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\Addon;
15 use Friendica\Core\Cache;
16 use Friendica\Core\Config;
17 use Friendica\Core\L10n;
18 use Friendica\Core\Protocol;
19 use Friendica\Core\System;
20 use Friendica\Model\Contact;
21 use Friendica\Model\Event;
22 use Friendica\Network\Probe;
23 use Friendica\Object\Image;
24 use Friendica\Util\Map;
25 use Friendica\Util\Network;
26 use Friendica\Util\ParseUrl;
27 use Friendica\Util\Proxy as ProxyUtils;
28 use League\HTMLToMarkdown\HtmlConverter;
30 class BBCode extends BaseObject
33 * @brief Fetches attachment data that were generated the old way
35 * @param string $body Message body
37 * 'type' -> Message type ("link", "video", "photo")
38 * 'text' -> Text before the shared message
39 * 'after' -> Text after the shared message
40 * 'image' -> Preview image of the message
41 * 'url' -> Url to the attached message
42 * 'title' -> Title of the attachment
43 * 'description' -> Description of the attachment
45 private static function getOldAttachmentData($body)
49 // Simplify image codes
50 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
52 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
53 foreach ($attached as $data) {
54 if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
58 $post["type"] = substr($data[1], 5);
60 $pos = strpos($body, $data[0]);
62 $post["text"] = trim(substr($body, 0, $pos));
63 $post["after"] = trim(substr($body, $pos + strlen($data[0])));
65 $post["text"] = trim(str_replace($data[0], "", $body));
68 $attacheddata = $data[2];
70 $URLSearchString = "^\[\]";
72 if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
74 $picturedata = Image::getInfoFromURL($matches[1]);
77 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
78 $post["image"] = $matches[1];
80 $post["preview"] = $matches[1];
85 if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
86 $post["url"] = $matches[1];
87 $post["title"] = $matches[2];
89 if (!empty($post["url"] == "") && (in_array($post["type"], ["link", "video"]))
90 && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
91 $post["url"] = $matches[1];
94 // Search for description
95 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
96 $post["description"] = $matches[1];
104 * @brief Fetches attachment data that were generated with the "attachment" element
106 * @param string $body Message body
108 * 'type' -> Message type ("link", "video", "photo")
109 * 'text' -> Text before the shared message
110 * 'after' -> Text after the shared message
111 * 'image' -> Preview image of the message
112 * 'url' -> Url to the attached message
113 * 'title' -> Title of the attachment
114 * 'description' -> Description of the attachment
116 public static function getAttachmentData($body)
120 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
121 return self::getOldAttachmentData($body);
124 $attributes = $match[2];
126 $data["text"] = trim($match[1]);
129 preg_match("/type='(.*?)'/ism", $attributes, $matches);
130 if (x($matches, 1)) {
131 $type = strtolower($matches[1]);
134 preg_match('/type="(.*?)"/ism', $attributes, $matches);
135 if (x($matches, 1)) {
136 $type = strtolower($matches[1]);
143 if (!in_array($type, ["link", "audio", "photo", "video"])) {
148 $data["type"] = $type;
152 preg_match("/url='(.*?)'/ism", $attributes, $matches);
153 if (x($matches, 1)) {
157 preg_match('/url="(.*?)"/ism', $attributes, $matches);
158 if (x($matches, 1)) {
163 $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
167 preg_match("/title='(.*?)'/ism", $attributes, $matches);
168 if (x($matches, 1)) {
169 $title = $matches[1];
172 preg_match('/title="(.*?)"/ism', $attributes, $matches);
173 if (x($matches, 1)) {
174 $title = $matches[1];
178 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
179 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
180 $title = str_replace(["[", "]"], ["[", "]"], $title);
181 $data["title"] = $title;
185 preg_match("/image='(.*?)'/ism", $attributes, $matches);
186 if (x($matches, 1)) {
187 $image = $matches[1];
190 preg_match('/image="(.*?)"/ism', $attributes, $matches);
191 if (x($matches, 1)) {
192 $image = $matches[1];
196 $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
200 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
201 if (x($matches, 1)) {
202 $preview = $matches[1];
205 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
206 if (x($matches, 1)) {
207 $preview = $matches[1];
210 if ($preview != "") {
211 $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
214 $data["description"] = trim($match[3]);
216 $data["after"] = trim($match[4]);
221 public static function getAttachedData($body, $item = [])
225 - type: link, video, photo
233 $has_title = !empty($item['title']);
234 $plink = (!empty($item['plink']) ? $item['plink'] : '');
235 $post = self::getAttachmentData($body);
237 // if nothing is found, it maybe having an image.
238 if (!isset($post["type"])) {
239 // Simplify image codes
240 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
242 $URLSearchString = "^\[\]";
244 $body = preg_replace("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
246 if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
247 if ((count($pictures) == 1) && !$has_title) {
248 // Checking, if the link goes to a picture
249 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
252 // Sometimes photo posts to the own album are not detected at the start.
253 // So we seem to cannot use the cache for these cases. That's strange.
