4 * @file src/Content/Text/BBCode.php
7 namespace Friendica\Content\Text;
12 use Friendica\BaseObject;
13 use Friendica\Content\OEmbed;
14 use Friendica\Content\Smilies;
15 use Friendica\Core\Addon;
16 use Friendica\Core\Cache;
17 use Friendica\Core\Config;
18 use Friendica\Core\L10n;
19 use Friendica\Core\PConfig;
20 use Friendica\Core\Protocol;
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 League\HTMLToMarkdown\HtmlConverter;
31 require_once "mod/proxy.php";
33 class BBCode extends BaseObject
36 * @brief Fetches attachment data that were generated the old way
38 * @param string $body Message body
40 * 'type' -> Message type ("link", "video", "photo")
41 * 'text' -> Text before the shared message
42 * 'after' -> Text after the shared message
43 * 'image' -> Preview image of the message
44 * 'url' -> Url to the attached message
45 * 'title' -> Title of the attachment
46 * 'description' -> Description of the attachment
48 private static function getOldAttachmentData($body)
52 // Simplify image codes
53 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
55 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
56 foreach ($attached as $data) {
57 if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
61 $post["type"] = substr($data[1], 5);
63 $pos = strpos($body, $data[0]);
65 $post["text"] = trim(substr($body, 0, $pos));
66 $post["after"] = trim(substr($body, $pos + strlen($data[0])));
68 $post["text"] = trim(str_replace($data[0], "", $body));
71 $attacheddata = $data[2];
73 $URLSearchString = "^\[\]";
75 if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/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("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
89 $post["url"] = $matches[1];
90 $post["title"] = $matches[2];
92 if (($post["url"] == "") && (in_array($post["type"], ["link", "video"]))
93 && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/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 $body Message body
111 * 'type' -> Message type ("link", "video", "photo")
112 * 'text' -> Text before the shared message
113 * 'after' -> Text after the shared message
114 * 'image' -> Preview image of the message
115 * 'url' -> Url to the attached message
116 * 'title' -> Title of the attachment
117 * 'description' -> Description of the attachment
119 public static function getAttachmentData($body)
123 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
124 return self::getOldAttachmentData($body);
127 $attributes = $match[2];
129 $data["text"] = trim($match[1]);
132 preg_match("/type='(.*?)'/ism", $attributes, $matches);
133 if (x($matches, 1)) {
134 $type = strtolower($matches[1]);
137 preg_match('/type="(.*?)"/ism', $attributes, $matches);
138 if (x($matches, 1)) {
139 $type = strtolower($matches[1]);
146 if (!in_array($type, ["link", "audio", "photo", "video"])) {
151 $data["type"] = $type;
155 preg_match("/url='(.*?)'/ism", $attributes, $matches);
156 if (x($matches, 1)) {
160 preg_match('/url="(.*?)"/ism', $attributes, $matches);
161 if (x($matches, 1)) {
166 $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
170 preg_match("/title='(.*?)'/ism", $attributes, $matches);
171 if (x($matches, 1)) {
172 $title = $matches[1];
175 preg_match('/title="(.*?)"/ism', $attributes, $matches);
176 if (x($matches, 1)) {
177 $title = $matches[1];
181 $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
182 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
183 $title = str_replace(["[", "]"], ["[", "]"], $title);
184 $data["title"] = $title;
188 preg_match("/image='(.*?)'/ism", $attributes, $matches);
189 if (x($matches, 1)) {
190 $image = $matches[1];
193 preg_match('/image="(.*?)"/ism', $attributes, $matches);
194 if (x($matches, 1)) {
195 $image = $matches[1];
199 $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
203 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
204 if (x($matches, 1)) {
205 $preview = $matches[1];
208 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
209 if (x($matches, 1)) {
210 $preview = $matches[1];
213 if ($preview != "") {
214 $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
217 $data["description"] = trim($match[3]);
219 $data["after"] = trim($match[4]);
224 public static function getAttachedData($body, $item = [])
228 - type: link, video, photo
236 $has_title = !empty($item['title']);
237 $plink = (!empty($item['plink']) ? $item['plink'] : '');
238 $post = self::getAttachmentData($body);
240 // if nothing is found, it maybe having an image.
241 if (!isset($post["type"])) {
242 // Simplify image codes
243 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
245 $URLSearchString = "^\[\]";
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 Convert a message into plaintext for connectors to other networks
345 * @param array $b The message array that is about to be posted
346 * @param int $limit The maximum number of characters when posting to that network
347 * @param bool $includedlinks Has an attached link to be included into the message?
348 * @param int $htmlmode This triggers the behaviour of the bbcode conversion
349 * @param string $target_network Name of the network where the post should go to.
351 * @return string The converted message
353 public static function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "")
355 // Remove the hash tags
356 $URLSearchString = "^\[\]";
357 $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]);
359 // Add an URL element if the text contains a raw link
360 $body = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2[/url]', $body);
362 // Remove the abstract
363 $body = self::stripAbstract($body);
365 // At first look at data that is attached via "type-..." stuff
366 // This will hopefully replaced with a dedicated bbcode later
367 //$post = self::getAttachedData($b["body"]);
368 $post = self::getAttachedData($body, $b);
370 if (($b["title"] != "") && ($post["text"] != "")) {
371 $post["text"] = trim($b["title"]."\n\n".$post["text"]);
372 } elseif ($b["title"] != "") {
373 $post["text"] = trim($b["title"]);
378 // Fetch the abstract from the given target network
379 if ($target_network != "") {
380 $default_abstract = self::getAbstract($b["body"]);
381 $abstract = self::getAbstract($b["body"], $target_network);
383 // If we post to a network with no limit we only fetch
384 // an abstract exactly for this network
385 if (($limit == 0) && ($abstract == $default_abstract)) {
388 } else {// Try to guess the correct target network
391 $abstract = self::getAbstract($b["body"], NETWORK_TWITTER);
394 $abstract = self::getAbstract($b["body"], NETWORK_STATUSNET);
397 $abstract = self::getAbstract($b["body"], NETWORK_APPNET);
399 default: // We don't know the exact target.
