the external links
306 preg_match_all("(\[url\](.*?)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
307 preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
309 $links = array_merge($links1, $links2);
311 // If there is only a single one, then use it.
312 // This should cover link posts via API.
313 if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
314 $post["type"] = "link";
315 $post["text"] = trim($body);
316 $post["url"] = $links[0][1];
319 // Now count the number of external media links
320 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
321 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
322 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
323 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
325 // Add them to the other external links
326 $links = array_merge($links, $links1, $links2, $links3, $links4);
328 // Are there more than one?
329 if (count($links) > 1) {
330 // The post will be the type "text", which means a blog post
331 unset($post["type"]);
332 $post["url"] = $plink;
335 if (!isset($post["type"])) {
336 $post["type"] = "text";
337 $post["text"] = trim($body);
339 } elseif (isset($post["url"]) && ($post["type"] == "video")) {
340 $data = ParseUrl::getSiteinfoCached($post["url"], true);
342 if (isset($data["images"][0])) {
343 $post["image"] = $data["images"][0]["src"];
351 * @brief Converts a BBCode text into plaintext
354 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
358 public static function toPlaintext($text, $keep_urls = true)
360 $naked_text = preg_replace('/\[.+?\]/','', $text);
362 $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text);
368 private static function proxyUrl($image, $simplehtml = false)
370 // Only send proxied pictures to API and for internal display
371 if (in_array($simplehtml, [false, 2])) {
372 return ProxyUtils::proxifyUrl($image);
378 public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
380 // Suppress "view full size"
381 if (intval(Config::get('system', 'no_view_full_size'))) {
382 $include_link = false;
385 // Picture addresses can contain special characters
386 $s = htmlspecialchars_decode($srctext);
389 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
391 foreach ($matches as $mtch) {
392 Logger::log('scale_external_image: ' . $mtch[1]);
394 $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
395 if (stristr($mtch[1], $hostname)) {
399 // $scale_replace, if passed, is an array of two elements. The
400 // first is the name of the full-size image. The second is the
401 // name of a remote, scaled-down version of the full size image.
402 // This allows Friendica to display the smaller remote image if
403 // one exists, while still linking to the full-size image
404 if ($scale_replace) {
405 $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
409 $i = Network::fetchUrl($scaled);
414 // guess mimetype from headers or filename
415 $type = Image::guessType($mtch[1], true);
418 $Image = new Image($i, $type);
419 if ($Image->isValid()) {
420 $orig_width = $Image->getWidth();
421 $orig_height = $Image->getHeight();
423 if ($orig_width > 640 || $orig_height > 640) {
424 $Image->scaleDown(640);
425 $new_width = $Image->getWidth();
426 $new_height = $Image->getHeight();
427 Logger::log('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], Logger::DEBUG);
430 '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
431 . "\n" . (($include_link)
432 ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
436 Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG);
443 // replace the special char encoding
444 $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
449 * The purpose of this function is to apply system message length limits to
450 * imported messages without including any embedded photos in the length
452 * @brief Truncates imported message body string length to max_import_size
453 * @param string $body
455 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
457 public static function limitBodySize($body)
459 $maxlen = Config::get('config', 'max_import_size', 0);
461 // If the length of the body, including the embedded images, is smaller
462 // than the maximum, then don't waste time looking for the images
463 if ($maxlen && (strlen($body) > $maxlen)) {
465 Logger::log('the total body length exceeds the limit', Logger::DEBUG);
471 $img_start = strpos($orig_body, '[img');
472 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
473 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
474 while (($img_st_close !== false) && ($img_end !== false)) {
476 $img_st_close++; // make it point to AFTER the closing bracket
477 $img_end += $img_start;
478 $img_end += strlen('[/img]');
480 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
481 // This is an embedded image
483 if (($textlen + $img_start) > $maxlen) {
484 if ($textlen < $maxlen) {
485 Logger::log('the limit happens before an embedded image', Logger::DEBUG);
486 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
490 $new_body = $new_body . substr($orig_body, 0, $img_start);
491 $textlen += $img_start;
494 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
497 if (($textlen + $img_end) > $maxlen) {
498 if ($textlen < $maxlen) {
499 Logger::log('the limit happens before the end of a non-embedded image', Logger::DEBUG);
500 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