254 if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
255 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
258 if ($data["type"] == "photo") {
259 $post["type"] = "photo";
260 if (isset($data["images"][0])) {
261 $post["image"] = $data["images"][0]["src"];
262 $post["url"] = $data["url"];
264 $post["image"] = $data["url"];
267 $post["preview"] = $pictures[0][2];
268 $post["text"] = str_replace($pictures[0][0], "", $body);
270 $imgdata = Image::getInfoFromURL($pictures[0][1]);
271 if ($imgdata && substr($imgdata["mime"], 0, 6) == "image/") {
272 $post["type"] = "photo";
273 $post["image"] = $pictures[0][1];
274 $post["preview"] = $pictures[0][2];
275 $post["text"] = str_replace($pictures[0][0], "", $body);
278 } elseif (count($pictures) > 0) {
279 $post["type"] = "link";
280 $post["url"] = $plink;
281 $post["image"] = $pictures[0][2];
282 $post["text"] = $body;
284 } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
285 if ((count($pictures) == 1) && !$has_title) {
286 $post["type"] = "photo";
287 $post["image"] = $pictures[0][1];
288 $post["text"] = str_replace($pictures[0][0], "", $body);
289 } elseif (count($pictures) > 0) {
290 $post["type"] = "link";
291 $post["url"] = $plink;
292 $post["image"] = $pictures[0][1];
293 $post["text"] = $body;
297 // Test for the external links
298 preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
299 preg_match_all("(\[url\=([$URLSearchString]*)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
301 $links = array_merge($links1, $links2);
303 // If there is only a single one, then use it.
304 // This should cover link posts via API.
305 if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
306 $post["type"] = "link";
307 $post["text"] = trim($body);
308 $post["url"] = $links[0][1];
311 // Now count the number of external media links
312 preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
313 preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
314 preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
315 preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
317 // Add them to the other external links
318 $links = array_merge($links, $links1, $links2, $links3, $links4);
320 // Are there more than one?
321 if (count($links) > 1) {
322 // The post will be the type "text", which means a blog post
323 unset($post["type"]);
324 $post["url"] = $plink;
327 if (!isset($post["type"])) {
328 $post["type"] = "text";
329 $post["text"] = trim($body);
331 } elseif (isset($post["url"]) && ($post["type"] == "video")) {
332 $data = ParseUrl::getSiteinfoCached($post["url"], true);
334 if (isset($data["images"][0])) {
335 $post["image"] = $data["images"][0]["src"];
343 * @brief Converts a BBCode text into plaintext
345 * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
349 public static function toPlaintext($text, $keep_urls = true)
351 $naked_text = preg_replace('/\[(.+?)\]/','', $text);
353 $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text);
359 private static function proxyUrl($image, $simplehtml = false)
361 // Only send proxied pictures to API and for internal display
362 if (in_array($simplehtml, [false, 2])) {
363 return ProxyUtils::proxifyUrl($image);
369 public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
371 // Suppress "view full size"
372 if (intval(Config::get('system', 'no_view_full_size'))) {
373 $include_link = false;
376 // Picture addresses can contain special characters
377 $s = htmlspecialchars_decode($srctext);
380 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
382 foreach ($matches as $mtch) {
383 logger('scale_external_image: ' . $mtch[1]);
385 $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
386 if (stristr($mtch[1], $hostname)) {
390 // $scale_replace, if passed, is an array of two elements. The
391 // first is the name of the full-size image. The second is the
392 // name of a remote, scaled-down version of the full size image.
393 // This allows Friendica to display the smaller remote image if
394 // one exists, while still linking to the full-size image
395 if ($scale_replace) {
396 $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
400 $i = Network::fetchUrl($scaled);
405 // guess mimetype from headers or filename
406 $type = Image::guessType($mtch[1], true);
409 $Image = new Image($i, $type);
410 if ($Image->isValid()) {
411 $orig_width = $Image->getWidth();
412 $orig_height = $Image->getHeight();
414 if ($orig_width > 640 || $orig_height > 640) {
415 $Image->scaleDown(640);
416 $new_width = $Image->getWidth();
417 $new_height = $Image->getHeight();
418 logger('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], LOGGER_DEBUG);
421 '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
422 . "\n" . (($include_link)
423 ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
427 logger('scale_external_images: new string: ' . $s, LOGGER_DEBUG);
434 // replace the special char encoding
435 $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
440 * The purpose of this function is to apply system message length limits to
441 * imported messages without including any embedded photos in the length
443 * @brief Truncates imported message body string length to max_import_size
444 * @param string $body
447 public static function limitBodySize($body)
449 $maxlen = Config::get('config', 'max_import_size', 0);
451 // If the length of the body, including the embedded images, is smaller
452 // than the maximum, then don't waste time looking for the images
453 if ($maxlen && (strlen($body) > $maxlen)) {
455 logger('the total body length exceeds the limit', LOGGER_DEBUG);
461 $img_start = strpos($orig_body, '[img');
462 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
463 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
464 while (($img_st_close !== false) && ($img_end !== false)) {
466 $img_st_close++; // make it point to AFTER the closing bracket