400 // We fetch an abstract since there is a posting limit.
402 $abstract = self::getAbstract($b["body"]);
407 if ($abstract != "") {
408 $post["text"] = $abstract;
410 if ($post["type"] == "text") {
411 $post["type"] = "link";
412 $post["url"] = $b["plink"];
416 $html = self::convert($post["text"].$post["after"], false, $htmlmode);
417 $msg = HTML::toPlaintext($html, 0, true);
418 $msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8'));
421 if ($includedlinks) {
422 if ($post["type"] == "link") {
423 $link = $post["url"];
424 } elseif ($post["type"] == "text") {
425 $link = $post["url"];
426 } elseif ($post["type"] == "video") {
427 $link = $post["url"];
428 } elseif ($post["type"] == "photo") {
429 $link = $post["image"];
432 if (($msg == "") && isset($post["title"])) {
433 $msg = trim($post["title"]);
436 if (($msg == "") && isset($post["description"])) {
437 $msg = trim($post["description"]);
440 // If the link is already contained in the post, then it neeedn't to be added again
441 // But: if the link is beyond the limit, then it has to be added.
442 if (($link != "") && strstr($msg, $link)) {
443 $pos = strpos($msg, $link);
445 // Will the text be shortened in the link?
446 // Or is the link the last item in the post?
447 if (($limit > 0) && ($pos < $limit) && (($pos + 23 > $limit) || ($pos + strlen($link) == strlen($msg)))) {
448 $msg = trim(str_replace($link, "", $msg));
449 } elseif (($limit == 0) || ($pos < $limit)) {
450 // The limit has to be increased since it will be shortened - but not now
451 // Only do it with Twitter (htmlmode = 8)
452 if (($limit > 0) && (strlen($link) > 23) && ($htmlmode == 8)) {
453 $limit = $limit - 23 + strlen($link);
458 if ($post["type"] == "text") {
466 // Reduce multiple spaces
467 // When posted to a network with limited space, we try to gain space where possible
468 while (strpos($msg, " ") !== false) {
469 $msg = str_replace(" ", " ", $msg);
472 // Twitter is using its own limiter, so we always assume that shortened links will have this length
473 if (iconv_strlen($link, "UTF-8") > 0) {
474 $limit = $limit - 23;
477 if (iconv_strlen($msg, "UTF-8") > $limit) {
478 if (($post["type"] == "text") && isset($post["url"])) {
479 $post["url"] = $b["plink"];
480 } elseif (!isset($post["url"])) {
481 $limit = $limit - 23;
482 $post["url"] = $b["plink"];
483 // Which purpose has this line? It is now uncommented, but left as a reminder
484 //} elseif (strpos($b["body"], "[share") !== false) {
485 // $post["url"] = $b["plink"];
486 } elseif (PConfig::get($b["uid"], "system", "no_intelligent_shortening")) {
487 $post["url"] = $b["plink"];
489 $msg = Plaintext::shorten($msg, $limit);
493 $post["text"] = trim($msg);
498 public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
500 // Suppress "view full size"
501 if (intval(Config::get('system', 'no_view_full_size'))) {
502 $include_link = false;
505 // Picture addresses can contain special characters
506 $s = htmlspecialchars_decode($srctext);
509 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
511 foreach ($matches as $mtch) {
512 logger('scale_external_image: ' . $mtch[1]);
514 $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
515 if (stristr($mtch[1], $hostname)) {
519 // $scale_replace, if passed, is an array of two elements. The
520 // first is the name of the full-size image. The second is the
521 // name of a remote, scaled-down version of the full size image.