504 $new_body = $new_body . substr($orig_body, 0, $img_end);
505 $textlen += $img_end;
508 $orig_body = substr($orig_body, $img_end);
510 if ($orig_body === false) {
511 // in case the body ends on a closing image tag
515 $img_start = strpos($orig_body, '[img');
516 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
517 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
520 if (($textlen + strlen($orig_body)) > $maxlen) {
521 if ($textlen < $maxlen) {
522 Logger::log('the limit happens after the end of the last image', Logger::DEBUG);
523 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
526 Logger::log('the text size with embedded images extracted did not violate the limit', Logger::DEBUG);
527 $new_body = $new_body . $orig_body;
537 * Processes [attachment] tags
539 * Note: Can produce a [bookmark] tag in the returned string
541 * @brief Processes [attachment] tags
542 * @param string $return
543 * @param bool|int $simplehtml
544 * @param bool $tryoembed
546 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
548 private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
550 $data = self::getAttachmentData($return);
551 if (empty($data) || empty($data["url"])) {
555 if (isset($data["title"])) {
556 $data["title"] = strip_tags($data["title"]);
557 $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
559 $data["title"] = null;
562 if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data["image"])) {
563 $data["preview"] = $data["image"];
568 if (in_array($simplehtml, [7, 9])) {
569 $return = self::convertUrlForOStatus($data["url"]);
570 } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
571 $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data["url"], $data["title"]);
574 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
575 $return = OEmbed::getHTML($data['url'], $data['title']);
577 throw new Exception('OEmbed is disabled for this attachment.');
579 } catch (Exception $e) {
580 $data["title"] = defaults($data, 'title', $data['url']);
582 if ($simplehtml != 4) {
583 $return = sprintf('<div class="type-%s">', $data["type"]);
586 if (!empty($data['title']) && !empty($data['url'])) {
587 if (!empty($data["image"]) && empty($data["text"]) && ($data["type"] == "photo")) {
588 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
590 if (!empty($data["image"])) {
591 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
592 } elseif (!empty($data["preview"])) {
593 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], self::proxyUrl($data["preview"], $simplehtml), $data["title"]);
595 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
599 if (!empty($data["description"]) && $data["description"] != $data["title"]) {
600 // Sanitize the HTML by converting it to BBCode
601 $bbcode = HTML::toBBCode($data["description"]);
602 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
605 if (!empty($data['url'])) {
606 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
609 if ($simplehtml != 4) {
615 return trim(defaults($data, 'text', '') . ' ' . $return . ' ' . defaults($data, 'after', ''));
618 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
620 $data = self::getAttachmentData($Text);
625 return $data["text"] . defaults($data, 'after', '');
628 $title = htmlentities(defaults($data, 'title', ''), ENT_QUOTES, 'UTF-8', false);
629 $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
630 if ($plaintext || (($title != "") && strstr($text, $title))) {
631 $data["title"] = $data["url"];
632 } elseif (($text != "") && strstr($title, $text)) {
633 $data["text"] = $data["title"];
634 $data["title"] = $data["url"];
637 if (empty($data["text"]) && !empty($data["title"]) && empty($data["url"])) {
638 return $data["title"] . $data["after"];
641 // If the link already is included in the post, don't add it again
642 if (!empty($data["url"]) && strpos($data["text"], $data["url"])) {
643 return $data["text"] . $data["after"];
646 $text = $data["text"];
648 if (!empty($data["url"]) && !empty($data["title"])) {
649 $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
650 } elseif (!empty($data["url"])) {
651 $text .= "\n[url]" . $data["url"] . "[/url]";
654 return $text . "\n" . $data["after"];
658 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
660 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
661 * @param array $match Array with the matching values
662 * @return string reformatted link including HTML codes
664 private static function convertUrlForOStatusCallback($match)
668 if (isset($match[2]) && ($match[1] != $match[2])) {
672 $parts = parse_url($url);
673 if (!isset($parts['scheme'])) {
677 return self::convertUrlForOStatus($url);
681 * @brief Converts [url] BBCodes in a format that looks fine on OStatus systems.
682 * @param string $url URL that is about to be reformatted
683 * @return string reformatted link including HTML codes
685 private static function convertUrlForOStatus($url)
687 $parts = parse_url($url);
688 $scheme = $parts['scheme'] . '://';
689 $styled_url = str_replace($scheme, '', $url);
691 if (strlen($styled_url) > 30) {
692 $styled_url = substr($styled_url, 0, 30) . "…";
695 $html = '<a href="%s" target="_blank">%s</a>';
697 return sprintf($html, $url, $styled_url);
701 * [noparse][i]italic[/i][/noparse] turns into
702 * [noparse][ i ]italic[ /i ][/noparse],
703 * to hide them from parser.