467 $img_end += $img_start;
468 $img_end += strlen('[/img]');
470 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
471 // This is an embedded image
473 if (($textlen + $img_start) > $maxlen) {
474 if ($textlen < $maxlen) {
475 logger('the limit happens before an embedded image', LOGGER_DEBUG);
476 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
480 $new_body = $new_body . substr($orig_body, 0, $img_start);
481 $textlen += $img_start;
484 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
487 if (($textlen + $img_end) > $maxlen) {
488 if ($textlen < $maxlen) {
489 logger('the limit happens before the end of a non-embedded image', LOGGER_DEBUG);
490 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
494 $new_body = $new_body . substr($orig_body, 0, $img_end);
495 $textlen += $img_end;
498 $orig_body = substr($orig_body, $img_end);
500 if ($orig_body === false) {
501 // in case the body ends on a closing image tag
505 $img_start = strpos($orig_body, '[img');
506 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
507 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
510 if (($textlen + strlen($orig_body)) > $maxlen) {
511 if ($textlen < $maxlen) {
512 logger('the limit happens after the end of the last image', LOGGER_DEBUG);
513 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
516 logger('the text size with embedded images extracted did not violate the limit', LOGGER_DEBUG);
517 $new_body = $new_body . $orig_body;
527 * Processes [attachment] tags
529 * Note: Can produce a [bookmark] tag in the returned string
531 * @brief Processes [attachment] tags
532 * @param string $return
533 * @param bool|int $simplehtml
534 * @param bool $tryoembed
537 private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
539 $data = self::getAttachmentData($return);
540 if (empty($data) || empty($data["url"])) {
544 if (isset($data["title"])) {
545 $data["title"] = strip_tags($data["title"]);
546 $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
548 $data["title"] = null;
551 if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data["image"])) {
552 $data["preview"] = $data["image"];
557 if ($simplehtml == 7) {
558 $return = self::convertUrlForOStatus($data["url"]);
559 } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
560 $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data["url"], $data["title"]);
563 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
564 $return = OEmbed::getHTML($data['url'], $data['title']);
566 throw new Exception('OEmbed is disabled for this attachment.');
568 } catch (Exception $e) {
569 $data["title"] = defaults($data, 'title', $data['url']);
571 if ($simplehtml != 4) {
572 $return = sprintf('<div class="type-%s">', $data["type"]);
575 if (!empty($data["image"])) {
576 $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), $data["title"]);
577 } elseif (!empty($data["preview"])) {
578 $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), $data["title"]);
581 if (($data["type"] == "photo") && !empty($data["url"]) && !empty($data["image"])) {
582 $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"]);
584 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
587 if (!empty($data["description"]) && $data["description"] != $data["title"]) {
588 // Sanitize the HTML by converting it to BBCode
589 $bbcode = HTML::toBBCode($data["description"]);
590 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
592 if ($data["type"] == "link") {
593 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
596 if ($simplehtml != 4) {
602 return trim($data["text"] . ' ' . $return . ' ' . $data["after"]);
605 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
607 $data = self::getAttachmentData($Text);
612 return $data["text"] . defaults($data, 'after', '');
615 $title = htmlentities(defaults($data, 'title', ''), ENT_QUOTES, 'UTF-8', false);
616 $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
617 if ($plaintext || (($title != "") && strstr($text, $title))) {
618 $data["title"] = $data["url"];
619 } elseif (($text != "") && strstr($title, $text)) {
620 $data["text"] = $data["title"];
621 $data["title"] = $data["url"];
624 if (($data["text"] == "") && ($data["title"] != "") && ($data["url"] == "")) {
625 return $data["title"] . $data["after"];
628 // If the link already is included in the post, don't add it again
629 if (!empty($data["url"]) && strpos($data["text"], $data["url"])) {
630 return $data["text"] . $data["after"];
633 $text = $data["text"];
635 if (!empty($data["url"]) && !empty($data["title"])) {
636 $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
637 } elseif (!empty($data["url"])) {
638 $text .= "\n[url]" . $data["url"] . "[/url]";
641 return $text . "\n" . $data["after"];
645 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
647 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
648 * @param array $match Array with the matching values
649 * @return string reformatted link including HTML codes
651 private static function convertUrlForOStatusCallback($match)
655 if (isset($match[2]) && ($match[1] != $match[2])) {
659 $parts = parse_url($url);
660 if (!isset($parts['scheme'])) {
664 return self::convertUrlForOStatus($url);
668 * @brief Converts [url] BBCodes in a format that looks fine on OStatus systems.
669 * @param string $url URL that is about to be reformatted
670 * @return string reformatted link including HTML codes
672 private static function convertUrlForOStatus($url)
674 $parts = parse_url($url);
675 $scheme = $parts['scheme'] . '://';
676 $styled_url = str_replace($scheme, '', $url);
678 if (strlen($styled_url) > 30) {
679 $styled_url = substr($styled_url, 0, 30) . "…";
682 $html = '<a href="%s" target="_blank">%s</a>';
684 return sprintf($html, $url, $styled_url);
688 * [noparse][i]italic[/i][/noparse] turns into
689 * [noparse][ i ]italic[ /i ][/noparse],
690 * to hide them from parser.