522 // This allows Friendica to display the smaller remote image if
523 // one exists, while still linking to the full-size image
524 if ($scale_replace) {
525 $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
529 $i = Network::fetchUrl($scaled);
534 // guess mimetype from headers or filename
535 $type = Image::guessType($mtch[1], true);
538 $Image = new Image($i, $type);
539 if ($Image->isValid()) {
540 $orig_width = $Image->getWidth();
541 $orig_height = $Image->getHeight();
543 if ($orig_width > 640 || $orig_height > 640) {
544 $Image->scaleDown(640);
545 $new_width = $Image->getWidth();
546 $new_height = $Image->getHeight();
547 logger('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], LOGGER_DEBUG);
550 '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
551 . "\n" . (($include_link)
552 ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
556 logger('scale_external_images: new string: ' . $s, LOGGER_DEBUG);
563 // replace the special char encoding
564 $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
569 * The purpose of this function is to apply system message length limits to
570 * imported messages without including any embedded photos in the length
572 * @brief Truncates imported message body string length to max_import_size
573 * @param string $body
576 public static function limitBodySize($body)
578 $maxlen = get_max_import_size();
580 // If the length of the body, including the embedded images, is smaller
581 // than the maximum, then don't waste time looking for the images
582 if ($maxlen && (strlen($body) > $maxlen)) {
584 logger('the total body length exceeds the limit', LOGGER_DEBUG);
590 $img_start = strpos($orig_body, '[img');
591 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
592 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
593 while (($img_st_close !== false) && ($img_end !== false)) {
595 $img_st_close++; // make it point to AFTER the closing bracket
596 $img_end += $img_start;
597 $img_end += strlen('[/img]');
599 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
600 // This is an embedded image
602 if (($textlen + $img_start) > $maxlen) {
603 if ($textlen < $maxlen) {
604 logger('the limit happens before an embedded image', LOGGER_DEBUG);
605 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
609 $new_body = $new_body . substr($orig_body, 0, $img_start);
610 $textlen += $img_start;
613 $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
616 if (($textlen + $img_end) > $maxlen) {
617 if ($textlen < $maxlen) {
618 logger('the limit happens before the end of a non-embedded image', LOGGER_DEBUG);
619 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
623 $new_body = $new_body . substr($orig_body, 0, $img_end);
624 $textlen += $img_end;
627 $orig_body = substr($orig_body, $img_end);
629 if ($orig_body === false) {
630 // in case the body ends on a closing image tag
634 $img_start = strpos($orig_body, '[img');
635 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
636 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
639 if (($textlen + strlen($orig_body)) > $maxlen) {
640 if ($textlen < $maxlen) {
641 logger('the limit happens after the end of the last image', LOGGER_DEBUG);
642 $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
645 logger('the text size with embedded images extracted did not violate the limit', LOGGER_DEBUG);
646 $new_body = $new_body . $orig_body;
656 * Processes [attachment] tags
658 * Note: Can produce a [bookmark] tag in the returned string
660 * @brief Processes [attachment] tags
661 * @param string $return
662 * @param bool|int $simplehtml
663 * @param bool $tryoembed
666 private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
668 $data = self::getAttachmentData($return);
673 if (isset($data["title"])) {
674 $data["title"] = strip_tags($data["title"]);
675 $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
678 if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) {
679 $data["preview"] = $data["image"];
684 if ($simplehtml == 7) {
685 $return = self::convertUrlForOStatus($data["url"]);
686 } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
687 $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data["url"], $data["title"]);
690 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
691 $return = OEmbed::getHTML($data['url'], $data['title']);
693 throw new Exception('OEmbed is disabled for this attachment.');
695 } catch (Exception $e) {
696 if ($simplehtml != 4) {
697 $return = sprintf('<div class="type-%s">', $data["type"]);
700 if ($data["image"] != "") {
701 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], proxy_url($data["image"]), $data["title"]);
702 } elseif ($data["preview"] != "") {
703 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], proxy_url($data["preview"]), $data["title"]);
706 if (($data["type"] == "photo") && ($data["url"] != "") && ($data["image"] != "")) {
707 $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], proxy_url($data["image"]), $data["title"]);
709 $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
712 if ($data["description"] != "" && $data["description"] != $data["title"]) {
713 // Sanitize the HTML by converting it to BBCode
714 $bbcode = HTML::toBBCode($data["description"]);
715 $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
717 if ($data["type"] == "link") {
718 $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
721 if ($simplehtml != 4) {
727 return trim($data["text"] . ' ' . $return . ' ' . $data["after"]);
730 public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
732 $data = self::getAttachmentData($Text);
737 return $data["text"] . $data["after"];
740 $title = htmlentities($data["title"], ENT_QUOTES, 'UTF-8', false);
741 $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
742 if ($plaintext || (($title != "") && strstr($text, $title))) {
743 $data["title"] = $data["url"];
744 } elseif (($text != "") && strstr($title, $text)) {
745 $data["text"] = $data["title"];
746 $data["title"] = $data["url"];
749 if (($data["text"] == "") && ($data["title"] != "") && ($data["url"] == "")) {
750 return $data["title"] . $data["after"];
753 // If the link already is included in the post, don't add it again
754 if (($data["url"] != "") && strpos($data["text"], $data["url"])) {
755 return $data["text"] . $data["after"];
758 $text = $data["text"];
760 if (($data["url"] != "") && ($data["title"] != "")) {
761 $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
762 } elseif (($data["url"] != "")) {
763 $text .= "\n[url]" . $data["url"] . "[/url]";
766 return $text . "\n" . $data["after"];
770 * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
772 * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
773 * @param array $match Array with the matching values
774 * @return string reformatted link including HTML codes
776 private static function convertUrlForOStatusCallback($match)
780 if (isset($match[2]) && ($match[1] != $match[2])) {
784 $parts = parse_url($url);
785 if (!isset($parts['scheme'])) {
789 return self::convertUrlForOStatus($url);
793 * @brief Converts [url] BBCodes in a format that looks fine on OStatus systems.
794 * @param string $url URL that is about to be reformatted
795 * @return string reformatted link including HTML codes
797 private static function convertUrlForOStatus($url)
799 $parts = parse_url($url);
800 $scheme = $parts['scheme'] . '://';
801 $styled_url = str_replace($scheme, '', $url);
803 if (strlen($styled_url) > 30) {
804 $styled_url = substr($styled_url, 0, 30) . "…";
807 $html = '<a href="%s" target="_blank">%s</a>';
809 return sprintf($html, $url, $styled_url);
813 * [noparse][i]italic[/i][/noparse] turns into
814 * [noparse][ i ]italic[ /i ][/noparse],
815 * to hide them from parser.