705 private static function escapeNoparseCallback($match)
707 $whole_match = $match[0];
708 $captured = $match[1];
709 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
710 $new_str = str_replace($captured, $spacefied, $whole_match);
715 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
716 * now turns back and the [noparse] tags are trimed
717 * returning [i]italic[/i]
719 private static function unescapeNoparseCallback($match)
721 $captured = $match[1];
722 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
727 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
730 * @param string $text Text to search
731 * @param string $name Tag name
732 * @param int $occurrences Number of first occurrences to skip
733 * @return boolean|array
735 public static function getTagPosition($text, $name, $occurrences = 0)
737 if ($occurrences < 0) {
742 for ($i = 0; $i <= $occurrences; $i++) {
743 if ($start_open !== false) {
744 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
748 if ($start_open === false) {
752 $start_equal = strpos($text, '=', $start_open);
753 $start_close = strpos($text, ']', $start_open);
755 if ($start_close === false) {
761 $end_open = strpos($text, '[/' . $name . ']', $start_close);
763 if ($end_open === false) {
769 'open' => $start_open,
770 'close' => $start_close
774 'close' => $end_open + strlen('[/' . $name . ']')
778 if ($start_equal !== false) {
779 $res['start']['equal'] = $start_equal + 1;
786 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
788 * @param string $pattern Preg pattern string
789 * @param string $replace Preg replace string
790 * @param string $name BBCode tag name
791 * @param string $text Text to search
794 public static function pregReplaceInTag($pattern, $replace, $name, $text)
797 $pos = self::getTagPosition($text, $name, $occurrences);
798 while ($pos !== false && $occurrences++ < 1000) {
799 $start = substr($text, 0, $pos['start']['open']);
800 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
801 $end = substr($text, $pos['end']['close']);
802 if ($end === false) {
806 $subject = preg_replace($pattern, $replace, $subject);
807 $text = $start . $subject . $end;
809 $pos = self::getTagPosition($text, $name, $occurrences);
815 private static function extractImagesFromItemBody($body)
822 $img_start = strpos($orig_body, '[img');
823 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
824 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
825 while (($img_st_close !== false) && ($img_end !== false)) {
826 $img_st_close++; // make it point to AFTER the closing bracket
827 $img_end += $img_start;
829 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
830 // This is an embedded image
831 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
832 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
836 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
839 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
841 if ($orig_body === false) {
842 // in case the body ends on a closing image tag
846 $img_start = strpos($orig_body, '[img');
847 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
848 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
851 $new_body = $new_body . $orig_body;
853 return ['body' => $new_body, 'images' => $saved_image];
856 private static function interpolateSavedImagesIntoItemBody($body, array $images)
861 foreach ($images as $image) {
862 // We're depending on the property of 'foreach' (specified on the PHP website) that
863 // it loops over the array starting from the first element and going sequentially
864 // to the last element
865 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
866 '<img src="' . self::proxyUrl($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
874 * This function converts a [share] block to text according to a provided callback function whose signature is:
876 * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
879 * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
881 * - $author_contact is a contact record array
882 * - $content is the inner content of the [share] block
883 * - $is_quote_share indicates whether there's any content before the [share] block
884 * - Return value is the string that should replace the [share] block in the provided text
886 * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
888 * @param string $text A BBCode string
889 * @param callable $callback
890 * @return string The BBCode string with all [share] blocks replaced
892 public static function convertShare($text, callable $callback)
894 $return = preg_replace_callback(
895 "/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
896 function ($match) use ($callback) {
897 $attribute_string = $match[2];
900 foreach(['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
901 preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
902 $attributes[$field] = html_entity_decode(defaults($matches, 2, ''), ENT_QUOTES, 'UTF-8');
905 // We only call this so that a previously unknown contact can be added.
906 // This is important for the function "Model\Contact::getDetailsByURL()".
907 // This function then can fetch an entry from the contact table.