692 private static function escapeNoparseCallback($match)
694 $whole_match = $match[0];
695 $captured = $match[1];
696 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
697 $new_str = str_replace($captured, $spacefied, $whole_match);
702 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
703 * now turns back and the [noparse] tags are trimed
704 * returning [i]italic[/i]
706 private static function unescapeNoparseCallback($match)
708 $captured = $match[1];
709 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
714 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
717 * @param string $text Text to search
718 * @param string $name Tag name
719 * @param int $occurrences Number of first occurrences to skip
720 * @return boolean|array
722 public static function getTagPosition($text, $name, $occurrences = 0)
724 if ($occurrences < 0) {
729 for ($i = 0; $i <= $occurrences; $i++) {
730 if ($start_open !== false) {
731 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
735 if ($start_open === false) {
739 $start_equal = strpos($text, '=', $start_open);
740 $start_close = strpos($text, ']', $start_open);
742 if ($start_close === false) {
748 $end_open = strpos($text, '[/' . $name . ']', $start_close);
750 if ($end_open === false) {
756 'open' => $start_open,
757 'close' => $start_close
761 'close' => $end_open + strlen('[/' . $name . ']')
765 if ($start_equal !== false) {
766 $res['start']['equal'] = $start_equal + 1;
773 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
775 * @param type $pattern Preg pattern string
776 * @param type $replace Preg replace string
777 * @param type $name BBCode tag name
778 * @param type $text Text to search
781 public static function pregReplaceInTag($pattern, $replace, $name, $text)
784 $pos = self::getTagPosition($text, $name, $occurrences);
785 while ($pos !== false && $occurrences++ < 1000) {
786 $start = substr($text, 0, $pos['start']['open']);
787 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
788 $end = substr($text, $pos['end']['close']);
789 if ($end === false) {
793 $subject = preg_replace($pattern, $replace, $subject);
794 $text = $start . $subject . $end;
796 $pos = self::getTagPosition($text, $name, $occurrences);
802 private static function extractImagesFromItemBody($body)
809 $img_start = strpos($orig_body, '[img');
810 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
811 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
812 while (($img_st_close !== false) && ($img_end !== false)) {
813 $img_st_close++; // make it point to AFTER the closing bracket
814 $img_end += $img_start;
816 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
817 // This is an embedded image
818 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
819 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
823 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
826 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
828 if ($orig_body === false) {
829 // in case the body ends on a closing image tag
833 $img_start = strpos($orig_body, '[img');
834 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
835 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
838 $new_body = $new_body . $orig_body;
840 return ['body' => $new_body, 'images' => $saved_image];
843 private static function interpolateSavedImagesIntoItemBody($body, array $images)
848 foreach ($images as $image) {
849 // We're depending on the property of 'foreach' (specified on the PHP website) that
850 // it loops over the array starting from the first element and going sequentially
851 // to the last element
852 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
853 '<img src="' . self::proxyUrl($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
861 * Processes [share] tags
863 * Note: Can produce a [bookmark] tag in the output
865 * @brief Processes [share] tags
866 * @param array $share preg_match_callback result array
867 * @param bool|int $simplehtml
870 private static function convertShare($share, $simplehtml)
872 $attributes = $share[2];
875 preg_match("/author='(.*?)'/ism", $attributes, $matches);
876 if (x($matches, 1)) {
877 $author = html_entity_decode($matches[1], ENT_QUOTES, 'UTF-8');
880 preg_match('/author="(.*?)"/ism', $attributes, $matches);
881 if (x($matches, 1)) {
882 $author = $matches[1];
886 preg_match("/profile='(.*?)'/ism", $attributes, $matches);
887 if (x($matches, 1)) {
888 $profile = $matches[1];
891 preg_match('/profile="(.*?)"/ism', $attributes, $matches);
892 if (x($matches, 1)) {
893 $profile = $matches[1];
897 preg_match("/avatar='(.*?)'/ism", $attributes, $matches);
898 if (x($matches, 1)) {
899 $avatar = $matches[1];
902 preg_match('/avatar="(.*?)"/ism', $attributes, $matches);
903 if (x($matches, 1)) {
904 $avatar = $matches[1];
908 preg_match("/link='(.*?)'/ism", $attributes, $matches);
909 if (x($matches, 1)) {
913 preg_match('/link="(.*?)"/ism', $attributes, $matches);
914 if (x($matches, 1)) {
920 preg_match("/posted='(.*?)'/ism", $attributes, $matches);
921 if (x($matches, 1)) {
922 $posted = $matches[1];
925 preg_match('/posted="(.*?)"/ism', $attributes, $matches);
926 if (x($matches, 1)) {
927 $posted = $matches[1];
930 // We only call this so that a previously unknown contact can be added.
931 // This is important for the function "Model\Contact::getDetailsByURL()".
932 // This function then can fetch an entry from the contact table.