817 private static function escapeNoparseCallback($match)
819 $whole_match = $match[0];
820 $captured = $match[1];
821 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
822 $new_str = str_replace($captured, $spacefied, $whole_match);
827 * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
828 * now turns back and the [noparse] tags are trimed
829 * returning [i]italic[/i]
831 private static function unescapeNoparseCallback($match)
833 $captured = $match[1];
834 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
839 * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
842 * @param string $text Text to search
843 * @param string $name Tag name
844 * @param int $occurrences Number of first occurrences to skip
845 * @return boolean|array
847 public static function getTagPosition($text, $name, $occurrences = 0)
849 if ($occurrences < 0) {
854 for ($i = 0; $i <= $occurrences; $i++) {
855 if ($start_open !== false) {
856 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
860 if ($start_open === false) {
864 $start_equal = strpos($text, '=', $start_open);
865 $start_close = strpos($text, ']', $start_open);
867 if ($start_close === false) {
873 $end_open = strpos($text, '[/' . $name . ']', $start_close);
875 if ($end_open === false) {
881 'open' => $start_open,
882 'close' => $start_close
886 'close' => $end_open + strlen('[/' . $name . ']')
890 if ($start_equal !== false) {
891 $res['start']['equal'] = $start_equal + 1;
898 * Performs a preg_replace within the boundaries of all named BBCode tags in a text
900 * @param type $pattern Preg pattern string
901 * @param type $replace Preg replace string
902 * @param type $name BBCode tag name
903 * @param type $text Text to search
906 public static function pregReplaceInTag($pattern, $replace, $name, $text)
909 $pos = self::getTagPosition($text, $name, $occurrences);
910 while ($pos !== false && $occurrences++ < 1000) {
911 $start = substr($text, 0, $pos['start']['open']);
912 $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
913 $end = substr($text, $pos['end']['close']);
914 if ($end === false) {
918 $subject = preg_replace($pattern, $replace, $subject);
919 $text = $start . $subject . $end;
921 $pos = self::getTagPosition($text, $name, $occurrences);
927 private static function extractImagesFromItemBody($body)
934 $img_start = strpos($orig_body, '[img');
935 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
936 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
937 while (($img_st_close !== false) && ($img_end !== false)) {
938 $img_st_close++; // make it point to AFTER the closing bracket
939 $img_end += $img_start;
941 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
942 // This is an embedded image
943 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
944 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
948 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
951 $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
953 if ($orig_body === false) {
954 // in case the body ends on a closing image tag
958 $img_start = strpos($orig_body, '[img');
959 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
960 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
963 $new_body = $new_body . $orig_body;
965 return ['body' => $new_body, 'images' => $saved_image];
968 private static function interpolateSavedImagesIntoItemBody($body, array $images)
973 foreach ($images as $image) {
974 // We're depending on the property of 'foreach' (specified on the PHP website) that
975 // it loops over the array starting from the first element and going sequentially
976 // to the last element
977 $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
978 '<img src="' . proxy_url($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
986 * Processes [share] tags
988 * Note: Can produce a [bookmark] tag in the output
990 * @brief Processes [share] tags
991 * @param array $share preg_match_callback result array
992 * @param bool|int $simplehtml
995 private static function convertShare($share, $simplehtml)
997 $attributes = $share[2];
1000 preg_match("/author='(.*?)'/ism", $attributes, $matches);
1001 if (x($matches, 1)) {
1002 $author = html_entity_decode($matches[1], ENT_QUOTES, 'UTF-8');
1005 preg_match('/author="(.*?)"/ism', $attributes, $matches);
1006 if (x($matches, 1)) {
1007 $author = $matches[1];
1011 preg_match("/profile='(.*?)'/ism", $attributes, $matches);
1012 if (x($matches, 1)) {
1013 $profile = $matches[1];
1016 preg_match('/profile="(.*?)"/ism', $attributes, $matches);
1017 if (x($matches, 1)) {
1018 $profile = $matches[1];
1022 preg_match("/avatar='(.*?)'/ism", $attributes, $matches);
1023 if (x($matches, 1)) {
1024 $avatar = $matches[1];
1027 preg_match('/avatar="(.*?)"/ism', $attributes, $matches);
1028 if (x($matches, 1)) {
1029 $avatar = $matches[1];
1033 preg_match("/link='(.*?)'/ism", $attributes, $matches);
1034 if (x($matches, 1)) {
1035 $link = $matches[1];
1038 preg_match('/link="(.*?)"/ism', $attributes, $matches);
1039 if (x($matches, 1)) {
1040 $link = $matches[1];
1045 preg_match("/posted='(.*?)'/ism", $attributes, $matches);
1046 if (x($matches, 1)) {
1047 $posted = $matches[1];
1050 preg_match('/posted="(.*?)"/ism', $attributes, $matches);
1051 if (x($matches, 1)) {
1052 $posted = $matches[1];
1055 // We only call this so that a previously unknown contact can be added.
1056 // This is important for the function "Model\Contact::getDetailsByURL()".
1057 // This function then can fetch an entry from the contact table.