908 $default['url'] = $attributes['profile'];
910 if (!empty($attributes['author'])) {
911 $default['name'] = $attributes['author'];
914 if (!empty($attributes['avatar'])) {
915 $default['photo'] = $attributes['avatar'];
918 Contact::getIdForURL($attributes['profile'], 0, true, $default);
920 $author_contact = Contact::getDetailsByURL($attributes['profile']);
921 $author_contact['addr'] = defaults($author_contact, 'addr' , Protocol::getAddrFromProfileUrl($attributes['profile']));
923 $attributes['author'] = defaults($author_contact, 'name' , $attributes['author']);
924 $attributes['avatar'] = defaults($author_contact, 'micro', $attributes['avatar']);
925 $attributes['profile'] = defaults($author_contact, 'url' , $attributes['profile']);
927 if ($attributes['avatar']) {
928 $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
931 return $match[1] . $callback($attributes, $author_contact, $match[3], trim($match[1]) != '');
940 * Default [share] tag conversion callback
942 * Note: Can produce a [bookmark] tag in the output
944 * @see BBCode::convertShare()
945 * @param array $attributes [share] block attribute values
946 * @param array $author_contact Contact row of the shared author
947 * @param string $content Inner content of the [share] block
948 * @param boolean $is_quote_share Whether there is content before the [share] block
949 * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style
951 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
953 private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
955 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
957 switch ($simplehtml) {
959 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' <a href="' . $attributes['profile'] . '">' . $mention . '</a>: </p>' . "\n" . '«' . $content . '»';
962 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
965 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
966 $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
968 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
970 if (!empty($attributes['posted']) && !empty($attributes['link'])) {
971 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
974 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
976 if (empty($attributes['posted']) && !empty($attributes['link'])) {
977 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
983 $headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
984 $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
985 $headline .= ':</b></p>' . "\n";
987 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
991 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
993 case 7: // statusnet/GNU Social
994 case 9: // ActivityPub
995 $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
998 // Transforms quoted tweets in rich attachments to avoid nested tweets
999 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) {
1001 $text = ($is_quote_share? '<br />' : '') . OEmbed::getHTML($attributes['link']);
1002 } catch (Exception $e) {
1003 $text = ($is_quote_share? '<br />' : '') . sprintf('[bookmark=%s]%s[/bookmark]', $attributes['link'], $content);
1006 $text = ($is_quote_share? "\n" : '');
1008 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1009 $text .= Renderer::replaceMacros($tpl, [
1010 '$profile' => $attributes['profile'],
1011 '$avatar' => $attributes['avatar'],
1012 '$author' => $attributes['author'],
1013 '$link' => $attributes['link'],
1014 '$posted' => $attributes['posted'],
1015 '$content' => trim($content)
1024 private static function removePictureLinksCallback($match)
1026 $text = Cache::get($match[1]);
1028 if (is_null($text)) {
1029 $a = self::getApp();
1031 $stamp1 = microtime(true);
1033 $ch = @curl_init($match[1]);
1034 @curl_setopt($ch, CURLOPT_NOBODY, true);
1035 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1036 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1038 $curl_info = @curl_getinfo($ch);
1040 $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack());
1042 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1043 $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
1045 $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
1047 // if its not a picture then look if its a page that contains a picture link
1048 $body = Network::fetchUrl($match[1]);
1050 $doc = new DOMDocument();
1051 @$doc->loadHTML($body);
1052 $xpath = new DOMXPath($doc);
1053 $list = $xpath->query("//meta[@name]");
1054 foreach ($list as $node) {
1057 if ($node->attributes->length) {
1058 foreach ($node->attributes as $attribute) {
1059 $attr[$attribute->name] = $attribute->value;
1063 if (strtolower($attr["name"]) == "twitter:image") {
1064 $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
1068 Cache::set($match[1], $text);
1074 private static function expandLinksCallback($match)
1076 if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1077 return ($match[1] . "[url]" . $match[2] . "[/url]");
1079 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1083 private static function cleanPictureLinksCallback($match)
1085 $text = Cache::get($match[1]);
1087 if (is_null($text)) {
1088 $a = self::getApp();
1090 $stamp1 = microtime(true);
1092 $ch = @curl_init($match[1]);
1093 @curl_setopt($ch, CURLOPT_NOBODY, true);
1094 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1095 @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
1097 $curl_info = @curl_getinfo($ch);
1099 $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack());
1101 // if its a link to a picture then embed this picture
1102 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1103 $text = "[img]" . $match[1] . "[/img]";
1105 $text = "[img]" . $match[2] . "[/img]";
1107 // if its not a picture then look if its a page that contains a picture link
1108 $body = Network::fetchUrl($match[1]);
1110 $doc = new DOMDocument();
1111 @$doc->loadHTML($body);
1112 $xpath = new DOMXPath($doc);
1113 $list = $xpath->query("//meta[@name]");
1114 foreach ($list as $node) {
1116 if ($node->attributes->length) {
1117 foreach ($node->attributes as $attribute) {
1118 $attr[$attribute->name] = $attribute->value;
1122 if (strtolower($attr["name"]) == "twitter:image") {
1123 $text = "[img]" . $attr["content"] . "[/img]";
1127 Cache::set($match[1], $text);
1133 public static function cleanPictureLinks($text)
1135 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1140 * @brief Converts a BBCode message to HTML message
1142 * BBcode 2 HTML was written by WAY2WEB.net