933 Contact::getIdForURL($profile, 0, true);
935 $data = Contact::getDetailsByURL($profile);
937 if (x($data, "name") && x($data, "addr")) {
938 $userid_compact = $data["name"] . " (" . $data["addr"] . ")";
940 $userid_compact = Protocol::getAddrFromProfileUrl($profile, $author);
943 if (x($data, "addr")) {
944 $userid = $data["addr"];
946 $userid = Protocol::formatMention($profile, $author);
949 if (x($data, "name")) {
950 $author = $data["name"];
953 if (x($data, "micro")) {
954 $avatar = $data["micro"];
957 $preshare = trim($share[1]);
958 if ($preshare != "") {
959 $preshare .= "<br />";
962 switch ($simplehtml) {
964 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' <a href="' . $profile . '">' . $userid . "</a>: <br />»" . $share[3] . "«";
967 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
970 $headline = '<b>' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . $userid . ':</b><br />';
972 $text = trim($share[1]);
978 if (stripos(normalise_link($link), 'http://twitter.com/') === 0) {
979 $text .= '<br /><a href="' . $link . '">' . $link . '</a>';
981 $text .= $headline . '<blockquote>' . trim($share[3]) . "</blockquote><br />";
984 $text .= '<br /><a href="' . $link . '">[l]</a>';
990 $headline = '<br /><b>' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8');
991 $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $link, $userid, $posted);
992 $headline .= ":</b><br />";
994 $text = trim($share[1]);
1000 $text .= $headline . '<blockquote class="shared_content">' . trim($share[3]) . "</blockquote><br />";
1004 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1006 case 7: // statusnet/GNU Social
1007 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3];
1010 $text = $preshare . "RT @" . $userid_compact . ": " . $share[3];
1013 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1016 $text .= "<br /><br />" . $link;
1020 // Transforms quoted tweets in rich attachments to avoid nested tweets
1021 if (stripos(normalise_link($link), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($link)) {
1023 $oembed = OEmbed::getHTML($link, $preshare);
1024 } catch (Exception $e) {
1025 $oembed = sprintf('[bookmark=%s]%s[/bookmark]', $link, $preshare);
1028 $text = $preshare . $oembed;
1030 $text = trim($share[1]) . "\n";
1032 $avatar = ProxyUtils::proxifyUrl($avatar, false, ProxyUtils::SIZE_THUMB);
1034 $tpl = get_markup_template('shared_content.tpl');
1035 $text .= replace_macros($tpl, [
1036 '$profile' => $profile,
1037 '$avatar' => $avatar,
1038 '$author' => $author,
1040 '$posted' => $posted,
1041 '$content' => trim($share[3])
1050 private static function removePictureLinksCallback($match)
1052 $text = Cache::get($match[1]);
1054 if (is_null($text)) {
1055 $a = self::getApp();
1057 $stamp1 = microtime(true);
1059 $ch = @curl_init($match[1]);
1060 @curl_setopt($ch, CURLOPT_NOBODY, true);
1061 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1062 @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
1064 $curl_info = @curl_getinfo($ch);
1066 $a->save_timestamp($stamp1, "network");
1068 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1069 $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
1071 $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
1073 // if its not a picture then look if its a page that contains a picture link
1074 $body = Network::fetchUrl($match[1]);
1076 $doc = new DOMDocument();
1077 @$doc->loadHTML($body);
1078 $xpath = new DOMXPath($doc);
1079 $list = $xpath->query("//meta[@name]");
1080 foreach ($list as $node) {
1083 if ($node->attributes->length) {
1084 foreach ($node->attributes as $attribute) {
1085 $attr[$attribute->name] = $attribute->value;
1089 if (strtolower($attr["name"]) == "twitter:image") {
1090 $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
1094 Cache::set($match[1], $text);
1100 private static function expandLinksCallback($match)
1102 if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1103 return ($match[1] . "[url]" . $match[2] . "[/url]");
1105 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1109 private static function cleanPictureLinksCallback($match)
1111 $text = Cache::get($match[1]);
1113 if (is_null($text)) {
1114 $a = self::getApp();
1116 $stamp1 = microtime(true);
1118 $ch = @curl_init($match[1]);
1119 @curl_setopt($ch, CURLOPT_NOBODY, true);
1120 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1121 @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
1123 $curl_info = @curl_getinfo($ch);
1125 $a->save_timestamp($stamp1, "network");
1127 // if its a link to a picture then embed this picture
1128 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1129 $text = "[img]" . $match[1] . "[/img]";
1131 $text = "[img]" . $match[2] . "[/img]";
1133 // if its not a picture then look if its a page that contains a picture link
1134 $body = Network::fetchUrl($match[1]);
1136 $doc = new DOMDocument();
1137 @$doc->loadHTML($body);
1138 $xpath = new DOMXPath($doc);
1139 $list = $xpath->query("//meta[@name]");
1140 foreach ($list as $node) {
1142 if ($node->attributes->length) {
1143 foreach ($node->attributes as $attribute) {
1144 $attr[$attribute->name] = $attribute->value;
1148 if (strtolower($attr["name"]) == "twitter:image") {
1149 $text = "[img]" . $attr["content"] . "[/img]";
1153 Cache::set($match[1], $text);
1159 public static function cleanPictureLinks($text)
1161 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1165 private static function textHighlightCallback($match)
1167 // Fallback in case the language doesn't exist
1168 $return = '[code]' . $match[2] . '[/code]';
1170 if (in_array(strtolower($match[1]),
1171 ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby',
1172 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh', 'bash'])
1174 $return = text_highlight($match[2], strtolower($match[1]));
1181 * @brief Converts a BBCode message to HTML message
1183 * BBcode 2 HTML was written by WAY2WEB.net
1184 * extended to work with Mistpark/Friendica - Mike Macgirvin
1186 * Simple HTML values meaning:
1187 * - 0: Friendica display
1189 * - 2: Used for Google+, Windows Phone push, Friendica API
1190 * - 3: Used before converting to Markdown in bb2diaspora.php
1191 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1194 * - 7: Used for dfrn, OStatus
1195 * - 8: Used for WP backlink text setting
1197 * @param string $text
1198 * @param bool $try_oembed
1199 * @param int $simple_html
1200 * @param bool $for_plaintext
1203 public static function convert($text, $try_oembed = true, $simple_html = false, $for_plaintext = false)
1205 $a = self::getApp();
1208 * preg_match_callback function to replace potential Oembed tags with Oembed content
1210 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1212 * $match[2] = $title or absent
1214 $try_oembed_callback = function ($match)
1217 $title = defaults($match, 2, null);
1220 $return = OEmbed::getHTML($url, $title);
1221 } catch (Exception $ex) {
1222 $return = $match[0];
1228 // Hide all [noparse] contained bbtags by spacefying them
1229 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1231 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1232 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1233 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1235 // Remove the abstract element. It is a non visible element.