1058 Contact::getIdForURL($profile, 0, true);
1060 $data = Contact::getDetailsByURL($profile);
1062 if (x($data, "name") && x($data, "addr")) {
1063 $userid_compact = $data["name"] . " (" . $data["addr"] . ")";
1065 $userid_compact = Protocol::getAddrFromProfileUrl($profile, $author);
1068 if (x($data, "addr")) {
1069 $userid = $data["addr"];
1071 $userid = Protocol::formatMention($profile, $author);
1074 if (x($data, "name")) {
1075 $author = $data["name"];
1078 if (x($data, "micro")) {
1079 $avatar = $data["micro"];
1082 $preshare = trim($share[1]);
1083 if ($preshare != "") {
1084 $preshare .= "<br />";
1087 switch ($simplehtml) {
1089 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' <a href="' . $profile . '">' . $userid . "</a>: <br />»" . $share[3] . "«";
1092 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1095 $headline = '<b>' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . $userid . ':</b><br />';
1097 $text = trim($share[1]);
1103 if (stripos(normalise_link($link), 'http://twitter.com/') === 0) {
1104 $text .= '<br /><a href="' . $link . '">' . $link . '</a>';
1106 $text .= $headline . '<blockquote>' . trim($share[3]) . "</blockquote><br />";
1109 $text .= '<br /><a href="' . $link . '">[l]</a>';
1115 $headline = '<br /><b>' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8');
1116 $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $link, $userid, $posted);
1117 $headline .= ":</b><br />";
1119 $text = trim($share[1]);
1125 $text .= $headline . '<blockquote class="shared_content">' . trim($share[3]) . "</blockquote><br />";
1129 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1132 $text = $preshare . ">> @" . $userid_compact . ": <br />" . $share[3];
1134 case 7: // statusnet/GNU Social
1135 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3];
1138 $text = $preshare . "RT @" . $userid_compact . ": " . $share[3];
1140 case 9: // Google+/Facebook
1141 $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1144 $text .= "<br /><br />" . $link;
1148 // Transforms quoted tweets in rich attachments to avoid nested tweets
1149 if (stripos(normalise_link($link), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($link)) {
1151 $oembed = OEmbed::getHTML($link, $preshare);
1152 } catch (Exception $e) {
1153 $oembed = sprintf('[bookmark=%s]%s[/bookmark]', $link, $preshare);
1156 $text = $preshare . $oembed;
1158 $text = trim($share[1]) . "\n";
1160 $avatar = proxy_url($avatar, false, PROXY_SIZE_THUMB);
1162 $tpl = get_markup_template('shared_content.tpl');
1163 $text .= replace_macros($tpl, [
1164 '$profile' => $profile,
1165 '$avatar' => $avatar,
1166 '$author' => $author,
1168 '$posted' => $posted,
1169 '$content' => trim($share[3])
1178 private static function removePictureLinksCallback($match)
1180 $text = Cache::get($match[1]);
1182 if (is_null($text)) {
1183 $a = self::getApp();
1185 $stamp1 = microtime(true);
1187 $ch = @curl_init($match[1]);
1188 @curl_setopt($ch, CURLOPT_NOBODY, true);
1189 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1190 @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
1192 $curl_info = @curl_getinfo($ch);
1194 $a->save_timestamp($stamp1, "network");
1196 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1197 $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
1199 $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
1201 // if its not a picture then look if its a page that contains a picture link
1202 $body = Network::fetchUrl($match[1]);
1204 $doc = new DOMDocument();
1205 @$doc->loadHTML($body);
1206 $xpath = new DOMXPath($doc);
1207 $list = $xpath->query("//meta[@name]");
1208 foreach ($list as $node) {
1211 if ($node->attributes->length) {
1212 foreach ($node->attributes as $attribute) {
1213 $attr[$attribute->name] = $attribute->value;
1217 if (strtolower($attr["name"]) == "twitter:image") {
1218 $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
1222 Cache::set($match[1], $text);
1228 private static function expandLinksCallback($match)
1230 if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1231 return ($match[1] . "[url]" . $match[2] . "[/url]");
1233 return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1237 private static function cleanPictureLinksCallback($match)
1239 $text = Cache::get($match[1]);
1241 if (is_null($text)) {
1242 $a = self::getApp();
1244 $stamp1 = microtime(true);
1246 $ch = @curl_init($match[1]);
1247 @curl_setopt($ch, CURLOPT_NOBODY, true);
1248 @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1249 @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
1251 $curl_info = @curl_getinfo($ch);
1253 $a->save_timestamp($stamp1, "network");
1255 // if its a link to a picture then embed this picture
1256 if (substr($curl_info["content_type"], 0, 6) == "image/") {
1257 $text = "[img]" . $match[1] . "[/img]";
1259 $text = "[img]" . $match[2] . "[/img]";
1261 // if its not a picture then look if its a page that contains a picture link
1262 $body = Network::fetchUrl($match[1]);
1264 $doc = new DOMDocument();
1265 @$doc->loadHTML($body);
1266 $xpath = new DOMXPath($doc);
1267 $list = $xpath->query("//meta[@name]");
1268 foreach ($list as $node) {
1270 if ($node->attributes->length) {
1271 foreach ($node->attributes as $attribute) {
1272 $attr[$attribute->name] = $attribute->value;
1276 if (strtolower($attr["name"]) == "twitter:image") {
1277 $text = "[img]" . $attr["content"] . "[/img]";
1281 Cache::set($match[1], $text);
1287 public static function cleanPictureLinks($text)
1289 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1293 private static function textHighlightCallback($match)
1295 // Fallback in case the language doesn't exist
1296 $return = '[code]' . $match[2] . '[/code]';
1298 if (in_array(strtolower($match[1]),
1299 ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby',
1300 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh', 'bash'])
1302 $return = text_highlight($match[2], strtolower($match[1]));
1309 * @brief Converts a BBCode message to HTML message
1311 * BBcode 2 HTML was written by WAY2WEB.net
1312 * extended to work with Mistpark/Friendica - Mike Macgirvin
1314 * Simple HTML values meaning:
1315 * - 0: Friendica display
1317 * - 2: Used for Facebook, Google+, Windows Phone push, Friendica API
1318 * - 3: Used before converting to Markdown in bb2diaspora.php
1319 * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1321 * - 6: Used for Appnet
1322 * - 7: Used for dfrn, OStatus
1323 * - 8: Used for WP backlink text setting
1325 * @param string $text
1326 * @param bool $try_oembed
1327 * @param int $simple_html
1328 * @param bool $for_plaintext
1331 public static function convert($text, $try_oembed = true, $simple_html = false, $for_plaintext = false)
1333 $a = self::getApp();
1336 * preg_match_callback function to replace potential Oembed tags with Oembed content
1338 * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1340 * $match[2] = $title or absent
1342 $try_oembed_callback = function ($match)
1345 $title = defaults($match, 2, null);
1348 $return = OEmbed::getHTML($url, $title);
1349 } catch (Exception $ex) {
1350 $return = $match[0];
1356 // Hide all [noparse] contained bbtags by spacefying them
1357 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1359 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1360 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1361 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1363 // Remove the abstract element. It is a non visible element.