1143 * extended to work with Mistpark/Friendica - Mike Macgirvin
1145 * Simple HTML values meaning:
1146 * - 0: Friendica display
1148 * - 2: Used for Windows Phone push, Friendica API
1149 * - 3: Used before converting to Markdown in bb2diaspora.php
1150 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1153 * - 7: Used for dfrn, OStatus
1154 * - 8: Used for WP backlink text setting
1157 * @param string $text
1158 * @param bool $try_oembed
1159 * @param int $simple_html
1160 * @param bool $for_plaintext
1162 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1164 public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false)
1166 $a = self::getApp();
1169 * preg_match_callback function to replace potential Oembed tags with Oembed content
1171 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1173 * $match[2] = $title or absent
1175 $try_oembed_callback = function ($match)
1178 $title = defaults($match, 2, null);
1181 $return = OEmbed::getHTML($url, $title);
1182 } catch (Exception $ex) {
1183 $return = $match[0];
1189 // Extracting code blocks before the whitespace processing and the autolinker
1192 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1193 function ($matches) use (&$codeblocks) {
1194 $return = '#codeblock-' . count($codeblocks) . '#';
1195 if (strpos($matches[2], "\n") !== false) {
1196 $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . trim($matches[2], "\n\r") . '</code></pre>';
1198 $codeblocks[] = '<code>' . $matches[2] . '</code>';
1206 // Hide all [noparse] contained bbtags by spacefying them
1207 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1209 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1210 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1211 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1213 // Remove the abstract element. It is a non visible element.
1214 $text = self::stripAbstract($text);
1216 // Move all spaces out of the tags
1217 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1218 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1220 // Extract the private images which use data urls since preg has issues with
1221 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1222 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1224 $extracted = self::extractImagesFromItemBody($text);
1225 $text = $extracted['body'];
1226 $saved_image = $extracted['images'];
1228 // If we find any event code, turn it into an event.
1229 // After we're finished processing the bbcode we'll
1230 // replace all of the event code with a reformatted version.
1232 $ev = Event::fromBBCode($text);
1234 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1235 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1237 $text = str_replace("<", "<", $text);
1238 $text = str_replace(">", ">", $text);
1240 // remove some newlines before the general conversion
1241 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1242 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1244 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1246 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1249 // Convert new line chars to html <br /> tags
1251 // nlbr seems to be hopelessly messed up
1252 // $Text = nl2br($Text);
1254 // We'll emulate it.
1256 $text = trim($text);
1257 $text = str_replace("\r\n", "\n", $text);
1259 // Remove linefeeds inside of the table elements. See issue #6799
1260 $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
1261 "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
1262 "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
1263 "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "];
1264 $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
1265 "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
1266 "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
1267 "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"];
1270 $text = str_replace($search, $replace, $text);
1271 } while ($oldtext != $text);
1273 // Replace these here only once
1274 $search = ["\n[table]", "[/table]\n"];
1275 $replace = ["[table]", "[/table]"];
1276 $text = str_replace($search, $replace, $text);
1278 // removing multiplicated newlines
1279 if (Config::get("system", "remove_multiplicated_lines")) {
1280 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1281 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1282 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1283 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1286 $text = str_replace($search, $replace, $text);
1287 } while ($oldtext != $text);
1290 // Handle attached links or videos
1291 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1293 // leave open the posibility of [map=something]
1294 // this is replaced in Item::prepareBody() which has knowledge of the item location
1295 if (strpos($text, '[/map]') !== false) {
1296 $text = preg_replace_callback(
1297 "/\[map\](.*?)\[\/map\]/ism",
1298 function ($match) use ($simple_html) {
1299 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1305 if (strpos($text, '[map=') !== false) {
1306 $text = preg_replace_callback(
1307 "/\[map=(.*?)\]/ism",
1308 function ($match) use ($simple_html) {
1309 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1315 if (strpos($text, '[map]') !== false) {
1316 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1319 // Check for headers
1320 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1321 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1322 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1323 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1324 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1325 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1327 // Check for paragraph
1328 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1330 // Check for bold text
1331 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1333 // Check for Italics text
1334 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1336 // Check for Underline text
1337 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1339 // Check for strike-through text
1340 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1342 // Check for over-line text
1343 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1345 // Check for colored text
1346 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1348 // Check for sized text
1349 // [size=50] --> font-size: 50px (with the unit).