1236 $text = self::stripAbstract($text);
1238 // Move all spaces out of the tags
1239 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1240 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1242 // Extract the private images which use data urls since preg has issues with
1243 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1244 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1246 $extracted = self::extractImagesFromItemBody($text);
1247 $text = $extracted['body'];
1248 $saved_image = $extracted['images'];
1250 // If we find any event code, turn it into an event.
1251 // After we're finished processing the bbcode we'll
1252 // replace all of the event code with a reformatted version.
1254 $ev = Event::fromBBCode($text);
1256 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1257 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1259 $text = str_replace("<", "<", $text);
1260 $text = str_replace(">", ">", $text);
1262 // remove some newlines before the general conversion
1263 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1264 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1266 $text = preg_replace("/\n\[code\]/ism", "[code]", $text);
1267 $text = preg_replace("/\[\/code\]\n/ism", "[/code]", $text);
1269 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1271 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1274 // Check for [code] text here, before the linefeeds are messed with.
1275 // The highlighter will unescape and re-escape the content.
1276 if (strpos($text, '[code=') !== false) {
1277 $text = preg_replace_callback("/\[code=(.*?)\](.*?)\[\/code\]/ism", 'self::textHighlightCallback', $text);
1279 // Convert new line chars to html <br /> tags
1281 // nlbr seems to be hopelessly messed up
1282 // $Text = nl2br($Text);
1284 // We'll emulate it.
1286 $text = trim($text);
1287 $text = str_replace("\r\n", "\n", $text);
1289 // removing multiplicated newlines
1290 if (Config::get("system", "remove_multiplicated_lines")) {
1291 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1292 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1293 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1294 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1297 $text = str_replace($search, $replace, $text);
1298 } while ($oldtext != $text);
1301 // Set up the parameters for a URL search string
1302 $URLSearchString = "^\[\]";
1303 // Set up the parameters for a MAIL search string
1304 $MAILSearchString = $URLSearchString;
1306 // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text
1307 if (!$for_plaintext) {
1308 // Autolink feature (thanks to http://code.seebz.net/p/autolink-php/)
1309 // Currently disabled, since the function is too greedy
1310 // $autolink_regex = "`([^\]\=\"']|^)(https?\://[^\s<]+[^\s<\.\)])`ism";
1311 $autolink_regex = "/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism";
1312 $text = preg_replace($autolink_regex, '$1[url]$2[/url]', $text);
1313 if ($simple_html == 7) {
1314 $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1315 $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1318 $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text);
1319 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1323 // Handle attached links or videos
1324 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1326 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1328 // Remove all hashtag addresses
1329 if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7])) {
1330 $text = preg_replace("/([#@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $text);
1331 } elseif ($simple_html == 3) {
1332 // The ! is converted to @ since Diaspora only understands the @
1333 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1334 '@<a href="$2">$3</a>',
1336 } elseif ($simple_html == 7) {
1337 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1338 '$1<span class="vcard"><a href="$2" class="url" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1340 } elseif (!$simple_html) {
1341 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1342 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1346 // Bookmarks in red - will be converted to bookmarks in friendica
1347 $text = preg_replace("/#\^\[url\]([$URLSearchString]*)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1348 $text = preg_replace("/#\^\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1349 $text = preg_replace("/#\[url\=[$URLSearchString]*\]\^\[\/url\]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i",
1350 "[bookmark=$1]$2[/bookmark]", $text);
1352 if (in_array($simple_html, [2, 6, 7, 8, 9])) {
1353 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1354 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1355 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1358 if ($simple_html == 5) {
1359 $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1362 // Perform URL Search
1364 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1367 if ($simple_html == 5) {
1368 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1370 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1373 // Handle Diaspora posts
1374 $text = preg_replace_callback(
1375 "&\[url=/posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1377 return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1381 $text = preg_replace_callback(
1382 "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1384 return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1388 // Server independent link to posts and comments
1389 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1390 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1391 $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
1393 $text = preg_replace("/([#])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1394 '$1<a href="' . System::baseUrl() . '/search?tag=$3" class="tag" title="$3">$3</a>', $text);
1396 $text = preg_replace("/\[url\=([$URLSearchString]*)\]#(.*?)\[\/url\]/ism",
1397 '#<a href="' . System::baseUrl() . '/search?tag=$2" class="tag" title="$2">$2</a>', $text);
1399 $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1400 $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1401 //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $Text);
1403 // Red compatibility, though the link can't be authenticated on Friendica
1404 $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1407 // we may need to restrict this further if it picks up too many strays
1408 // link acct:user@host to a webfinger profile redirector
1410 $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);
1412 // Perform MAIL Search
1413 $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1414 $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1416 // leave open the posibility of [map=something]
1417 // this is replaced in prepare_body() which has knowledge of the item location
1419 if (strpos($text, '[/map]') !== false) {
1420 $text = preg_replace_callback(
1421 "/\[map\](.*?)\[\/map\]/ism",
1422 function ($match) use ($simple_html) {
1423 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1428 if (strpos($text, '[map=') !== false) {
1429 $text = preg_replace_callback(
1430 "/\[map=(.*?)\]/ism",
1431 function ($match) use ($simple_html) {
1432 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1437 if (strpos($text, '[map]') !== false) {
1438 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1441 // Check for headers
1442 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1443 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1444 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1445 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1446 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1447 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1449 // Check for paragraph
1450 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1452 // Check for bold text
1453 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1455 // Check for Italics text
1456 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1458 // Check for Underline text
1459 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1461 // Check for strike-through text
1462 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1464 // Check for over-line text
1465 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1467 // Check for colored text
1468 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1470 // Check for sized text
1471 // [size=50] --> font-size: 50px (with the unit).