1364 $text = self::stripAbstract($text);
1366 // Move all spaces out of the tags
1367 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1368 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1370 // Extract the private images which use data urls since preg has issues with
1371 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1372 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1374 $extracted = self::extractImagesFromItemBody($text);
1375 $text = $extracted['body'];
1376 $saved_image = $extracted['images'];
1378 // If we find any event code, turn it into an event.
1379 // After we're finished processing the bbcode we'll
1380 // replace all of the event code with a reformatted version.
1382 $ev = Event::fromBBCode($text);
1384 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1385 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1387 $text = str_replace("<", "<", $text);
1388 $text = str_replace(">", ">", $text);
1390 // remove some newlines before the general conversion
1391 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1392 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1394 $text = preg_replace("/\n\[code\]/ism", "[code]", $text);
1395 $text = preg_replace("/\[\/code\]\n/ism", "[/code]", $text);
1397 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1399 $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1402 // Check for [code] text here, before the linefeeds are messed with.
1403 // The highlighter will unescape and re-escape the content.
1404 if (strpos($text, '[code=') !== false) {
1405 $text = preg_replace_callback("/\[code=(.*?)\](.*?)\[\/code\]/ism", 'self::textHighlightCallback', $text);
1407 // Convert new line chars to html <br /> tags
1409 // nlbr seems to be hopelessly messed up
1410 // $Text = nl2br($Text);
1412 // We'll emulate it.
1414 $text = trim($text);
1415 $text = str_replace("\r\n", "\n", $text);
1417 // removing multiplicated newlines
1418 if (Config::get("system", "remove_multiplicated_lines")) {
1419 $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1420 "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1421 $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1422 "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1425 $text = str_replace($search, $replace, $text);
1426 } while ($oldtext != $text);
1429 // Set up the parameters for a URL search string
1430 $URLSearchString = "^\[\]";
1431 // Set up the parameters for a MAIL search string
1432 $MAILSearchString = $URLSearchString;
1434 // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text
1435 if (!$for_plaintext) {
1436 // Autolink feature (thanks to http://code.seebz.net/p/autolink-php/)
1437 // Currently disabled, since the function is too greedy
1438 // $autolink_regex = "`([^\]\=\"']|^)(https?\://[^\s<]+[^\s<\.\)])`ism";
1439 $autolink_regex = "/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism";
1440 $text = preg_replace($autolink_regex, '$1[url]$2[/url]', $text);
1441 if ($simple_html == 7) {
1442 $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1443 $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
1446 $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text);
1447 $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1451 // Handle attached links or videos
1452 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1454 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1456 // Remove all hashtag addresses
1457 if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7])) {
1458 $text = preg_replace("/([#@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $text);
1459 } elseif ($simple_html == 3) {
1460 // The ! is converted to @ since Diaspora only understands the @
1461 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1462 '@<a href="$2">$3</a>',
1464 } elseif ($simple_html == 7) {
1465 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1466 '$1<span class="vcard"><a href="$2" class="url" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1468 } elseif (!$simple_html) {
1469 $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1470 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1474 // Bookmarks in red - will be converted to bookmarks in friendica
1475 $text = preg_replace("/#\^\[url\]([$URLSearchString]*)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1476 $text = preg_replace("/#\^\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1477 $text = preg_replace("/#\[url\=[$URLSearchString]*\]\^\[\/url\]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i",
1478 "[bookmark=$1]$2[/bookmark]", $text);
1480 if (in_array($simple_html, [2, 6, 7, 8, 9])) {
1481 $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1482 //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1483 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1486 if ($simple_html == 5) {
1487 $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1490 // Perform URL Search
1492 $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1495 if ($simple_html == 5) {
1496 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1498 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1501 // Handle Diaspora posts
1502 $text = preg_replace_callback(
1503 "&\[url=/posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1505 return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1509 // Server independent link to posts and comments
1510 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1511 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1512 $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
1514 $text = preg_replace("/([#])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1515 '$1<a href="' . System::baseUrl() . '/search?tag=$3" class="tag" title="$3">$3</a>', $text);
1517 $text = preg_replace("/\[url\=([$URLSearchString]*)\]#(.*?)\[\/url\]/ism",
1518 '#<a href="' . System::baseUrl() . '/search?tag=$2" class="tag" title="$2">$2</a>', $text);
1520 $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1521 $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1522 //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $Text);
1524 // Red compatibility, though the link can't be authenticated on Friendica
1525 $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1528 // we may need to restrict this further if it picks up too many strays
1529 // link acct:user@host to a webfinger profile redirector
1531 $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);
1533 // Perform MAIL Search
1534 $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1535 $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1537 // leave open the posibility of [map=something]
1538 // this is replaced in prepare_body() which has knowledge of the item location
1540 if (strpos($text, '[/map]') !== false) {
1541 $text = preg_replace_callback(
1542 "/\[map\](.*?)\[\/map\]/ism",
1543 function ($match) use ($simple_html) {
1544 return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1549 if (strpos($text, '[map=') !== false) {
1550 $text = preg_replace_callback(
1551 "/\[map=(.*?)\]/ism",
1552 function ($match) use ($simple_html) {
1553 return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1558 if (strpos($text, '[map]') !== false) {
1559 $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1562 // Check for headers
1563 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1564 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1565 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1566 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1567 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1568 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1570 // Check for paragraph
1571 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1573 // Check for bold text
1574 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1576 // Check for Italics text
1577 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1579 // Check for Underline text
1580 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1582 // Check for strike-through text
1583 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<strike>$1</strike>', $text);
1585 // Check for over-line text
1586 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1588 // Check for colored text
1589 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1591 // Check for sized text
1592 // [size=50] --> font-size: 50px (with the unit).