1350 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1351 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1353 // Check for centered text
1354 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1356 // Check for list text
1357 $text = str_replace("[*]", "<li>", $text);
1359 // Check for style sheet commands
1360 $text = preg_replace_callback(
1361 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1363 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1368 // Check for CSS classes
1369 $text = preg_replace_callback(
1370 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1372 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1377 // handle nested lists
1380 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1381 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1382 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1383 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1384 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1385 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1386 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1387 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1388 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1389 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1390 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1391 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1392 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1393 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1396 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1397 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1398 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1399 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1401 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1402 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1404 $text = str_replace('[hr]', '<hr />', $text);
1406 // This is actually executed in Item::prepareBody()
1408 $nosmile = strpos($text, '[nosmile]') !== false;
1409 $text = str_replace('[nosmile]', '', $text);
1411 // Check for font change text
1412 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1414 // Declare the format for [spoiler] layout
1415 $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
1417 // Check for [spoiler] text
1418 // handle nested quotes
1420 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1421 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
1424 // Check for [spoiler=Author] text
1426 $t_wrote = L10n::t('$1 wrote:');
1428 // handle nested quotes
1430 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1431 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1432 "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
1436 // Declare the format for [quote] layout
1437 $QuoteLayout = '<blockquote>$1</blockquote>';
1439 // Check for [quote] text
1440 // handle nested quotes
1442 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1443 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1446 // Check for [quote=Author] text
1448 $t_wrote = L10n::t('$1 wrote:');
1450 // handle nested quotes
1452 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1453 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1454 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1459 // [img=widthxheight]image source[/img]
1460 $text = preg_replace_callback(
1461 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1462 function ($matches) use ($simple_html) {
1463 if (strpos($matches[3], "data:image/") === 0) {
1467 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1468 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1473 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1474 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1476 $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/ism",
1477 function ($matches) use ($simple_html) {
1478 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1479 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1480 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '">';
1485 // [img]pathtoimage[/img]
1486 $text = preg_replace_callback(
1487 "/\[img\](.*?)\[\/img\]/ism",
1488 function ($matches) use ($simple_html) {
1489 if (strpos($matches[1], "data:image/") === 0) {
1493 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1494 return "[img]" . $matches[1] . "[/img]";
1499 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1500 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1502 $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . L10n::t('Encrypted content') . '" /><br />', $text);
1503 $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $text);
1504 //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $Text);
1508 $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);
1509 $text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3).*?)\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
1511 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1512 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1514 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1515 '<a href="$1" target="_blank">$1</a>', $text);
1516 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1517 '<a href="$1" target="_blank">$1</a>', $text);
1520 // html5 video and audio
1524 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1526 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1529 // Youtube extensions
1531 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1532 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1533 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1536 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1537 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1538 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1541 $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);
1543 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1544 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1548 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1549 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1552 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1553 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1556 $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);
1558 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1559 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1563 $text = OEmbed::BBCode2HTML($text);
1565 // Avoid triple linefeeds through oembed
1566 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1568 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1569 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1570 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1571 // start which is always required). Allow desc with a missing summary for compatibility.