1472 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1473 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1475 // Check for centered text
1476 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1478 // Check for list text
1479 $text = str_replace("[*]", "<li>", $text);
1481 // Check for style sheet commands
1482 $text = preg_replace_callback(
1483 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1485 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1490 // Check for CSS classes
1491 $text = preg_replace_callback(
1492 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1494 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1499 // handle nested lists
1502 while ((((strpos($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 // This is actually executed in prepare_body()
1530 $text = str_replace('[nosmile]', '', $text);
1532 // Check for font change text
1533 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1535 // Declare the format for [code] layout
1537 $CodeLayout = '<code>$1</code>';
1538 // Check for [code] text
1539 $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text);
1541 // Declare the format for [spoiler] layout
1542 $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
1544 // Check for [spoiler] text
1545 // handle nested quotes
1547 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1548 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
1551 // Check for [spoiler=Author] text
1553 $t_wrote = L10n::t('$1 wrote:');
1555 // handle nested quotes
1557 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1558 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1559 "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
1563 // Declare the format for [quote] layout
1564 $QuoteLayout = '<blockquote>$1</blockquote>';
1566 // Check for [quote] text
1567 // handle nested quotes
1569 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1570 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1573 // Check for [quote=Author] text
1575 $t_wrote = L10n::t('$1 wrote:');
1577 // handle nested quotes
1579 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1580 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1581 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1586 // [img=widthxheight]image source[/img]
1587 $text = preg_replace_callback(
1588 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1589 function ($matches) use ($simple_html) {
1590 if (strpos($matches[3], "data:image/") === 0) {
1594 $matches[3] = self::proxyUrl($matches[3], $simple_html);
1595 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1600 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1601 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1603 $text = preg_replace_callback("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism",
1604 function ($matches) use ($simple_html) {
1605 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1606 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1607 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '">';
1612 // [img]pathtoimage[/img]
1613 $text = preg_replace_callback(
1614 "/\[img\](.*?)\[\/img\]/ism",
1615 function ($matches) use ($simple_html) {
1616 if (strpos($matches[1], "data:image/") === 0) {
1620 $matches[1] = self::proxyUrl($matches[1], $simple_html);
1621 return "[img]" . $matches[1] . "[/img]";
1626 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1627 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1630 $text = preg_replace_callback("/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
1631 function ($match) use ($simple_html) {
1632 return self::convertShare($match, $simple_html);
1635 $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);
1636 $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 />', $text);
1637 //$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);
1641 $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);
1642 $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);
1644 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1645 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1647 $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1648 '<a href="$1" target="_blank">$1</a>', $text);
1649 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1650 '<a href="$1" target="_blank">$1</a>', $text);
1653 // html5 video and audio
1657 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1659 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1662 // Youtube extensions
1664 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1665 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1666 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1669 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1670 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1671 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1674 $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);
1676 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1677 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1681 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1682 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1685 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1686 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1689 $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);
1691 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1692 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1696 $text = OEmbed::BBCode2HTML($text);
1698 // Avoid triple linefeeds through oembed
1699 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1701 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1702 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1703 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1704 // start which is always required). Allow desc with a missing summary for compatibility.
1706 if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) {
1707 $sub = Event::getHTML($ev, $simple_html);
1709 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1710 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1711 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1712 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1713 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1714 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1715 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1718 // Replace non graphical smilies for external posts
1720 $text = Smilies::replace($text, false, true);
1723 // Replace inline code blocks
1724 $text = preg_replace_callback("|(?!<br[^>]*>)<code>([^<]*)</code>(?!<br[^>]*>)|ism",
1725 function ($match) use ($simple_html) {
1726 $return = '<key>' . $match[1] . '</key>';
1727 // Use <code> for Diaspora inline code blocks
1728 if ($simple_html === 3) {
1729 $return = '<code>' . $match[1] . '</code>';
1735 // Unhide all [noparse] contained bbtags unspacefying them
1736 // and triming the [noparse] tag.
1738 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1739 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1740 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1742 /// @todo What is the meaning of these lines?
1743 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1744 $text = preg_replace('/\&\#039\;/', '\'', $text);
1746 // Currently deactivated, it made problems with " inside of alt texts.
1747 //$text = preg_replace('/\"\;/', '"', $text);
1749 // fix any escaped ampersands that may have been converted into links
1750 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1752 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1753 $allowed_src_protocols = ['http', 'redir', 'cid'];
1754 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1755 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
1757 // sanitize href attributes (only whitelisted protocols URLs)
1758 // default value for backward compatibility
1759 $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']);
1761 // Always allowed protocol even if config isn't set or not including it
1762 $allowed_link_protocols[] = 'http';
1763 $allowed_link_protocols[] = 'redir/';
1765 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1766 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
1769 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1772 // Clean up the HTML by loading and saving the HTML with the DOM.
1773 // Bad structured html can break a whole page.
1774 // For performance reasons do it only with ativated item cache or at export.
1775 if (!$try_oembed || (get_itemcachepath() != "")) {
1776 $doc = new DOMDocument();
1777 $doc->preserveWhiteSpace = false;
1779 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1781 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1782 $encoding = '<?xml encoding="UTF-8">';
1783 @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
1784 $doc->encoding = 'UTF-8';
1785 $text = $doc->saveHTML();
1786 $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
1788 $text = str_replace('<br></li>', '</li>', $text);
1790 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1793 // Clean up some useless linebreaks in lists
1794 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1795 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1796 //$Text = str_replace('</li><br />', '</li>', $Text);
1797 //$Text = str_replace('<br /><li>', '<li>', $Text);
1798 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1800 Addon::callHooks('bbcode', $text);
1806 * @brief Strips the "abstract" tag from the provided text
1808 * @param string $text The text with BBCode
1809 * @return string The same text - but without "abstract" element
1811 public static function stripAbstract($text)
1813 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1814 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1820 * @brief Returns the value of the "abstract" element
1822 * @param string $text The text that maybe contains the element
1823 * @param string $addon The addon for which the abstract is meant for
1824 * @return string The abstract
1826 public static function getAbstract($text, $addon = "")
1830 $addon = strtolower($addon);
1832 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1833 foreach ($results AS $result) {
1834 $abstracts[strtolower($result[1])] = $result[2];
1838 if (isset($abstracts[$addon])) {
1839 $abstract = $abstracts[$addon];
1842 if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1843 $abstract = $result[1];
1850 * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
1852 * @param array $match Matching values for the callback
1853 * @return string Replaced mention
1855 private static function bbCodeMention2DiasporaCallback($match)
1857 $contact = Contact::getDetailsByURL($match[3]);
1859 if (empty($contact['addr'])) {
1860 $contact = Probe::uri($match[3]);
1863 if (empty($contact['addr'])) {
1867 $mention = '@{' . $match[2] . '; ' . $contact['addr'] . '}';
1872 * @brief Converts a BBCode text into Markdown
1874 * This function converts a BBCode item body to be sent to Markdown-enabled
1875 * systems like Diaspora and Libertree
1877 * @param string $text
1878 * @param bool $for_diaspora Diaspora requires more changes than Libertree
1881 public static function toMarkdown($text, $for_diaspora = true)
1883 $a = self::getApp();
1885 $original_text = $text;
1887 // Since Diaspora is creating a summary for links, this function removes them before posting
1888 if ($for_diaspora) {
1889 $text = self::removeShareInformation($text);
1893 * Transform #tags, strip off the [url] and replace spaces with underscore
1895 $url_search_string = "^\[\]";
1896 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
1897 function ($matches) {
1898 return '#' . str_replace(' ', '_', $matches[2]);
1903 // Converting images with size parameters to simple images. Markdown doesn't know it.
1904 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
1906 // Extracting multi-line code blocks before the whitespace processing/code highlighter in self::convert()
1909 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#is",
1910 function ($matches) use (&$codeblocks) {
1911 $return = $matches[0];
1912 if (strpos($matches[2], "\n") !== false) {
1913 $return = '#codeblock-' . count($codeblocks) . '#';
1915 $prefix = '````' . $matches[1] . PHP_EOL;
1916 $codeblocks[] = $prefix . trim($matches[2]) . PHP_EOL . '````';
1923 // Convert it to HTML - don't try oembed
1924 if ($for_diaspora) {
1925 $text = self::convert($text, false, 3);
1927 // Add all tags that maybe were removed
1928 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
1930 foreach ($tags[2] as $tag) {
1931 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
1932 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
1933 $tagline .= '#' . $tag . ' ';
1936 $text = $text . " " . $tagline;
1939 $text = self::convert($text, false, 4);
1942 // mask some special HTML chars from conversation to markdown
1943 $text = str_replace(['<', '>', '&'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text);
1945 // If a link is followed by a quote then there should be a newline before it
1946 // Maybe we should make this newline at every time before a quote.
1947 $text = str_replace(["</a><blockquote>"], ["</a><br><blockquote>"], $text);
1949 $stamp1 = microtime(true);
1951 // Now convert HTML to Markdown
1952 $converter = new HtmlConverter();
1953 $text = $converter->convert($text);
1955 // unmask the special chars back to HTML
1956 $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
1958 $a->save_timestamp($stamp1, "parser");
1960 // Libertree has a problem with escaped hashtags.
1961 $text = str_replace(['\#'], ['#'], $text);
1963 // Remove any leading or trailing whitespace, as this will mess up
1964 // the Diaspora signature verification and cause the item to disappear
1965 $text = trim($text);
1967 if ($for_diaspora) {
1968 $url_search_string = "^\[\]";
1969 $text = preg_replace_callback(
1970 "/([@]\[(.*?)\])\(([$url_search_string]*?)\)/ism",
1971 ['self', 'bbCodeMention2DiasporaCallback'],
1976 // Restore code blocks
1977 $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
1978 function ($matches) use ($codeblocks) {
1980 if (isset($codeblocks[intval($matches[1])])) {
1981 $return = $codeblocks[$matches[1]];
1988 Addon::callHooks('bb2diaspora', $text);