1593 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1594 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1596 // Check for centered text
1597 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1599 // Check for list text
1600 $text = str_replace("[*]", "<li>", $text);
1602 // Check for style sheet commands
1603 $text = preg_replace_callback(
1604 "(\[style=(.*?)\](.*?)\[\/style\])ism",
1606 return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
1611 // Check for CSS classes
1612 $text = preg_replace_callback(
1613 "(\[class=(.*?)\](.*?)\[\/class\])ism",
1615 return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
1620 // handle nested lists
1623 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1624 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1625 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1626 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1627 $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1628 $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1629 $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1630 $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1631 $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1632 $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1633 $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1634 $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1635 $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1636 $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1639 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1640 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1641 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1642 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1644 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1645 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1647 $text = str_replace('[hr]', '<hr />', $text);
1649 // This is actually executed in prepare_body()
1651 $text = str_replace('[nosmile]', '', $text);
1653 // Check for font change text
1654 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1656 // Declare the format for [code] layout
1658 $CodeLayout = '<code>$1</code>';
1659 // Check for [code] text
1660 $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text);
1662 // Declare the format for [spoiler] layout
1663 $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
1665 // Check for [spoiler] text
1666 // handle nested quotes
1668 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1669 $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
1672 // Check for [spoiler=Author] text
1674 $t_wrote = L10n::t('$1 wrote:');
1676 // handle nested quotes
1678 while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1679 $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1680 "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
1684 // Declare the format for [quote] layout
1685 $QuoteLayout = '<blockquote>$1</blockquote>';
1687 // Check for [quote] text
1688 // handle nested quotes
1690 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1691 $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1694 // Check for [quote=Author] text
1696 $t_wrote = L10n::t('$1 wrote:');
1698 // handle nested quotes
1700 while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1701 $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1702 "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
1707 // [img=widthxheight]image source[/img]
1708 $text = preg_replace_callback(
1709 "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1710 function ($matches) {
1711 if (strpos($matches[3], "data:image/") === 0) {
1715 $matches[3] = proxy_url($matches[3]);
1716 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1721 $text = preg_replace_callback("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism",
1722 function ($matches) {
1723 $matches[1] = proxy_url($matches[1]);
1724 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1725 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '">';
1729 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1730 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1733 // [img]pathtoimage[/img]
1734 $text = preg_replace_callback(
1735 "/\[img\](.*?)\[\/img\]/ism",
1736 function ($matches) {
1737 if (strpos($matches[1], "data:image/") === 0) {
1741 $matches[1] = proxy_url($matches[1]);
1742 return "[img]" . $matches[1] . "[/img]";
1747 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1748 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1751 $text = preg_replace_callback("/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
1752 function ($match) use ($simple_html) {
1753 return self::convertShare($match, $simple_html);
1756 $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);
1757 $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 />', $text);
1758 //$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);
1762 $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);
1763 $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);
1765 $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1766 $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1768 $text = preg_replace("/\[video\](.*?)\[\/video\]/",
1769 '<a href="$1" target="_blank">$1</a>', $text);
1770 $text = preg_replace("/\[audio\](.*?)\[\/audio\]/",
1771 '<a href="$1" target="_blank">$1</a>', $text);
1774 // html5 video and audio
1778 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1780 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1783 // Youtube extensions
1785 $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1786 $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1787 $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1790 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1791 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1792 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1795 $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);
1797 $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1798 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1802 $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1803 $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1806 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1807 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1810 $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);
1812 $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1813 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1817 $text = OEmbed::BBCode2HTML($text);
1819 // Avoid triple linefeeds through oembed
1820 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1822 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1823 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1824 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1825 // start which is always required). Allow desc with a missing summary for compatibility.
1827 if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) {
1828 $sub = Event::getHTML($ev, $simple_html);
1830 $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1831 $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1832 $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1833 $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1834 $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1835 $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1836 $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1839 // Replace non graphical smilies for external posts
1841 $text = Smilies::replace($text, false, true);
1844 // Replace inline code blocks
1845 $text = preg_replace_callback("|(?!<br[^>]*>)<code>([^<]*)</code>(?!<br[^>]*>)|ism",
1846 function ($match) use ($simple_html) {
1847 $return = '<key>' . $match[1] . '</key>';
1848 // Use <code> for Diaspora inline code blocks
1849 if ($simple_html === 3) {
1850 $return = '<code>' . $match[1] . '</code>';
1856 // Unhide all [noparse] contained bbtags unspacefying them
1857 // and triming the [noparse] tag.
1859 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1860 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1861 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1863 /// @todo What is the meaning of these lines?
1864 $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1865 $text = preg_replace('/\&\#039\;/', '\'', $text);
1867 // Currently deactivated, it made problems with " inside of alt texts.