1573 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1574 $sub = Event::getHTML($ev, $simple_html);
1576 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1577 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1578 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1579 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1580 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1581 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1582 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1585 // Replace non graphical smilies for external posts
1586 if (!$nosmile && !$for_plaintext) {
1587 $text = Smilies::replace($text);
1590 // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text
1591 if (!$for_plaintext) {
1592 $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1593 if (in_array($simple_html, [7, 9])) {
1594 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1595 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1598 $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1599 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1602 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1604 // Remove all hashtag addresses
1605 if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7, 9])) {
1606 $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1607 } elseif ($simple_html == 3) {
1608 // The ! is converted to @ since Diaspora only understands the @
1609 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1610 '@<a href="$2">$3</a>',
1612 } elseif (in_array($simple_html, [7, 9])) {
1613 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1614 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1616 } elseif (!$simple_html) {
1617 $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1618 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1622 // Bookmarks in red - will be converted to bookmarks in friendica
1623 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1624 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1625 $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1626 "[bookmark=$1]$2[/bookmark]", $text);
1628 if (in_array($simple_html, [2, 6, 7, 8])) {
1629 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1630 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1631 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1634 if ($simple_html == 5) {
1635 $text = preg_replace("/[^#@!]\[url\=(.*?)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1638 // Perform URL Search
1640 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1643 if ($simple_html == 5) {
1644 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1646 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1649 // Handle Diaspora posts
1650 $text = preg_replace_callback(
1651 "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1653 return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1657 $text = preg_replace_callback(
1658 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1660 return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1664 // Server independent link to posts and comments
1665 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1666 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1667 $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
1671 * - #[url=<anything>]<term>[/url]
1672 * - [url=<anything>]#<term>[/url]
1674 $text = preg_replace_callback("/(?:#\[url\=.*?\]|\[url\=.*?\]#)(.*?)\[\/url\]/ism", function($matches) {
1676 . System::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1677 . '" class="tag" title="' . XML::escape($matches[1]) . '">'
1678 . XML::escape($matches[1])
1682 // We need no target="_blank" for local links
1683 // convert links start with System::baseUrl() as local link without the target="_blank" attribute
1684 $escapedBaseUrl = preg_quote(System::baseUrl(), '/');
1685 $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1686 $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1688 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1689 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1691 // Red compatibility, though the link can't be authenticated on Friendica
1692 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1695 // we may need to restrict this further if it picks up too many strays
1696 // link acct:user@host to a webfinger profile redirector
1698 $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="' . System::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
1700 // Perform MAIL Search
1701 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1702 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1704 // Unhide all [noparse] contained bbtags unspacefying them
1705 // and triming the [noparse] tag.
1707 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1708 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1709 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1711 /// @todo What is the meaning of these lines?
1712 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1713 $text = preg_replace('/\&\#039\;/', '\'', $text);
1715 // Currently deactivated, it made problems with " inside of alt texts.
1716 //$text = preg_replace('/\"\;/', '"', $text);
1718 // fix any escaped ampersands that may have been converted into links
1719 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1721 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1722 $allowed_src_protocols = ['http', 'redir', 'cid'];
1723 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1724 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
1726 // sanitize href attributes (only whitelisted protocols URLs)
1727 // default value for backward compatibility
1728 $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']);
1730 // Always allowed protocol even if config isn't set or not including it
1731 $allowed_link_protocols[] = 'http';
1732 $allowed_link_protocols[] = 'redir/';
1734 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1735 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
1738 $text = self::convertShare(
1740 function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1741 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1746 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1749 // Restore code blocks
1750 $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
1751 function ($matches) use ($codeblocks) {
1752 $return = $matches[0];
1753 if (isset($codeblocks[intval($matches[1])])) {
1754 $return = $codeblocks[$matches[1]];
1761 // Clean up the HTML by loading and saving the HTML with the DOM.
1762 // Bad structured html can break a whole page.
1763 // For performance reasons do it only with activated item cache or at export.
1764 if (!$try_oembed || (get_itemcachepath() != "")) {
1765 $doc = new DOMDocument();
1766 $doc->preserveWhiteSpace = false;
1768 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1770 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1771 $encoding = '<?xml encoding="UTF-8">';
1772 @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
1773 $doc->encoding = 'UTF-8';
1774 $text = $doc->saveHTML();
1775 $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
1777 $text = str_replace('<br></li>', '</li>', $text);
1779 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1782 // Clean up some useless linebreaks in lists
1783 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1784 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1785 //$Text = str_replace('</li><br />', '</li>', $Text);
1786 //$Text = str_replace('<br /><li>', '<li>', $Text);
1787 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1789 Hook::callAll('bbcode', $text);
1795 * @brief Strips the "abstract" tag from the provided text
1797 * @param string $text The text with BBCode
1798 * @return string The same text - but without "abstract" element
1800 public static function stripAbstract($text)
1802 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1803 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1809 * @brief Returns the value of the "abstract" element
1811 * @param string $text The text that maybe contains the element
1812 * @param string $addon The addon for which the abstract is meant for
1813 * @return string The abstract
1815 public static function getAbstract($text, $addon = "")
1819 $addon = strtolower($addon);
1821 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1822 foreach ($results AS $result) {
1823 $abstracts[strtolower($result[1])] = $result[2];
1827 if (isset($abstracts[$addon])) {
1828 $abstract = $abstracts[$addon];
1831 if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1832 $abstract = $result[1];
1839 * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
1841 * @param array $match Matching values for the callback
1842 * [1] = Mention type (! or @)
1845 * @return string Replaced mention
1846 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1847 * @throws \ImagickException
1849 private static function bbCodeMention2DiasporaCallback($match)
1851 $contact = Contact::getDetailsByURL($match[3]);
1853 if (empty($contact['addr'])) {
1854 $contact = Probe::uri($match[3]);
1857 if (empty($contact['addr'])) {
1861 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
1866 * @brief Converts a BBCode text into Markdown
1868 * This function converts a BBCode item body to be sent to Markdown-enabled
1869 * systems like Diaspora and Libertree
1871 * @param string $text
1872 * @param bool $for_diaspora Diaspora requires more changes than Libertree
1874 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1876 public static function toMarkdown($text, $for_diaspora = true)
1878 $a = self::getApp();
1880 $original_text = $text;
1882 // Since Diaspora is creating a summary for links, this function removes them before posting
1883 if ($for_diaspora) {
1884 $text = self::removeShareInformation($text);
1888 * Transform #tags, strip off the [url] and replace spaces with underscore
1890 $url_search_string = "^\[\]";
1891 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
1892 function ($matches) {
1893 return '#' . str_replace(' ', '_', $matches[2]);
1898 // Converting images with size parameters to simple images. Markdown doesn't know it.