1868 //$text = preg_replace('/\"\;/', '"', $text);
1870 // fix any escaped ampersands that may have been converted into links
1871 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1873 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1874 $allowed_src_protocols = ['http', 'redir', 'cid'];
1875 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1876 '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
1878 // sanitize href attributes (only whitelisted protocols URLs)
1879 // default value for backward compatibility
1880 $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']);
1882 // Always allowed protocol even if config isn't set or not including it
1883 $allowed_link_protocols[] = 'http';
1884 $allowed_link_protocols[] = 'redir/';
1886 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1887 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
1890 $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
1893 // Clean up the HTML by loading and saving the HTML with the DOM.
1894 // Bad structured html can break a whole page.
1895 // For performance reasons do it only with ativated item cache or at export.
1896 if (!$try_oembed || (get_itemcachepath() != "")) {
1897 $doc = new DOMDocument();
1898 $doc->preserveWhiteSpace = false;
1900 $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1902 $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1903 $encoding = '<?xml encoding="UTF-8">';
1904 @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
1905 $doc->encoding = 'UTF-8';
1906 $text = $doc->saveHTML();
1907 $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
1909 $text = str_replace('<br></li>', '</li>', $text);
1911 //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1914 // Clean up some useless linebreaks in lists
1915 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1916 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1917 //$Text = str_replace('</li><br />', '</li>', $Text);
1918 //$Text = str_replace('<br /><li>', '<li>', $Text);
1919 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1921 Addon::callHooks('bbcode', $text);
1927 * @brief Strips the "abstract" tag from the provided text
1929 * @param string $text The text with BBCode
1930 * @return string The same text - but without "abstract" element
1932 public static function stripAbstract($text)
1934 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1935 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1941 * @brief Returns the value of the "abstract" element
1943 * @param string $text The text that maybe contains the element
1944 * @param string $addon The addon for which the abstract is meant for
1945 * @return string The abstract
1947 private static function getAbstract($text, $addon = "")
1951 $addon = strtolower($addon);
1953 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1954 foreach ($results AS $result) {
1955 $abstracts[strtolower($result[1])] = $result[2];
1959 if (isset($abstracts[$addon])) {
1960 $abstract = $abstracts[$addon];
1963 if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1964 $abstract = $result[1];
1971 * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
1973 * @param array $match Matching values for the callback
1974 * @return string Replaced mention
1976 private static function bbCodeMention2DiasporaCallback($match)
1978 $contact = Contact::getDetailsByURL($match[3]);
1980 if (empty($contact['addr'])) {
1981 $contact = Probe::uri($match[3]);
1984 if (empty($contact['addr'])) {
1988 $mention = '@{' . $match[2] . '; ' . $contact['addr'] . '}';
1993 * @brief Converts a BBCode text into Markdown
1995 * This function converts a BBCode item body to be sent to Markdown-enabled
1996 * systems like Diaspora and Libertree
1998 * @param string $text
1999 * @param bool $for_diaspora Diaspora requires more changes than Libertree
2002 public static function toMarkdown($text, $for_diaspora = true)
2004 $a = self::getApp();
2006 $original_text = $text;
2008 // Since Diaspora is creating a summary for links, this function removes them before posting
2009 if ($for_diaspora) {
2010 $text = self::removeShareInformation($text);
2014 * Transform #tags, strip off the [url] and replace spaces with underscore
2016 $url_search_string = "^\[\]";
2017 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
2018 function ($matches) {
2019 return '#' . str_replace(' ', '_', $matches[2]);
2024 // Converting images with size parameters to simple images. Markdown doesn't know it.
2025 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2027 // Extracting multi-line code blocks before the whitespace processing/code highlighter in self::convert()
2030 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#is",
2031 function ($matches) use (&$codeblocks) {
2032 $return = $matches[0];
2033 if (strpos($matches[2], "\n") !== false) {
2034 $return = '#codeblock-' . count($codeblocks) . '#';
2036 $prefix = '````' . $matches[1] . PHP_EOL;
2037 $codeblocks[] = $prefix . trim($matches[2]) . PHP_EOL . '````';
2044 // Convert it to HTML - don't try oembed
2045 if ($for_diaspora) {
2046 $text = self::convert($text, false, 3);
2048 // Add all tags that maybe were removed
2049 if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2051 foreach ($tags[2] as $tag) {
2052 $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2053 if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2054 $tagline .= '#' . $tag . ' ';
2057 $text = $text . " " . $tagline;
2060 $text = self::convert($text, false, 4);
2063 // mask some special HTML chars from conversation to markdown
2064 $text = str_replace(['<', '>', '&'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text);
2066 // If a link is followed by a quote then there should be a newline before it
2067 // Maybe we should make this newline at every time before a quote.
2068 $text = str_replace(["</a><blockquote>"], ["</a><br><blockquote>"], $text);
2070 $stamp1 = microtime(true);
2072 // Now convert HTML to Markdown
2073 $converter = new HtmlConverter();
2074 $text = $converter->convert($text);
2076 // unmask the special chars back to HTML
2077 $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
2079 $a->save_timestamp($stamp1, "parser");
2081 // Libertree has a problem with escaped hashtags.
2082 $text = str_replace(['\#'], ['#'], $text);
2084 // Remove any leading or trailing whitespace, as this will mess up
2085 // the Diaspora signature verification and cause the item to disappear
2086 $text = trim($text);
2088 if ($for_diaspora) {
2089 $url_search_string = "^\[\]";
2090 $text = preg_replace_callback(
2091 "/([@]\[(.*?)\])\(([$url_search_string]*?)\)/ism",
2092 ['self', 'bbCodeMention2DiasporaCallback'],
2097 // Restore code blocks
2098 $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
2099 function ($matches) use ($codeblocks) {
2101 if (isset($codeblocks[intval($matches[1])])) {
2102 $return = $codeblocks[$matches[1]];
2109 Addon::callHooks('bb2diaspora', $text);