1899 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
1901 // Convert it to HTML - don't try oembed
1902 if ($for_diaspora) {
1903 $text = self::convert($text, false, 3);
1905 // Add all tags that maybe were removed
1906 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
1908 foreach ($tags[2] as $tag) {
1909 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
1910 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
1911 $tagline .= '#' . $tag . ' ';
1914 $text = $text . " " . $tagline;
1917 $text = self::convert($text, false, 4);
1920 // mask some special HTML chars from conversation to markdown
1921 $text = str_replace(['<', '>', '&'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text);
1923 // If a link is followed by a quote then there should be a newline before it
1924 // Maybe we should make this newline at every time before a quote.
1925 $text = str_replace(["</a><blockquote>"], ["</a><br><blockquote>"], $text);
1927 $stamp1 = microtime(true);
1929 // Now convert HTML to Markdown
1930 $text = HTML::toMarkdown($text);
1932 // unmask the special chars back to HTML
1933 $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
1935 $a->getProfiler()->saveTimestamp($stamp1, "parser", System::callstack());
1937 // Libertree has a problem with escaped hashtags.
1938 $text = str_replace(['\#'], ['#'], $text);
1940 // Remove any leading or trailing whitespace, as this will mess up
1941 // the Diaspora signature verification and cause the item to disappear
1942 $text = trim($text);
1944 if ($for_diaspora) {
1945 $url_search_string = "^\[\]";
1946 $text = preg_replace_callback(
1947 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
1948 ['self', 'bbCodeMention2DiasporaCallback'],
1953 Hook::callAll('bb2diaspora', $text);
1959 * @brief Pull out all #hashtags and @person tags from $string.
1961 * We also get @person@domain.com - which would make
1962 * the regex quite complicated as tags can also
1963 * end a sentence. So we'll run through our results
1964 * and strip the period from any tags which end with one.
1965 * Returns array of tags found, or empty array.
1967 * @param string $string Post content
1969 * @return array List of tag and person names
1971 public static function getTags($string)
1975 // Convert hashtag links to hashtags
1976 $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2', $string);
1978 // ignore anything in a code block
1979 $string = preg_replace('/\[code.*?\].*?\[\/code\]/sm', '', $string);
1981 // Force line feeds at bbtags
1982 $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
1984 // ignore anything in a bbtag
1985 $string = preg_replace('/\[(.*?)\]/sm', '', $string);
1987 // Match full names against @tags including the space between first and last
1988 // We will look these up afterward to see if they are full names or not recognisable.
1990 if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
1991 foreach ($matches[1] as $match) {
1992 if (strstr($match, ']')) {
1993 // we might be inside a bbcode color tag - leave it alone
1997 if (substr($match, -1, 1) === '.') {
1998 $ret[] = substr($match, 0, -1);
2005 // Otherwise pull out single word tags. These can be @nickname, @first_last
2008 if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
2009 foreach ($matches[1] as $match) {
2010 if (strstr($match, ']')) {
2011 // we might be inside a bbcode color tag - leave it alone
2014 if (substr($match, -1, 1) === '.') {
2015 $match = substr($match,0,-1);
2017 // ignore strictly numeric tags like #1
2018 if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
2021 // try not to catch url fragments
2022 if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {