src/Content/Text/BBCode.php

   1 <?php
   2 /**
   3  * @copyright Copyright (C) 2010-2021, the Friendica project
   4  *
   5  * @license GNU AGPL version 3 or any later version
   6  *
   7  * This program is free software: you can redistribute it and/or modify
   8  * it under the terms of the GNU Affero General Public License as
   9  * published by the Free Software Foundation, either version 3 of the
  10  * License, or (at your option) any later version.
  11  *
  12  * This program is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15  * GNU Affero General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU Affero General Public License
  18  * along with this program.  If not, see <https://www.gnu.org/licenses/>.
  19  *
  20  */
  21
  22 namespace Friendica\Content\Text;
  23
  24 use DOMDocument;
  25 use DOMXPath;
  26 use Exception;
  27 use Friendica\Content\ContactSelector;
  28 use Friendica\Content\Item;
  29 use Friendica\Content\OEmbed;
  30 use Friendica\Content\PageInfo;
  31 use Friendica\Content\Smilies;
  32 use Friendica\Core\Hook;
  33 use Friendica\Core\Logger;
  34 use Friendica\Core\Protocol;
  35 use Friendica\Core\Renderer;
  36 use Friendica\DI;
  37 use Friendica\Model\Contact;
  38 use Friendica\Model\Event;
  39 use Friendica\Model\Photo;
  40 use Friendica\Model\Post;
  41 use Friendica\Model\Tag;
  42 use Friendica\Object\Image;
  43 use Friendica\Protocol\Activity;
  44 use Friendica\Util\Images;
  45 use Friendica\Util\Map;
  46 use Friendica\Util\ParseUrl;
  47 use Friendica\Util\Proxy as ProxyUtils;
  48 use Friendica\Util\Strings;
  49 use Friendica\Util\XML;
  50
  51 class BBCode
  52 {
  53         // Update this value to the current date whenever changes are made to BBCode::convert
  54         const VERSION = '2021-05-21';
  55
  56         const INTERNAL = 0;
  57         const EXTERNAL = 1;
  58         const API = 2;
  59         const DIASPORA = 3;
  60         const CONNECTORS = 4;
  61         const OSTATUS = 7;
  62         const TWITTER = 8;
  63         const BACKLINK = 8;
  64         const ACTIVITYPUB = 9;
  65
  66         const TOP_ANCHOR = '<br class="top-anchor">';
  67         const BOTTOM_ANCHOR = '<br class="button-anchor">';
  68         /**
  69          * Fetches attachment data that were generated the old way
  70          *
  71          * @param string $body Message body
  72          * @return array
  73          *                     'type' -> Message type ('link', 'video', 'photo')
  74          *                     'text' -> Text before the shared message
  75          *                     'after' -> Text after the shared message
  76          *                     'image' -> Preview image of the message
  77          *                     'url' -> Url to the attached message
  78          *                     'title' -> Title of the attachment
  79          *                     'description' -> Description of the attachment
  80          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
  81          */
  82         private static function getOldAttachmentData($body)
  83         {
  84                 $post = [];
  85
  86                 // Simplify image codes
  87                 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
  88
  89                 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
  90                         foreach ($attached as $data) {
  91                                 if (!in_array($data[1], ['type-link', 'type-video', 'type-photo'])) {
  92                                         continue;
  93                                 }
  94
  95                                 $post['type'] = substr($data[1], 5);
  96
  97                                 $pos = strpos($body, $data[0]);
  98                                 if ($pos > 0) {
  99                                         $post['text'] = trim(substr($body, 0, $pos));
 100                                         $post['after'] = trim(substr($body, $pos + strlen($data[0])));
 101                                 } else {
 102                                         $post['text'] = trim(str_replace($data[0], '', $body));
 103                                         $post['after'] = '';
 104                                 }
 105
 106                                 $attacheddata = $data[2];
 107
 108                                 if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
 109
 110                                         $picturedata = Images::getInfoFromURLCached($matches[1]);
 111
 112                                         if ($picturedata) {
 113                                                 if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
 114                                                         $post['image'] = $matches[1];
 115                                                 } else {
 116                                                         $post['preview'] = $matches[1];
 117                                                 }
 118                                         }
 119                                 }
 120
 121                                 if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
 122                                         $post['url'] = $matches[1];
 123                                         $post['title'] = $matches[2];
 124                                 }
 125                                 if (!empty($post['url']) && (in_array($post['type'], ['link', 'video']))
 126                                         && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
 127                                         $post['url'] = $matches[1];
 128                                 }
 129
 130                                 // Search for description
 131                                 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
 132                                         $post['description'] = $matches[1];
 133                                 }
 134                         }
 135                 }
 136                 return $post;
 137         }
 138
 139         /**
 140          * Fetches attachment data that were generated with the "attachment" element
 141          *
 142          * @param string $body Message body
 143          * @return array
 144          *                     'type' -> Message type ('link', 'video', 'photo')
 145          *                     'text' -> Text before the shared message
 146          *                     'after' -> Text after the shared message
 147          *                     'image' -> Preview image of the message
 148          *                     'url' -> Url to the attached message
 149          *                     'title' -> Title of the attachment
 150          *                     'description' -> Description of the attachment
 151          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
 152          */
 153         public static function getAttachmentData($body)
 154         {
 155                 $data = [
 156                         'type'          => '',
 157                         'text'          => '',
 158                         'after'         => '',
 159                         'image'         => null,
 160                         'url'           => '',
 161                         'author_name'   => '',
 162                         'author_url'    => '',
 163                         'provider_name' => '',
 164                         'provider_url'  => '',
 165                         'title'         => '',
 166                         'description'   => '',
 167                 ];
 168
 169                 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
 170                         return self::getOldAttachmentData($body);
 171                 }
 172
 173                 $attributes = $match[2];
 174
 175                 $data['text'] = trim($match[1]);
 176
 177                 foreach (['type', 'url', 'title', 'image', 'preview', 'publisher_name', 'publisher_url', 'author_name', 'author_url'] as $field) {
 178                         preg_match('/' . preg_quote($field, '/') . '=("|\')(.*?)\1/ism', $attributes, $matches);
 179                         $value = $matches[2] ?? '';
 180
 181                         if ($value != '') {
 182                                 switch ($field) {
 183                                         case 'publisher_name':
 184                                                 $data['provider_name'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
 185                                                 break;
 186                                         case 'publisher_url':
 187                                                 $data['provider_url'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
 188                                                 break;
 189                                         case 'author_name':
 190                                                 $data['author_name'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
 191                                                 if ($data['provider_name'] == $data['author_name']) {
 192                                                         $data['author_name'] = '';
 193                                                 }
 194                                                 break;
 195                                         case 'author_url':
 196                                                 $data['author_url'] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
 197                                                 if ($data['provider_url'] == $data['author_url']) {
 198                                                         $data['author_url'] = '';
 199                                                 }
 200                                                 break;
 201                                         case 'title':
 202                                                 $value = self::convert(html_entity_decode($value, ENT_QUOTES, 'UTF-8'), false, true);
 203                                                 $value = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
 204                                                 $value = str_replace(['[', ']'], ['&#91;', '&#93;'], $value);
 205                                                 $data['title'] = $value;
 206                                         default:
 207                                                 $data[$field] = html_entity_decode($value, ENT_QUOTES, 'UTF-8');
 208                                                 break;
 209                                 }
 210                         }
 211                 }
 212
 213                 if (!in_array($data['type'], ['link', 'audio', 'photo', 'video'])) {
 214                         return [];
 215                 }
 216
 217                 $data['description'] = trim($match[3]);
 218
 219                 $data['after'] = trim($match[4]);
 220
 221                 $parts = parse_url($data['url']);
 222                 if (!empty($parts['scheme']) && !empty($parts['host'])) {
 223                         if (empty($data['provider_name'])) {
 224                                 $data['provider_name'] = $parts['host'];
 225                         }
 226                         if (empty($data['provider_url']) || empty(parse_url($data['provider_url'], PHP_URL_SCHEME))) {
 227                                 $data['provider_url'] = $parts['scheme'] . '://' . $parts['host'];
 228
 229                                 if (!empty($parts['port'])) {
 230                                         $data['provider_url'] .= ':' . $parts['port'];
 231                                 }
 232                         }
 233                 }
 234
 235                 return $data;
 236         }
 237
 238         public static function getAttachedData($body, $item = [])
 239         {
 240                 /*
 241                 - text:
 242                 - type: link, video, photo
 243                 - title:
 244                 - url:
 245                 - image:
 246                 - description:
 247                 - (thumbnail)
 248                 */
 249
 250                 $has_title = !empty($item['title']);
 251                 $plink = $item['plink'] ?? '';
 252                 $post = self::getAttachmentData($body);
 253
 254                 // Get all linked images with alternative image description
 255                 if (preg_match_all("/\[img=([^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
 256                         foreach ($pictures as $picture) {
 257                                 if (Photo::isLocal($picture[1])) {
 258                                         $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2]];
 259                                 } else {
 260                                         $post['remote_images'][] = ['url' => $picture[1], 'description' => $picture[2]];
 261                                 }
 262                         }
 263                         if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
 264                                 $post['image_description'] = $post['images'][0]['description'];
 265                         }
 266                 }
 267
 268                 if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
 269                         foreach ($pictures as $picture) {
 270                                 if (Photo::isLocal($picture[1])) {
 271                                         $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => ''];
 272                                 } else {
 273                                         $post['remote_images'][] = ['url' => $picture[1], 'description' => ''];
 274                                 }
 275                         }
 276                 }
 277
 278                 // if nothing is found, it maybe having an image.
 279                 if (!isset($post['type'])) {
 280                         // Simplify image codes
 281                         $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
 282                         $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
 283                         $post['text'] = $body;
 284
 285                         if (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $body, $pictures, PREG_SET_ORDER)) {
 286                                 if ((count($pictures) == 1) && !$has_title) {
 287                                         if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
 288                                                 // Replace the preview picture with the real picture
 289                                                 $url = str_replace('-1.', '-0.', $pictures[0][2]);
 290                                                 $data = ['url' => $url, 'type' => 'photo'];
 291                                         } else {
 292                                                 // Checking, if the link goes to a picture
 293                                                 $data = ParseUrl::getSiteinfoCached($pictures[0][1]);
 294                                         }
 295
 296                                         // Workaround:
 297                                         // Sometimes photo posts to the own album are not detected at the start.
 298                                         // So we seem to cannot use the cache for these cases. That's strange.
 299                                         if (($data['type'] != 'photo') && strstr($pictures[0][1], "/photos/")) {
 300                                                 $data = ParseUrl::getSiteinfo($pictures[0][1]);
 301                                         }
 302
 303                                         if ($data['type'] == 'photo') {
 304                                                 $post['type'] = 'photo';
 305                                                 if (isset($data['images'][0])) {
 306                                                         $post['image'] = $data['images'][0]['src'];
 307                                                         $post['url'] = $data['url'];
 308                                                 } else {
 309                                                         $post['image'] = $data['url'];
 310                                                 }
 311
 312                                                 $post['preview'] = $pictures[0][2];
 313                                                 $post['text'] = trim(str_replace($pictures[0][0], '', $body));
 314                                         } else {
 315                                                 $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
 316                                                 if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
 317                                                         $post['type'] = 'photo';
 318                                                         $post['image'] = $pictures[0][1];
 319                                                         $post['preview'] = $pictures[0][2];
 320                                                         $post['text'] = trim(str_replace($pictures[0][0], '', $body));
 321                                                 }
 322                                         }
 323                                 } elseif (count($pictures) > 0) {
 324                                         $post['type'] = 'link';
 325                                         $post['url'] = $plink;
 326                                         $post['image'] = $pictures[0][2];
 327                                         $post['text'] = $body;
 328
 329                                         foreach ($pictures as $picture) {
 330                                                 $post['text'] = trim(str_replace($picture[0], '', $post['text']));
 331                                         }
 332                                 }
 333                         } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
 334                                 if ($has_title) {
 335                                         $post['type'] = 'link';
 336                                         $post['url'] = $plink;
 337                                 } else {
 338                                         $post['type'] = 'photo';
 339                                 }
 340
 341                                 $post['image'] = $pictures[0][1];
 342                                 $post['text'] = $body;
 343                                 foreach ($pictures as $picture) {
 344                                         $post['text'] = trim(str_replace($picture[0], '', $post['text']));
 345                                 }
 346                         }
 347
 348                         // Test for the external links
 349                         preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER);
 350                         preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $links2, PREG_SET_ORDER);
 351
 352                         $links = array_merge($links1, $links2);
 353
 354                         // If there is only a single one, then use it.
 355                         // This should cover link posts via API.
 356                         if ((count($links) == 1) && !isset($post['preview']) && !$has_title) {
 357                                 $post['type'] = 'link';
 358                                 $post['url'] = $links[0][1];
 359                         }
 360
 361                         // Simplify "video" element
 362                         $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
 363
 364                         // Now count the number of external media links
 365                         preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
 366                         preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
 367                         preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER);
 368                         preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $links4, PREG_SET_ORDER);
 369
 370                         // Add them to the other external links
 371                         $links = array_merge($links, $links1, $links2, $links3, $links4);
 372
 373                         // Are there more than one?
 374                         if (count($links) > 1) {
 375                                 // The post will be the type "text", which means a blog post
 376                                 unset($post['type']);
 377                                 $post['url'] = $plink;
 378                         }
 379
 380                         if (!isset($post['type'])) {
 381                                 $post['type'] = "text";
 382                                 $post['text'] = trim($body);
 383                         }
 384
 385                         if (($post['type'] == 'photo') && empty($post['images']) && !empty($post['remote_images'])) {
 386                                 $post['images'] = $post['remote_images'];
 387                                 $post['image'] = $post['images'][0]['url'];
 388                                 if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
 389                                         $post['image_description'] = $post['images'][0]['description'];
 390                                 }
 391                         }
 392                         unset($post['remote_images']);
 393                 } elseif (isset($post['url']) && ($post['type'] == 'video')) {
 394                         $data = ParseUrl::getSiteinfoCached($post['url']);
 395
 396                         if (isset($data['images'][0])) {
 397                                 $post['image'] = $data['images'][0]['src'];
 398                         }
 399                 }
 400
 401                 return $post;
 402         }
 403
 404         /**
 405          * Remove [attachment] BBCode and replaces it with a regular [url]
 406          *
 407          * @param string  $body
 408          * @param boolean $no_link_desc No link description
 409          *
 410          * @return string with replaced body
 411          */
 412         public static function removeAttachment($body, $no_link_desc = false)
 413         {
 414                 return preg_replace_callback("/\s*\[attachment (.*?)\](.*?)\[\/attachment\]\s*/ism",
 415                         function ($match) use ($no_link_desc) {
 416                                 $attach_data = self::getAttachmentData($match[0]);
 417                                 if (empty($attach_data['url'])) {
 418                                         return $match[0];
 419                                 } elseif (empty($attach_data['title']) || $no_link_desc) {
 420                                         return "\n[url]" . $attach_data['url'] . "[/url]\n";
 421                                 } else {
 422                                         return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
 423                                 }
 424                 }, $body);
 425         }
 426
 427         /**
 428          * Converts a BBCode text into plaintext
 429          *
 430          * @param      $text
 431          * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
 432          *
 433          * @return string
 434          */
 435         public static function toPlaintext($text, $keep_urls = true)
 436         {
 437                 // Remove pictures in advance to avoid unneeded proxy calls
 438                 $text = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $2 ', $text);
 439                 $text = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $text);
 440
 441                 // Remove attachment
 442                 $text = self::removeAttachment($text);
 443
 444                 $naked_text = HTML::toPlaintext(self::convert($text, false, 0, true), 0, !$keep_urls);
 445
 446                 return $naked_text;
 447         }
 448
 449         private static function proxyUrl($image, $simplehtml = self::INTERNAL, $uriid = 0)
 450         {
 451                 // Only send proxied pictures to API and for internal display
 452                 if (!in_array($simplehtml, [self::INTERNAL, self::API])) {
 453                         return $image;
 454                 } elseif ($uriid) {
 455                         return Post\Link::getByLink($uriid, $image);
 456                 } else {
 457                         return ProxyUtils::proxifyUrl($image);
 458                 }
 459         }
 460
 461         /**
 462          * This function changing the visual size (not the real size) of images.
 463          * The function does not work for pictures with an alternate text description.
 464          * This could only be changed by using some new "img" BBCode format.
 465          *
 466          * @param string $srctext The body with images
 467          * @return string The body with possibly scaled images
 468          */
 469         public static function scaleExternalImages(string $srctext)
 470         {
 471                 $s = $srctext;
 472
 473                 // Simplify image links
 474                 $s = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $s);
 475
 476                 $matches = null;
 477                 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
 478                 if ($c) {
 479                         foreach ($matches as $mtch) {
 480                                 Logger::info('scale_external_image', ['image' => $mtch[1]]);
 481
 482                                 $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
 483                                 if (stristr($mtch[1], $hostname)) {
 484                                         continue;
 485                                 }
 486
 487                                 $curlResult = DI::httpRequest()->get($mtch[1]);
 488                                 if (!$curlResult->isSuccess()) {
 489                                         continue;
 490                                 }
 491
 492                                 $i = $curlResult->getBody();
 493                                 $type = $curlResult->getContentType();
 494                                 $type = Images::getMimeTypeByData($i, $mtch[1], $type);
 495
 496                                 if ($i) {
 497                                         $Image = new Image($i, $type);
 498                                         if ($Image->isValid()) {
 499                                                 $orig_width = $Image->getWidth();
 500                                                 $orig_height = $Image->getHeight();
 501
 502                                                 if ($orig_width > 640 || $orig_height > 640) {
 503                                                         $Image->scaleDown(640);
 504                                                         $new_width = $Image->getWidth();
 505                                                         $new_height = $Image->getHeight();
 506                                                         Logger::info('External images scaled', ['orig_width' => $orig_width, 'new_width' => $new_width, 'orig_height' => $orig_height, 'new_height' => $new_height, 'match' => $mtch[0]]);
 507                                                         $s = str_replace(
 508                                                                 $mtch[0],
 509                                                                 '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
 510                                                                 . "\n",
 511                                                                 $s
 512                                                         );
 513                                                         Logger::info('New string', ['image' => $s]);
 514                                                 }
 515                                         }
 516                                 }
 517                         }
 518                 }
 519
 520                 return $s;
 521         }
 522
 523         /**
 524          * Truncates imported message body string length to max_import_size
 525          *
 526          * The purpose of this function is to apply system message length limits to
 527          * imported messages without including any embedded photos in the length
 528          *
 529          * @param string $body
 530          * @return string
 531          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
 532          */
 533         public static function limitBodySize($body)
 534         {
 535                 $maxlen = DI::config()->get('config', 'max_import_size', 0);
 536
 537                 // If the length of the body, including the embedded images, is smaller
 538                 // than the maximum, then don't waste time looking for the images
 539                 if ($maxlen && (strlen($body) > $maxlen)) {
 540
 541                         Logger::info('the total body length exceeds the limit', ['maxlen' => $maxlen, 'body_len' => strlen($body)]);
 542
 543                         $orig_body = $body;
 544                         $new_body = '';
 545                         $textlen = 0;
 546
 547                         $img_start = strpos($orig_body, '[img');
 548                         $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
 549                         $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
 550                         while (($img_st_close !== false) && ($img_end !== false)) {
 551
 552                                 $img_st_close++; // make it point to AFTER the closing bracket
 553                                 $img_end += $img_start;
 554                                 $img_end += strlen('[/img]');
 555
 556                                 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
 557                                         // This is an embedded image
 558
 559                                         if (($textlen + $img_start) > $maxlen) {
 560                                                 if ($textlen < $maxlen) {
 561                                                         Logger::info('the limit happens before an embedded image');
 562                                                         $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
 563                                                         $textlen = $maxlen;
 564                                                 }
 565                                         } else {
 566                                                 $new_body = $new_body . substr($orig_body, 0, $img_start);
 567                                                 $textlen += $img_start;
 568                                         }
 569
 570                                         $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
 571                                 } else {
 572
 573                                         if (($textlen + $img_end) > $maxlen) {
 574                                                 if ($textlen < $maxlen) {
 575                                                         Logger::info('the limit happens before the end of a non-embedded image');
 576                                                         $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
 577                                                         $textlen = $maxlen;
 578                                                 }
 579                                         } else {
 580                                                 $new_body = $new_body . substr($orig_body, 0, $img_end);
 581                                                 $textlen += $img_end;
 582                                         }
 583                                 }
 584                                 $orig_body = substr($orig_body, $img_end);
 585
 586                                 if ($orig_body === false) {
 587                                         // in case the body ends on a closing image tag
 588                                         $orig_body = '';
 589                                 }
 590
 591                                 $img_start = strpos($orig_body, '[img');
 592                                 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
 593                                 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
 594                         }
 595
 596                         if (($textlen + strlen($orig_body)) > $maxlen) {
 597                                 if ($textlen < $maxlen) {
 598                                         Logger::info('the limit happens after the end of the last image');
 599                                         $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
 600                                 }
 601                         } else {
 602                                 Logger::info('the text size with embedded images extracted did not violate the limit');
 603                                 $new_body = $new_body . $orig_body;
 604                         }
 605
 606                         return $new_body;
 607                 } else {
 608                         return $body;
 609                 }
 610         }
 611
 612         /**
 613          * Processes [attachment] tags
 614          *
 615          * Note: Can produce a [bookmark] tag in the returned string
 616          *
 617          * @param string  $text
 618          * @param integer $simplehtml
 619          * @param bool    $tryoembed
 620          * @param array   $data
 621          * @return string
 622          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
 623          */
 624         public static function convertAttachment($text, $simplehtml = self::INTERNAL, $tryoembed = true, array $data = [])
 625         {
 626                 $data = $data ?: self::getAttachmentData($text);
 627                 if (empty($data) || empty($data['url'])) {
 628                         return $text;
 629                 }
 630
 631                 $stamp1 = microtime(true);
 632
 633                 if (isset($data['title'])) {
 634                         $data['title'] = strip_tags($data['title']);
 635                         $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
 636                 } else {
 637                         $data['title'] = null;
 638                 }
 639
 640                 if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
 641                         $data['preview'] = $data['image'];
 642                         $data['image'] = '';
 643                 }
 644
 645                 $return = '';
 646                 try {
 647                         if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
 648                                 $return = OEmbed::getHTML($data['url'], $data['title']);
 649                         } else {
 650                                 throw new Exception('OEmbed is disabled for this attachment.');
 651                         }
 652                 } catch (Exception $e) {
 653                         $data['title'] = ($data['title'] ?? '') ?: $data['url'];
 654
 655                         if ($simplehtml != self::CONNECTORS) {
 656                                 $return = sprintf('<div class="type-%s">', $data['type']);
 657                         }
 658
 659                         if (!empty($data['title']) && !empty($data['url'])) {
 660                                 if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
 661                                         $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
 662                                 } else {
 663                                         if (!empty($data['image'])) {
 664                                                 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br>', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
 665                                         } elseif (!empty($data['preview'])) {
 666                                                 $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br>', $data['url'], self::proxyUrl($data['preview'], $simplehtml), $data['title']);
 667                                         }
 668                                         $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
 669                                 }
 670                         }
 671
 672                         if (!empty($data['description']) && $data['description'] != $data['title']) {
 673                                 // Sanitize the HTML
 674                                 $return .= sprintf('<blockquote>%s</blockquote>', trim(HTML::purify($data['description'])));
 675                         }
 676
 677                         if (!empty($data['provider_url']) && !empty($data['provider_name'])) {
 678                                 if (!empty($data['author_name'])) {
 679                                         $return .= sprintf('<sup><a href="%s">%s (%s)</a></sup>', $data['provider_url'], $data['author_name'], $data['provider_name']);
 680                                 } else {
 681                                         $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['provider_url'], $data['provider_name']);
 682                                 }
 683                         }
 684
 685                         if ($simplehtml != self::CONNECTORS) {
 686                                 $return .= '</div>';
 687                         }
 688                 }
 689
 690                 DI::profiler()->saveTimestamp($stamp1, 'rendering');
 691                 return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
 692         }
 693
 694         public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
 695         {
 696                 $data = self::getAttachmentData($Text);
 697
 698                 if (!$data) {
 699                         return $Text;
 700                 } elseif ($nolink) {
 701                         return $data['text'] . ($data['after'] ?? '');
 702                 }
 703
 704                 $title = htmlentities($data['title'] ?? '', ENT_QUOTES, 'UTF-8', false);
 705                 $text = htmlentities($data['text'], ENT_QUOTES, 'UTF-8', false);
 706                 if ($plaintext || (($title != '') && strstr($text, $title))) {
 707                         $data['title'] = $data['url'];
 708                 } elseif (($text != '') && strstr($title, $text)) {
 709                         $data['text'] = $data['title'];
 710                         $data['title'] = $data['url'];
 711                 }
 712
 713                 if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
 714                         return $data['title'] . $data['after'];
 715                 }
 716
 717                 // If the link already is included in the post, don't add it again
 718                 if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
 719                         return $data['text'] . $data['after'];
 720                 }
 721
 722                 $text = $data['text'];
 723
 724                 if (!empty($data['url']) && !empty($data['title'])) {
 725                         $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
 726                 } elseif (!empty($data['url'])) {
 727                         $text .= "\n[url]" . $data['url'] . '[/url]';
 728                 }
 729
 730                 return $text . "\n" . $data['after'];
 731         }
 732
 733         /**
 734          * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
 735          *
 736          * @param array $match Array with the matching values
 737          * @return string reformatted link including HTML codes
 738          */
 739         private static function convertUrlForActivityPubCallback($match)
 740         {
 741                 $url = $match[1];
 742
 743                 if (isset($match[2]) && ($match[1] != $match[2])) {
 744                         return $match[0];
 745                 }
 746
 747                 $parts = parse_url($url);
 748                 if (!isset($parts['scheme'])) {
 749                         return $match[0];
 750                 }
 751
 752                 return self::convertUrlForActivityPub($url);
 753         }
 754
 755         /**
 756          * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
 757          *
 758          * @param string $url URL that is about to be reformatted
 759          * @return string reformatted link including HTML codes
 760          */
 761         private static function convertUrlForActivityPub($url)
 762         {
 763                 $html = '<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>';
 764                 return sprintf($html, $url, self::getStyledURL($url));
 765         }
 766
 767         /**
 768          * Converts an URL in a nicer format (without the scheme and possibly shortened)
 769          *
 770          * @param string $url URL that is about to be reformatted
 771          * @return string reformatted link
 772          */
 773         private static function getStyledURL($url)
 774         {
 775                 $parts = parse_url($url);
 776                 $scheme = $parts['scheme'] . '://';
 777                 $styled_url = str_replace($scheme, '', $url);
 778
 779                 if (strlen($styled_url) > 30) {
 780                         $styled_url = substr($styled_url, 0, 30) . "…";
 781                 }
 782
 783                 return $styled_url;
 784         }
 785
 786         /*
 787          * [noparse][i]italic[/i][/noparse] turns into
 788          * [noparse][ i ]italic[ /i ][/noparse],
 789          * to hide them from parser.
 790          */
 791         private static function escapeNoparseCallback($match)
 792         {
 793                 $whole_match = $match[0];
 794                 $captured = $match[1];
 795                 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
 796                 $new_str = str_replace($captured, $spacefied, $whole_match);
 797                 return $new_str;
 798         }
 799
 800         /*
 801          * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
 802          * now turns back and the [noparse] tags are trimed
 803          * returning [i]italic[/i]
 804          */
 805         private static function unescapeNoparseCallback($match)
 806         {
 807                 $captured = $match[1];
 808                 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
 809                 return $unspacefied;
 810         }
 811
 812         /**
 813          * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
 814          * occurrences
 815          *
 816          * @param string $text        Text to search
 817          * @param string $name        Tag name
 818          * @param int    $occurrences Number of first occurrences to skip
 819          * @return boolean|array
 820          */
 821         public static function getTagPosition($text, $name, $occurrences = 0)
 822         {
 823                 if ($occurrences < 0) {
 824                         $occurrences = 0;
 825                 }
 826
 827                 $start_open = -1;
 828                 for ($i = 0; $i <= $occurrences; $i++) {
 829                         if ($start_open !== false) {
 830                                 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
 831                         }
 832                 }
 833
 834                 if ($start_open === false) {
 835                         return false;
 836                 }
 837
 838                 $start_equal = strpos($text, '=', $start_open);
 839                 $start_close = strpos($text, ']', $start_open);
 840
 841                 if ($start_close === false) {
 842                         return false;
 843                 }
 844
 845                 $start_close++;
 846
 847                 $end_open = strpos($text, '[/' . $name . ']', $start_close);
 848
 849                 if ($end_open === false) {
 850                         return false;
 851                 }
 852
 853                 $res = [
 854                         'start' => [
 855                                 'open' => $start_open,
 856                                 'close' => $start_close
 857                         ],
 858                         'end' => [
 859                                 'open' => $end_open,
 860                                 'close' => $end_open + strlen('[/' . $name . ']')
 861                         ],
 862                 ];
 863
 864                 if ($start_equal !== false) {
 865                         $res['start']['equal'] = $start_equal + 1;
 866                 }
 867
 868                 return $res;
 869         }
 870
 871         /**
 872          * Performs a preg_replace within the boundaries of all named BBCode tags in a text
 873          *
 874          * @param string $pattern Preg pattern string
 875          * @param string $replace Preg replace string
 876          * @param string $name    BBCode tag name
 877          * @param string $text    Text to search
 878          * @return string
 879          */
 880         public static function pregReplaceInTag($pattern, $replace, $name, $text)
 881         {
 882                 $occurrences = 0;
 883                 $pos = self::getTagPosition($text, $name, $occurrences);
 884                 while ($pos !== false && $occurrences++ < 1000) {
 885                         $start = substr($text, 0, $pos['start']['open']);
 886                         $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
 887                         $end = substr($text, $pos['end']['close']);
 888                         if ($end === false) {
 889                                 $end = '';
 890                         }
 891
 892                         $subject = preg_replace($pattern, $replace, $subject);
 893                         $text = $start . $subject . $end;
 894
 895                         $pos = self::getTagPosition($text, $name, $occurrences);
 896                 }
 897
 898                 return $text;
 899         }
 900
 901         private static function extractImagesFromItemBody($body)
 902         {
 903                 $saved_image = [];
 904                 $orig_body = $body;
 905                 $new_body = '';
 906
 907                 $cnt = 0;
 908                 $img_start = strpos($orig_body, '[img');
 909                 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
 910                 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
 911                 while (($img_st_close !== false) && ($img_end !== false)) {
 912                         $img_st_close++; // make it point to AFTER the closing bracket
 913                         $img_end += $img_start;
 914
 915                         if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
 916                                 // This is an embedded image
 917                                 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
 918                                 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
 919
 920                                 $cnt++;
 921                         } else {
 922                                 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
 923                         }
 924
 925                         $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
 926
 927                         if ($orig_body === false) {
 928                                 // in case the body ends on a closing image tag
 929                                 $orig_body = '';
 930                         }
 931
 932                         $img_start = strpos($orig_body, '[img');
 933                         $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
 934                         $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
 935                 }
 936
 937                 $new_body = $new_body . $orig_body;
 938
 939                 return ['body' => $new_body, 'images' => $saved_image];
 940         }
 941
 942         private static function interpolateSavedImagesIntoItemBody($uriid, $body, array $images)
 943         {
 944                 $newbody = $body;
 945
 946                 $cnt = 0;
 947                 foreach ($images as $image) {
 948                         // We're depending on the property of 'foreach' (specified on the PHP website) that
 949                         // it loops over the array starting from the first element and going sequentially
 950                         // to the last element
 951                         $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
 952                                 '<img src="' . self::proxyUrl($image, self::INTERNAL, $uriid) . '" alt="' . DI::l10n()->t('Image/photo') . '" />', $newbody);
 953                         $cnt++;
 954                 }
 955
 956                 return $newbody;
 957         }
 958
 959         /**
 960          *
 961          * @param  string   $text     A BBCode string
 962          * @return array share attributes
 963          */
 964         public static function fetchShareAttributes($text)
 965         {
 966                 // See Issue https://github.com/friendica/friendica/issues/10454
 967                 // Hashtags in usernames are expanded to links. This here is a quick fix.
 968                 $text = preg_replace('/([@!#])\[url\=.*?\](.*?)\[\/url\]/ism', '$1$2', $text);
 969
 970                 $attributes = [];
 971                 if (!preg_match("/(.*?)\[share(.*?)\](.*)\[\/share\]/ism", $text, $matches)) {
 972                         return $attributes;
 973                 }
 974
 975                 $attribute_string = $matches[2];
 976                 foreach (['author', 'profile', 'avatar', 'link', 'posted', 'guid'] as $field) {
 977                         preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
 978                         $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
 979                 }
 980                 return $attributes;
 981         }
 982
 983         /**
 984          * This function converts a [share] block to text according to a provided callback function whose signature is:
 985          *
 986          * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string
 987          *
 988          * Where:
 989          * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact
 990          * data lookup
 991          * - $author_contact is a contact record array
 992          * - $content is the inner content of the [share] block
 993          * - $is_quote_share indicates whether there's any content before the [share] block
 994          * - Return value is the string that should replace the [share] block in the provided text
 995          *
 996          * This function is intended to be used by addon connector to format a share block like the target network is expecting it.
 997          *
 998          * @param  string   $text     A BBCode string
 999          * @param  callable $callback
1000          * @return string The BBCode string with all [share] blocks replaced
1001          */
1002         public static function convertShare($text, callable $callback, int $uriid = 0)
1003         {
1004                 $return = preg_replace_callback(
1005                         "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
1006                         function ($match) use ($callback, $uriid) {
1007                                 $attribute_string = $match[2];
1008                                 $attributes = [];
1009                                 foreach (['author', 'profile', 'avatar', 'link', 'posted', 'guid'] as $field) {
1010                                         preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
1011                                         $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
1012                                 }
1013
1014                                 $author_contact = Contact::getByURL($attributes['profile'], false, ['id', 'url', 'addr', 'name', 'micro']);
1015                                 $author_contact['url'] = ($author_contact['url'] ?? $attributes['profile']);
1016                                 $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']);
1017
1018                                 $attributes['author']   = ($author_contact['name']  ?? '') ?: $attributes['author'];
1019                                 $attributes['avatar']   = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
1020                                 $attributes['profile']  = ($author_contact['url']   ?? '') ?: $attributes['profile'];
1021
1022                                 if (!empty($author_contact['id'])) {
1023                                         $attributes['avatar'] = Contact::getAvatarUrlForId($author_contact['id'], ProxyUtils::SIZE_THUMB);
1024                                 } elseif ($attributes['avatar'] && $uriid) {
1025                                         $attributes['avatar'] = Post\Link::getByLink($uriid, $attributes['avatar'], ProxyUtils::SIZE_THUMB);
1026                                 } elseif ($attributes['avatar']) {
1027                                         $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], ProxyUtils::SIZE_THUMB);
1028                                 }
1029
1030                                 $content = preg_replace(Strings::autoLinkRegEx(), '<a href="$1">$1</a>', $match[3]);
1031
1032                                 return $match[1] . $callback($attributes, $author_contact, $content, trim($match[1]) != '');
1033                         },
1034                         $text
1035                 );
1036
1037                 return $return;
1038         }
1039
1040         /**
1041          * Default [share] tag conversion callback
1042          *
1043          * Note: Can produce a [bookmark] tag in the output
1044          *
1045          * @see BBCode::convertShare()
1046          * @param array   $attributes     [share] block attribute values
1047          * @param array   $author_contact Contact row of the shared author
1048          * @param string  $content        Inner content of the [share] block
1049          * @param boolean $is_quote_share Whether there is content before the [share] block
1050          * @param integer $simplehtml     Mysterious integer value depending on the target network/formatting style
1051          * @return string
1052          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1053          */
1054         private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml)
1055         {
1056                 $mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
1057
1058                 switch ($simplehtml) {
1059                         case self::API:
1060                                 $text = ($is_quote_share? '<br>' : '') .
1061                                 '<p><b><a href="' . $attributes['link'] . '">' . html_entity_decode('&#x2672; ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . "</a>:</b> </p>\n" .
1062                                 '<blockquote class="shared_content">' . $content . '</blockquote>';
1063                                 break;
1064                         case self::DIASPORA:
1065                                 if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
1066                                         $text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
1067                                 } else {
1068                                         $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a>:</b></p>' . "\n";
1069
1070                                         if (!empty($attributes['posted']) && !empty($attributes['link'])) {
1071                                                 $headline = '<p><b>♲ <a href="' . $attributes['profile'] . '">' . $attributes['author'] . '</a></b> - <a href="' . $attributes['link'] . '">' . $attributes['posted'] . ' GMT</a></p>' . "\n";
1072                                         }
1073
1074                                         $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote>' . trim($content) . '</blockquote>' . "\n";
1075
1076                                         if (empty($attributes['posted']) && !empty($attributes['link'])) {
1077                                                 $text .= '<p><a href="' . $attributes['link'] . '">[Source]</a></p>' . "\n";
1078                                         }
1079                                 }
1080
1081                                 break;
1082                         case self::CONNECTORS:
1083                                 $headline = '<p><b>' . html_entity_decode('&#x2672; ', ENT_QUOTES, 'UTF-8');
1084                                 $headline .= DI::l10n()->t('<a href="%1$s" target="_blank" rel="noopener noreferrer">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
1085                                 $headline .= ':</b></p>' . "\n";
1086
1087                                 $text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
1088
1089                                 break;
1090                         case self::OSTATUS:
1091                                 $text = ($is_quote_share? '<br>' : '') . '<p>' . html_entity_decode('&#x2672; ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
1092                                 break;
1093                         case self::ACTIVITYPUB:
1094                                 $author = '@<span class="vcard"><a href="' . $author_contact['url'] . '" class="url u-url mention" title="' . $author_contact['addr'] . '"><span class="fn nickname mention">' . $author_contact['addr'] . '</span></a>:</span>';
1095                                 $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('&#x2672;', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
1096                                 break;
1097                         default:
1098                                 $text = ($is_quote_share? "\n" : '');
1099
1100                                 $contact = Contact::getByURL($attributes['profile'], false, ['network']);
1101                                 $network = $contact['network'] ?? Protocol::PHANTOM;
1102
1103                                 $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
1104                                 $text .= Renderer::replaceMacros($tpl, [
1105                                         '$profile'      => $attributes['profile'],
1106                                         '$avatar'       => $attributes['avatar'],
1107                                         '$author'       => $attributes['author'],
1108                                         '$link'         => $attributes['link'],
1109                                         '$link_title'   => DI::l10n()->t('Link to source'),
1110                                         '$posted'       => $attributes['posted'],
1111                                         '$guid'         => $attributes['guid'],
1112                                         '$network_name' => ContactSelector::networkToName($network, $attributes['profile']),
1113                                         '$network_icon' => ContactSelector::networkToIcon($network, $attributes['profile']),
1114                                         '$content'      => self::TOP_ANCHOR . self::setMentions(trim($content), 0, $network) . self::BOTTOM_ANCHOR,
1115                                 ]);
1116                                 break;
1117                 }
1118
1119                 return $text;
1120         }
1121
1122         private static function removePictureLinksCallback($match)
1123         {
1124                 $cache_key = 'remove:' . $match[1];
1125                 $text = DI::cache()->get($cache_key);
1126
1127                 if (is_null($text)) {
1128                         $curlResult = DI::httpRequest()->head($match[1], ['timeout' => DI::config()->get('system', 'xrd_timeout')]);
1129                         if ($curlResult->isSuccess()) {
1130                                 $mimetype = $curlResult->getHeader('Content-Type');
1131                         } else {
1132                                 $mimetype = '';
1133                         }
1134
1135                         if (substr($mimetype, 0, 6) == 'image/') {
1136                                 $text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
1137                         } else {
1138                                 $text = "[url=" . $match[2] . ']' . $match[2] . "[/url]";
1139
1140                                 // if its not a picture then look if its a page that contains a picture link
1141                                 $body = DI::httpRequest()->fetch($match[1]);
1142                                 if (empty($body)) {
1143                                         DI::cache()->set($cache_key, $text);
1144                                         return $text;
1145                                 }
1146
1147                                 $doc = new DOMDocument();
1148                                 @$doc->loadHTML($body);
1149                                 $xpath = new DOMXPath($doc);
1150                                 $list = $xpath->query("//meta[@name]");
1151                                 foreach ($list as $node) {
1152                                         $attr = [];
1153
1154                                         if ($node->attributes->length) {
1155                                                 foreach ($node->attributes as $attribute) {
1156                                                         $attr[$attribute->name] = $attribute->value;
1157                                                 }
1158                                         }
1159
1160                                         if (strtolower($attr['name']) == 'twitter:image') {
1161                                                 $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
1162                                         }
1163                                 }
1164                         }
1165                         DI::cache()->set($cache_key, $text);
1166                 }
1167
1168                 return $text;
1169         }
1170
1171         private static function expandLinksCallback($match)
1172         {
1173                 if (($match[3] == '') || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1174                         return ($match[1] . "[url]" . $match[2] . "[/url]");
1175                 } else {
1176                         return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1177                 }
1178         }
1179
1180         private static function cleanPictureLinksCallback($match)
1181         {
1182                 // When the picture link is the own photo path then we can avoid fetching the link
1183                 $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
1184                 if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
1185                         if (!empty($match[3])) {
1186                                 $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
1187                         } else {
1188                                 $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
1189                         }
1190                         return $text;
1191                 }
1192
1193                 $cache_key = 'clean:' . $match[1];
1194                 $text = DI::cache()->get($cache_key);
1195                 if (!is_null($text)) {
1196                         return $text;
1197                 }
1198
1199                 $curlResult = DI::httpRequest()->head($match[1], ['timeout' => DI::config()->get('system', 'xrd_timeout')]);
1200                 if ($curlResult->isSuccess()) {
1201                         $mimetype = $curlResult->getHeader('Content-Type');
1202                 } else {
1203                         $mimetype = '';
1204                 }
1205
1206                 // if its a link to a picture then embed this picture
1207                 if (substr($mimetype, 0, 6) == 'image/') {
1208                         $text = '[img]' . $match[1] . '[/img]';
1209                 } else {
1210                         if (!empty($match[3])) {
1211                                 $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
1212                         } else {
1213                                 $text = '[img]' . $match[2] . '[/img]';
1214                         }
1215
1216                         // if its not a picture then look if its a page that contains a picture link
1217                         $body = DI::httpRequest()->fetch($match[1]);
1218                         if (empty($body)) {
1219                                 DI::cache()->set($cache_key, $text);
1220                                 return $text;
1221                         }
1222
1223                         $doc = new DOMDocument();
1224                         @$doc->loadHTML($body);
1225                         $xpath = new DOMXPath($doc);
1226                         $list = $xpath->query("//meta[@name]");
1227                         foreach ($list as $node) {
1228                                 $attr = [];
1229                                 if ($node->attributes->length) {
1230                                         foreach ($node->attributes as $attribute) {
1231                                                 $attr[$attribute->name] = $attribute->value;
1232                                         }
1233                                 }
1234
1235                                 if (strtolower($attr['name']) == "twitter:image") {
1236                                         if (!empty($match[3])) {
1237                                                 $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
1238                                         } else {
1239                                                 $text = "[img]" . $attr['content'] . "[/img]";
1240                                         }
1241                                 }
1242                         }
1243                 }
1244                 DI::cache()->set($cache_key, $text);
1245
1246                 return $text;
1247         }
1248
1249         public static function cleanPictureLinks($text)
1250         {
1251                 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1252                 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $return);
1253                 return $return;
1254         }
1255
1256         public static function removeLinks(string $bbcode)
1257         {
1258                 $bbcode = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $1 ', $bbcode);
1259                 $bbcode = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $bbcode);
1260
1261                 $bbcode = preg_replace('/[@!#]\[url\=.*?\].*?\[\/url\]/ism', '', $bbcode);
1262                 $bbcode = preg_replace("/\[url=[^\[\]]*\](.*)\[\/url\]/Usi", ' $1 ', $bbcode);
1263                 $bbcode = preg_replace('/[@!#]?\[url.*?\[\/url\]/ism', '', $bbcode);
1264                 return $bbcode;
1265         }
1266
1267         /**
1268          * Converts a BBCode message to HTML message
1269          *
1270          * BBcode 2 HTML was written by WAY2WEB.net
1271          * extended to work with Mistpark/Friendica - Mike Macgirvin
1272          *
1273          * Simple HTML values meaning:
1274          * - 0: Friendica display
1275          * - 1: Unused
1276          * - 2: Used for Windows Phone push, Friendica API
1277          * - 3: Used before converting to Markdown in bb2diaspora.php
1278          * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1279          * - 5: Unused
1280          * - 6: Unused
1281          * - 7: Used for dfrn, OStatus
1282          * - 8: Used for Twitter, WP backlink text setting
1283          * - 9: ActivityPub
1284          *
1285          * @param string $text
1286          * @param bool   $try_oembed
1287          * @param int    $simple_html
1288          * @param bool   $for_plaintext
1289          * @param int    $uriid
1290          * @return string
1291          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1292          */
1293         public static function convert(string $text = null, $try_oembed = true, $simple_html = self::INTERNAL, $for_plaintext = false, $uriid = 0)
1294         {
1295                 // Accounting for null default column values
1296                 if (is_null($text) || $text === '') {
1297                         return '';
1298                 }
1299
1300                 Hook::callAll('bbcode', $text);
1301
1302                 $a = DI::app();
1303
1304                 $text = self::performWithEscapedTags($text, ['code'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a, $uriid) {
1305                         $text = self::performWithEscapedTags($text, ['noparse', 'nobb', 'pre'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a, $uriid) {
1306                                 /*
1307                                  * preg_match_callback function to replace potential Oembed tags with Oembed content
1308                                  *
1309                                  * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1310                                  * $match[1] = $url
1311                                  * $match[2] = $title or absent
1312                                  */
1313                                 $try_oembed_callback = function ($match)
1314                                 {
1315                                         $url = $match[1];
1316                                         $title = $match[2] ?? null;
1317
1318                                         try {
1319                                                 $return = OEmbed::getHTML($url, $title);
1320                                         } catch (Exception $ex) {
1321                                                 $return = $match[0];
1322                                         }
1323
1324                                         return $return;
1325                                 };
1326
1327                                 // Remove the abstract element. It is a non visible element.
1328                                 $text = self::stripAbstract($text);
1329
1330                                 // Line ending normalisation
1331                                 $text = str_replace("\r\n", "\n", $text);
1332
1333                                 // Move new lines outside of tags
1334                                 $text = preg_replace("#\[(\w*)](\n*)#ism", '$2[$1]', $text);
1335                                 $text = preg_replace("#(\n*)\[/(\w*)]#ism", '[/$2]$1', $text);
1336
1337                                 // Extract the private images which use data urls since preg has issues with
1338                                 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1339                                 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1340
1341                                 $extracted = self::extractImagesFromItemBody($text);
1342                                 $text = $extracted['body'];
1343                                 $saved_image = $extracted['images'];
1344
1345                                 // If we find any event code, turn it into an event.
1346                                 // After we're finished processing the bbcode we'll
1347                                 // replace all of the event code with a reformatted version.
1348
1349                                 $ev = Event::fromBBCode($text);
1350
1351                                 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1352                                 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1353
1354                                 $text = str_replace("<", "&lt;", $text);
1355                                 $text = str_replace(">", "&gt;", $text);
1356
1357                                 // remove some newlines before the general conversion
1358                                 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1359                                 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1360
1361                                 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1362                                 if (!$try_oembed) {
1363                                         $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1364                                 }
1365
1366                                 // Remove linefeeds inside of the table elements. See issue #6799
1367                                 $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
1368                                         "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
1369                                         "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
1370                                         "\n[hr]", "[hr]\n", " [hr]", "[hr] ",
1371                                         "\n[attachment ", " [attachment ", "\n[/attachment]", "[/attachment]\n", " [/attachment]", "[/attachment] ",
1372                                         "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "];
1373                                 $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
1374                                         "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
1375                                         "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
1376                                         "[hr]", "[hr]", "[hr]", "[hr]",
1377                                         "[attachment ", "[attachment ", "[/attachment]", "[/attachment]", "[/attachment]", "[/attachment]",
1378                                         "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"];
1379                                 do {
1380                                         $oldtext = $text;
1381                                         $text = str_replace($search, $replace, $text);
1382                                 } while ($oldtext != $text);
1383
1384                                 // Replace these here only once
1385                                 $search = ["\n[table]", "[/table]\n"];
1386                                 $replace = ["[table]", "[/table]"];
1387                                 $text = str_replace($search, $replace, $text);
1388
1389                                 // Trim new lines regardless of the system.remove_multiplicated_lines config value
1390                                 $text = trim($text, "\n");
1391
1392                                 // removing multiplicated newlines
1393                                 if (DI::config()->get('system', 'remove_multiplicated_lines')) {
1394                                         $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[*]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1395                                                         "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1396                                         $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[*]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1397                                                         "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1398                                         do {
1399                                                 $oldtext = $text;
1400                                                 $text = str_replace($search, $replace, $text);
1401                                         } while ($oldtext != $text);
1402                                 }
1403
1404                                 // Add HTML new lines
1405                                 $text = str_replace("\n", '<br>', $text);
1406
1407                                 /// @todo Have a closer look at the different html modes
1408                                 // Handle attached links or videos
1409                                 if ($simple_html == self::ACTIVITYPUB) {
1410                                         $text = self::removeAttachment($text);
1411                                 } elseif (!in_array($simple_html, [self::INTERNAL, self::EXTERNAL, self::CONNECTORS])) {
1412                                         $text = self::removeAttachment($text, true);
1413                                 } else {
1414                                         $text = self::convertAttachment($text, $simple_html, $try_oembed);
1415                                 }
1416
1417                                 $nosmile = strpos($text, '[nosmile]') !== false;
1418                                 $text = str_replace('[nosmile]', '', $text);
1419
1420                                 // Replace non graphical smilies for external posts
1421                                 if (!$nosmile) {
1422                                         $text = self::performWithEscapedTags($text, ['img'], function ($text) use ($simple_html, $for_plaintext) {
1423                                                 return Smilies::replace($text, ($simple_html != self::INTERNAL) || $for_plaintext);
1424                                         });
1425                                 }
1426
1427                                 // leave open the posibility of [map=something]
1428                                 // this is replaced in Item::prepareBody() which has knowledge of the item location
1429                                 if (strpos($text, '[/map]') !== false) {
1430                                         $text = preg_replace_callback(
1431                                                 "/\[map\](.*?)\[\/map\]/ism",
1432                                                 function ($match) use ($simple_html) {
1433                                                         return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
1434                                                 },
1435                                                 $text
1436                                         );
1437                                 }
1438
1439                                 if (strpos($text, '[map=') !== false) {
1440                                         $text = preg_replace_callback(
1441                                                 "/\[map=(.*?)\]/ism",
1442                                                 function ($match) use ($simple_html) {
1443                                                         return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
1444                                                 },
1445                                                 $text
1446                                         );
1447                                 }
1448
1449                                 if (strpos($text, '[map]') !== false) {
1450                                         $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
1451                                 }
1452
1453                                 // Check for headers
1454                                 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1455                                 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1456                                 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1457                                 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1458                                 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1459                                 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1460
1461                                 // Check for paragraph
1462                                 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1463
1464                                 // Check for bold text
1465                                 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1466
1467                                 // Check for Italics text
1468                                 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1469
1470                                 // Check for Underline text
1471                                 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1472
1473                                 // Check for strike-through text
1474                                 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
1475
1476                                 // Check for over-line text
1477                                 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1478
1479                                 // Check for colored text
1480                                 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1481
1482                                 // Check for sized text
1483                                 // [size=50] --> font-size: 50px (with the unit).
1484                                 if ($simple_html != self::DIASPORA) {
1485                                         $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", '<span style="font-size:$1px;line-height:normal;">$2</span>', $text);
1486                                         $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", '<span style="font-size:$1;line-height:normal;">$2</span>', $text);
1487                                 } else {
1488                                         // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
1489                                         $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text);
1490                                 }
1491
1492
1493                                 // Check for centered text
1494                                 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", '<div style="text-align:center;">$1</div>', $text);
1495
1496                                 // Check for list text
1497                                 $text = str_replace("[*]", "<li>", $text);
1498
1499                                 // Check for style sheet commands
1500                                 $text = preg_replace("(\[style=(.*?)\](.*?)\[\/style\])ism", '<span style="$1">$2</span>', $text);
1501
1502                                 // Check for CSS classes
1503                                 $text = preg_replace("(\[class=(.*?)\](.*?)\[\/class\])ism", '<span class="$1">$2</span>', $text);
1504
1505                                 // handle nested lists
1506                                 $endlessloop = 0;
1507
1508                                 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1509                                                 ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1510                                                 ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1511                                                 ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1512                                         $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1513                                         $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1514                                         $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1515                                         $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1516                                         $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1517                                         $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1518                                         $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1519                                         $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1520                                         $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1521                                         $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1522                                 }
1523
1524                                 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1525                                 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1526                                 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1527                                 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1528
1529                                 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1530                                 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1531
1532                                 $text = str_replace('[hr]', '<hr />', $text);
1533
1534                                 if (!$for_plaintext) {
1535                                         $text = self::performWithEscapedTags($text, ['url', 'img', 'audio', 'video', 'youtube', 'vimeo', 'share', 'attachment', 'iframe', 'bookmark'], function ($text) {
1536                                                 return preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
1537                                         });
1538                                 }
1539
1540                                 // Check for font change text
1541                                 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1542
1543                                 // Declare the format for [spoiler] layout
1544                                 $SpoilerLayout = '<details class="spoiler"><summary>' . DI::l10n()->t('Click to open/close') . '</summary>$1</details>';
1545
1546                                 // Check for [spoiler] text
1547                                 // handle nested quotes
1548                                 $endlessloop = 0;
1549                                 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1550                                         $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", $SpoilerLayout, $text);
1551                                 }
1552
1553                                 // Check for [spoiler=Title] text
1554
1555                                 // handle nested quotes
1556                                 $endlessloop = 0;
1557                                 while ((strpos($text, "[/spoiler]")!== false)  && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1558                                         $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1559                                                 '<details class="spoiler"><summary>$1</summary>$2</details>',
1560                                                 $text);
1561                                 }
1562
1563                                 // Declare the format for [quote] layout
1564                                 $QuoteLayout = '<blockquote>$1</blockquote>';
1565
1566                                 // Check for [quote] text
1567                                 // handle nested quotes
1568                                 $endlessloop = 0;
1569                                 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1570                                         $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1571                                 }
1572
1573                                 // Check for [quote=Author] text
1574
1575                                 $t_wrote = DI::l10n()->t('$1 wrote:');
1576
1577                                 // handle nested quotes
1578                                 $endlessloop = 0;
1579                                 while ((strpos($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>",
1582                                                 $text);
1583                                 }
1584
1585
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, $uriid) {
1590                                                 if (strpos($matches[3], "data:image/") === 0) {
1591                                                         return $matches[0];
1592                                                 }
1593
1594                                                 $matches[3] = self::proxyUrl($matches[3], $simple_html, $uriid);
1595                                                 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1596                                         },
1597                                         $text
1598                                 );
1599
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);
1602
1603                                 $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/ism",
1604                                         function ($matches) use ($simple_html, $uriid) {
1605                                                 $matches[1] = self::proxyUrl($matches[1], $simple_html, $uriid);
1606                                                 $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
1607                                                 return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '" title="' . $matches[2] . '">';
1608                                         },
1609                                         $text);
1610
1611                                 // Images
1612                                 // [img]pathtoimage[/img]
1613                                 $text = preg_replace_callback(
1614                                         "/\[img\](.*?)\[\/img\]/ism",
1615                                         function ($matches) use ($simple_html, $uriid) {
1616                                                 if (strpos($matches[1], "data:image/") === 0) {
1617                                                         return $matches[0];
1618                                                 }
1619
1620                                                 $matches[1] = self::proxyUrl($matches[1], $simple_html, $uriid);
1621                                                 return "[img]" . $matches[1] . "[/img]";
1622                                         },
1623                                         $text
1624                                 );
1625
1626                                 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1627                                 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
1628
1629                                 $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . DI::l10n()->t('Encrypted content') . '" title="' . DI::l10n()->t('Encrypted content') . '" /><br>', $text);
1630                                 $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . DI::l10n()->t('Encrypted content') . '" title="' . '$1' . ' ' . DI::l10n()->t('Encrypted content') . '" /><br>', $text);
1631                                 //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . DI::l10n()->t('Encrypted content') . '" title="' . '$1' . ' ' . DI::l10n()->t('Encrypted content') . '" /><br>', $Text);
1632
1633                                 // Simplify "video" element
1634                                 $text = preg_replace('(\[video[^\]]*?\ssrc\s?=\s?([^\s\]]+)[^\]]*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
1635
1636                                 if ($try_oembed) {
1637                                         // html5 video and audio
1638                                         $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism",
1639                                                 '<video src="$1" controls width="100%" height="auto"><a href="$1">$1</a></video>', $text);
1640
1641                                         $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1642                                         $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1643
1644                                         $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1645                                                 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1646                                         $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", '<audio src="$1" controls><a href="$1">$1</a></audio>', $text);
1647                                 } else {
1648                                         $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
1649                                                 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1650                                         $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
1651                                                 '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1652                                 }
1653
1654                                 // Backward compatibility, [iframe] support has been removed in version 2020.12
1655                                 $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1656
1657                                 // Youtube extensions
1658                                 if ($try_oembed) {
1659                                         $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1660                                         $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1661                                         $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1662                                 }
1663
1664                                 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1665                                 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1666                                 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1667
1668                                 if ($try_oembed) {
1669                                         $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);
1670                                 } else {
1671                                         $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1672                                                 '<a href="https://www.youtube.com/watch?v=$1" target="_blank" rel="noopener noreferrer">https://www.youtube.com/watch?v=$1</a>', $text);
1673                                 }
1674
1675                                 if ($try_oembed) {
1676                                         $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1677                                         $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1678                                 }
1679
1680                                 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1681                                 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1682
1683                                 if ($try_oembed) {
1684                                         $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);
1685                                 } else {
1686                                         $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1687                                                 '<a href="https://vimeo.com/$1" target="_blank" rel="noopener noreferrer">https://vimeo.com/$1</a>', $text);
1688                                 }
1689
1690                                 // oembed tag
1691                                 $text = OEmbed::BBCode2HTML($text);
1692
1693                                 // Avoid triple linefeeds through oembed
1694                                 $text = str_replace("<br style='clear:left'></span><br><br>", "<br style='clear:left'></span><br>", $text);
1695
1696                                 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1697                                 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1698                                 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1699                                 // start which is always required). Allow desc with a missing summary for compatibility.
1700
1701                                 if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
1702                                         $sub = Event::getHTML($ev, $simple_html);
1703
1704                                         $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1705                                         $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1706                                         $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1707                                         $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1708                                         $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1709                                         $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1710                                         $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1711                                 }
1712
1713                                 if (!$for_plaintext && DI::config()->get('system', 'big_emojis') && ($simple_html != self::DIASPORA)) {
1714                                         $conv = html_entity_decode(str_replace([' ', "\n", "\r"], '', $text));
1715                                         // Emojis are always 4 byte Unicode characters
1716                                         if (!empty($conv) && (strlen($conv) / mb_strlen($conv) == 4)) {
1717                                                 $text = '<span style="font-size: xx-large; line-height: normal;">' . $text . '</span>';
1718                                         }
1719                                 }
1720
1721                                 if (!$for_plaintext) {
1722                                         if (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1723                                                 $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1724                                                 $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
1725                                         }
1726                                 } else {
1727                                         $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
1728                                         $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1729                                 }
1730
1731                                 // Remove all hashtag addresses
1732                                 if ($simple_html && !in_array($simple_html, [self::DIASPORA, self::OSTATUS, self::ACTIVITYPUB])) {
1733                                         $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
1734                                 } elseif ($simple_html == self::DIASPORA) {
1735                                         // The ! is converted to @ since Diaspora only understands the @
1736                                         $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1737                                                 '@<a href="$2">$3</a>',
1738                                                 $text);
1739                                 } elseif (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
1740                                         $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1741                                                 '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1742                                                 $text);
1743                                 } elseif (!$simple_html) {
1744                                         $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
1745                                                 '<bdi>$1<a href="$2" class="userinfo mention" title="$3">$3</a></bdi>',
1746                                                 $text);
1747                                 }
1748
1749                                 // Bookmarks in red - will be converted to bookmarks in friendica
1750                                 $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1751                                 $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1752                                 $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
1753                                                         "[bookmark=$1]$2[/bookmark]", $text);
1754
1755                                 if (in_array($simple_html, [self::API, self::OSTATUS, self::TWITTER])) {
1756                                         $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1757                                         //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1758                                         $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1759                                 }
1760
1761                                 // Perform URL Search
1762                                 if ($try_oembed) {
1763                                         $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1764                                 }
1765
1766                                 $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1767
1768                                 // Handle Diaspora posts
1769                                 $text = preg_replace_callback(
1770                                         "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1771                                         function ($match) {
1772                                                 return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1773                                         }, $text
1774                                 );
1775
1776                                 $text = preg_replace_callback(
1777                                         "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
1778                                         function ($match) {
1779                                                 return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
1780                                         }, $text
1781                                 );
1782
1783                                 // Server independent link to posts and comments
1784                                 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1785                                 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1786                                 $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
1787
1788                                 /* Tag conversion
1789                                  * Supports:
1790                                  * - #[url=<anything>]<term>[/url]
1791                                  * - [url=<anything>]#<term>[/url]
1792                                  */
1793                                 $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) use ($simple_html) {
1794                                         if ($simple_html == BBCode::ACTIVITYPUB) {
1795                                                 return '<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1796                                                         . '" data-tag="' . XML::escape($matches[1]) . '" rel="tag ugc">#'
1797                                                         . XML::escape($matches[1]) . '</a>';
1798                                         } else {
1799                                                 return '#<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
1800                                                         . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
1801                                                         . XML::escape($matches[1]) . '</a>';
1802                                         }
1803                                 }, $text);
1804
1805                                 // We need no target="_blank" rel="noopener noreferrer" for local links
1806                                 // convert links start with DI::baseUrl() as local link without the target="_blank" rel="noopener noreferrer" attribute
1807                                 $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
1808                                 $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
1809                                 $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
1810
1811                                 $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
1812                                 $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1813
1814                                 // Red compatibility, though the link can't be authenticated on Friendica
1815                                 $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
1816
1817
1818                                 // we may need to restrict this further if it picks up too many strays
1819                                 // link acct:user@host to a webfinger profile redirector
1820
1821                                 $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', '<a href="' . DI::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
1822
1823                                 // Perform MAIL Search
1824                                 $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1825                                 $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1826
1827                                 /// @todo What is the meaning of these lines?
1828                                 $text = preg_replace('/\[\&amp\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1829                                 $text = preg_replace('/\&\#039\;/', '\'', $text);
1830
1831                                 // Currently deactivated, it made problems with " inside of alt texts.
1832                                 //$text = preg_replace('/\&quot\;/', '"', $text);
1833
1834                                 // fix any escaped ampersands that may have been converted into links
1835                                 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&amp\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1836
1837                                 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1838                                 $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:'];
1839
1840                                 array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
1841
1842                                 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1843                                                          '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $text);
1844
1845                                 // sanitize href attributes (only allowlisted protocols URLs)
1846                                 // default value for backward compatibility
1847                                 $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
1848
1849                                 // Always allowed protocol even if config isn't set or not including it
1850                                 $allowed_link_protocols[] = '//';
1851                                 $allowed_link_protocols[] = 'http://';
1852                                 $allowed_link_protocols[] = 'https://';
1853                                 $allowed_link_protocols[] = 'redir/';
1854
1855                                 array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
1856
1857                                 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1858                                 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
1859
1860                                 // Shared content
1861                                 $text = self::convertShare(
1862                                         $text,
1863                                         function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
1864                                                 return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
1865                                         }, $uriid
1866                                 );
1867
1868                                 $text = self::interpolateSavedImagesIntoItemBody($uriid, $text, $saved_image);
1869
1870                                 return $text;
1871                         }); // Escaped noparse, nobb, pre
1872
1873                         // Remove escaping tags and replace new lines that remain
1874                         $text = preg_replace_callback('/\[(noparse|nobb)](.*?)\[\/\1]/ism', function ($match) {
1875                                 return str_replace("\n", "<br>", $match[2]);
1876                         }, $text);
1877
1878                         // Additionally, [pre] tags preserve spaces
1879                         $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", function ($match) {
1880                                 return str_replace([' ', "\n"], ['&nbsp;', "<br>"], htmlentities($match[1], ENT_NOQUOTES,'UTF-8'));
1881                         }, $text);
1882
1883                         return $text;
1884                 }); // Escaped code
1885
1886                 $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
1887                         function ($matches) {
1888                                 if (strpos($matches[2], "\n") !== false) {
1889                                         $return = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlentities(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
1890                                 } else {
1891                                         $return = '<code>' . htmlentities($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
1892                                 }
1893
1894                                 return $return;
1895                         },
1896                         $text
1897                 );
1898
1899                 // Default iframe allowed domains/path
1900                 $allowedIframeDomains = [
1901                         DI::baseUrl()->getHostname()
1902                         . (DI::baseUrl()->getUrlPath() ? '/' . DI::baseUrl()->getUrlPath() : '')
1903                         . '/oembed/', # The path part has to change with the source in Content\Oembed::iframe
1904                         'www.youtube.com/embed/',
1905                         'player.vimeo.com/video/',
1906                 ];
1907
1908                 $allowedIframeDomains = array_merge(
1909                         $allowedIframeDomains,
1910                         DI::config()->get('system', 'allowed_oembed') ?
1911                                 explode(',', DI::config()->get('system', 'allowed_oembed'))
1912                                 : []
1913                 );
1914
1915                 $text = HTML::purify($text, $allowedIframeDomains);
1916
1917                 return trim($text);
1918         }
1919
1920         /**
1921          * Strips the "abstract" tag from the provided text
1922          *
1923          * @param string $text The text with BBCode
1924          * @return string The same text - but without "abstract" element
1925          */
1926         public static function stripAbstract($text)
1927         {
1928                 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1929                 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1930
1931                 return $text;
1932         }
1933
1934         /**
1935          * Returns the value of the "abstract" element
1936          *
1937          * @param string $text The text that maybe contains the element
1938          * @param string $addon The addon for which the abstract is meant for
1939          * @return string The abstract
1940          */
1941         public static function getAbstract($text, $addon = '')
1942         {
1943                 $abstract = '';
1944                 $abstracts = [];
1945                 $addon = strtolower($addon);
1946
1947                 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1948                         foreach ($results AS $result) {
1949                                 $abstracts[strtolower($result[1])] = $result[2];
1950                         }
1951                 }
1952
1953                 if (isset($abstracts[$addon])) {
1954                         $abstract = $abstracts[$addon];
1955                 }
1956
1957                 if ($abstract == '' && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1958                         $abstract = $result[1];
1959                 }
1960
1961                 return $abstract;
1962         }
1963
1964         /**
1965          * Callback function to replace a Friendica style mention in a mention for Diaspora
1966          *
1967          * @param array $match Matching values for the callback
1968          *                     [1] = Mention type (! or @)
1969          *                     [2] = Name
1970          *                     [3] = Address
1971          * @return string Replaced mention
1972          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1973          * @throws \ImagickException
1974          */
1975         private static function bbCodeMention2DiasporaCallback($match)
1976         {
1977                 $contact = Contact::getByURL($match[3], false, ['addr']);
1978                 if (empty($contact['addr'])) {
1979                         return $match[0];
1980                 }
1981
1982                 $mention = $match[1] . '{' . $match[2] . '; ' . $contact['addr'] . '}';
1983                 return $mention;
1984         }
1985
1986         /**
1987          * Converts a BBCode text into Markdown
1988          *
1989          * This function converts a BBCode item body to be sent to Markdown-enabled
1990          * systems like Diaspora and Libertree
1991          *
1992          * @param string $text
1993          * @param bool   $for_diaspora Diaspora requires more changes than Libertree
1994          * @return string
1995          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1996          */
1997         public static function toMarkdown($text, $for_diaspora = true)
1998         {
1999                 $original_text = $text;
2000
2001                 // Since Diaspora is creating a summary for links, this function removes them before posting
2002                 if ($for_diaspora) {
2003                         $text = self::removeShareInformation($text);
2004                 }
2005
2006                 /**
2007                  * Transform #tags, strip off the [url] and replace spaces with underscore
2008                  */
2009                 $url_search_string = "^\[\]";
2010                 $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
2011                         function ($matches) {
2012                                 return '#' . str_replace(' ', '_', $matches[2]);
2013                         },
2014                         $text
2015                 );
2016
2017                 // Converting images with size parameters to simple images. Markdown doesn't know it.
2018                 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
2019
2020                 // Convert it to HTML - don't try oembed
2021                 if ($for_diaspora) {
2022                         $text = self::convert($text, false, self::DIASPORA);
2023
2024                         // Add all tags that maybe were removed
2025                         if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
2026                                 $tagline = '';
2027                                 foreach ($tags[2] as $tag) {
2028                                         $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
2029                                         if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
2030                                                 $tagline .= '#' . $tag . ' ';
2031                                         }
2032                                 }
2033                                 $text = $text . " " . $tagline;
2034                         }
2035                 } else {
2036                         $text = self::convert($text, false, self::CONNECTORS);
2037                 }
2038
2039                 // If a link is followed by a quote then there should be a newline before it
2040                 // Maybe we should make this newline at every time before a quote.
2041                 $text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
2042
2043                 $stamp1 = microtime(true);
2044
2045                 // Now convert HTML to Markdown
2046                 $text = HTML::toMarkdown($text);
2047
2048                 DI::profiler()->saveTimestamp($stamp1, "parser");
2049
2050                 // Libertree has a problem with escaped hashtags.
2051                 $text = str_replace(['\#'], ['#'], $text);
2052
2053                 // Remove any leading or trailing whitespace, as this will mess up
2054                 // the Diaspora signature verification and cause the item to disappear
2055                 $text = trim($text);
2056
2057                 if ($for_diaspora) {
2058                         $url_search_string = "^\[\]";
2059                         $text = preg_replace_callback(
2060                                 "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism",
2061                                 ['self', 'bbCodeMention2DiasporaCallback'],
2062                                 $text
2063                         );
2064                 }
2065
2066                 Hook::callAll('bb2diaspora', $text);
2067
2068                 return $text;
2069         }
2070
2071         /**
2072          * Pull out all #hashtags and @person tags from $string.
2073          *
2074          * We also get @person@domain.com - which would make
2075          * the regex quite complicated as tags can also
2076          * end a sentence. So we'll run through our results
2077          * and strip the period from any tags which end with one.
2078          * Returns array of tags found, or empty array.
2079          *
2080          * @param string $string Post content
2081          *
2082          * @return array List of tag and person names
2083          */
2084         public static function getTags($string)
2085         {
2086                 $ret = [];
2087
2088                 BBCode::performWithEscapedTags($string, ['noparse', 'pre', 'code', 'img'], function ($string) use (&$ret) {
2089                         // Convert hashtag links to hashtags
2090                         $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2 ', $string);
2091
2092                         // Force line feeds at bbtags
2093                         $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
2094
2095                         // ignore anything in a bbtag
2096                         $string = preg_replace('/\[(.*?)\]/sm', '', $string);
2097
2098                         // Match full names against @tags including the space between first and last
2099                         // We will look these up afterward to see if they are full names or not recognisable.
2100
2101                         if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
2102                                 foreach ($matches[1] as $match) {
2103                                         if (strstr($match, ']')) {
2104                                                 // we might be inside a bbcode color tag - leave it alone
2105                                                 continue;
2106                                         }
2107
2108                                         if (substr($match, -1, 1) === '.') {
2109                                                 $ret[] = substr($match, 0, -1);
2110                                         } else {
2111                                                 $ret[] = $match;
2112                                         }
2113                                 }
2114                         }
2115
2116                         // Otherwise pull out single word tags. These can be @nickname, @first_last
2117                         // and #hash tags.
2118
2119                         if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?\']*[^\^ \x0D\x0A,;:?!\'.])/', $string, $matches)) {
2120                                 foreach ($matches[1] as $match) {
2121                                         if (strstr($match, ']')) {
2122                                                 // we might be inside a bbcode color tag - leave it alone
2123                                                 continue;
2124                                         }
2125
2126                                         // try not to catch url fragments
2127                                         if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
2128                                                 continue;
2129                                         }
2130
2131                                         $ret[] = $match;
2132                                 }
2133                         }
2134                 });
2135
2136                 return array_unique($ret);
2137         }
2138
2139         /**
2140          * Expand tags to URLs
2141          *
2142          * @param string $body
2143          * @return string body with expanded tags
2144          */
2145         public static function expandTags(string $body)
2146         {
2147                 return preg_replace_callback("/([!#@])([^\^ \x0D\x0A,;:?\']*[^\^ \x0D\x0A,;:?!\'.])/",
2148                         function ($match) {
2149                                 switch ($match[1]) {
2150                                         case '!':
2151                                         case '@':
2152                                                 $contact = Contact::getByURL($match[2]);
2153                                                 if (!empty($contact)) {
2154                                                         return $match[1] . '[url=' . $contact['url'] . ']' . $contact['name'] . '[/url]';
2155                                                 } else {
2156                                                         return $match[1] . $match[2];
2157                                                 }
2158                                                 break;
2159                                         case '#':
2160                                                 return $match[1] . '[url=' . 'https://' . DI::baseUrl() . '/search?tag=' . $match[2] . ']' . $match[2] . '[/url]';
2161                                 }
2162                         }, $body);
2163         }
2164
2165         /**
2166          * Perform a custom function on a text after having escaped blocks enclosed in the provided tag list.
2167          *
2168          * @param string   $text
2169          * @param array    $tagList A list of tag names, e.g ['noparse', 'nobb', 'pre']
2170          * @param callable $callback
2171          * @return string
2172          * @throws Exception
2173          *@see Strings::performWithEscapedBlocks
2174          *
2175          */
2176         public static function performWithEscapedTags(string $text, array $tagList, callable $callback)
2177         {
2178                 $tagList = array_map('preg_quote', $tagList);
2179
2180                 return Strings::performWithEscapedBlocks($text, '#\[(?:' . implode('|', $tagList) . ').*?\[/(?:' . implode('|', $tagList) . ')]#ism', $callback);
2181         }
2182
2183         /**
2184          * Replaces mentions in the provided message body for the provided user and network if any
2185          *
2186          * @param $body
2187          * @param $profile_uid
2188          * @param $network
2189          * @return string
2190          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2191          * @throws \ImagickException
2192          */
2193         public static function setMentions($body, $profile_uid = 0, $network = '')
2194         {
2195                 BBCode::performWithEscapedTags($body, ['noparse', 'pre', 'code', 'img'], function ($body) use ($profile_uid, $network) {
2196                         $tags = BBCode::getTags($body);
2197
2198                         $tagged = [];
2199                         $inform = '';
2200
2201                         foreach ($tags as $tag) {
2202                                 $tag_type = substr($tag, 0, 1);
2203
2204                                 if ($tag_type == Tag::TAG_CHARACTER[Tag::HASHTAG]) {
2205                                         continue;
2206                                 }
2207
2208                                 /*
2209                                  * If we already tagged 'Robert Johnson', don't try and tag 'Robert'.
2210                                  * Robert Johnson should be first in the $tags array
2211                                  */
2212                                 foreach ($tagged as $nextTag) {
2213                                         if (stristr($nextTag, $tag . ' ')) {
2214                                                 continue 2;
2215                                         }
2216                                 }
2217
2218                                 if (($success = Item::replaceTag($body, $inform, $profile_uid, $tag, $network)) && $success['replaced']) {
2219                                         $tagged[] = $tag;
2220                                 }
2221                         }
2222
2223                         return $body;
2224                 });
2225
2226                 return $body;
2227         }
2228
2229         /**
2230          * @param string      $author  Author display name
2231          * @param string      $profile Author profile URL
2232          * @param string      $avatar  Author profile picture URL
2233          * @param string      $link    Post source URL
2234          * @param string      $posted  Post created date
2235          * @param string|null $guid    Post guid (if any)
2236          * @return string
2237          * @TODO Rewrite to handle over whole record array
2238          */
2239         public static function getShareOpeningTag(string $author, string $profile, string $avatar, string $link, string $posted, string $guid = null)
2240         {
2241                 $header = "[share author='" . str_replace(["'", "[", "]"], ["&#x27;", "&#x5B;", "&#x5D;"], $author) .
2242                         "' profile='" . str_replace(["'", "[", "]"], ["&#x27;", "&#x5B;", "&#x5D;"], $profile) .
2243                         "' avatar='" . str_replace(["'", "[", "]"], ["&#x27;", "&#x5B;", "&#x5D;"], $avatar) .
2244                         "' link='" . str_replace(["'", "[", "]"], ["&#x27;", "&#x5B;", "&#x5D;"], $link) .
2245                         "' posted='" . str_replace(["'", "[", "]"], ["&#x27;", "&#x5B;", "&#x5D;"], $posted);
2246
2247                 if ($guid) {
2248                         $header .= "' guid='" . str_replace(["'", "[", "]"], ["&#x27;", "&#x5B;", "&#x5D;"], $guid);
2249                 }
2250
2251                 $header  .= "']";
2252
2253                 return $header;
2254         }
2255
2256         /**
2257          * Returns the BBCode relevant to embed the provided URL in a post body.
2258          * For media type, it will return [img], [video] and [audio] tags.
2259          * For regular web pages, it will either output a [bookmark] tag if title and description were provided,
2260          * an [attachment] tag or a simple [url] tag depending on $tryAttachment.
2261          *
2262          * @param string      $url
2263          * @param bool        $tryAttachment
2264          * @param string|null $title
2265          * @param string|null $description
2266          * @param string|null $tags
2267          * @return string
2268          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2269          *@see ParseUrl::getSiteinfoCached
2270          *
2271          */
2272         public static function embedURL(string $url, bool $tryAttachment = true, string $title = null, string $description = null, string $tags = null): string
2273         {
2274                 DI::logger()->info($url);
2275
2276                 // If there is already some content information submitted we don't
2277                 // need to parse the url for content.
2278                 if (!empty($title) && !empty($description)) {
2279                         $title = str_replace(["\r", "\n"], ['', ''], $title);
2280
2281                         $description = '[quote]' . trim($description) . '[/quote]' . "\n";
2282
2283                         $str_tags = '';
2284                         if (!empty($tags)) {
2285                                 $arr_tags = ParseUrl::convertTagsToArray($tags);
2286                                 if (count($arr_tags)) {
2287                                         $str_tags = "\n" . implode(' ', $arr_tags) . "\n";
2288                                 }
2289                         }
2290
2291                         $result = sprintf('[bookmark=%s]%s[/bookmark]%s', $url, ($title) ? $title : $url, $description) . $str_tags;
2292
2293                         DI::logger()->info('(unparsed): returns: ' . $result);
2294
2295                         return $result;
2296                 }
2297
2298                 $siteinfo = ParseUrl::getSiteinfoCached($url);
2299
2300                 if (in_array($siteinfo['type'], ['image', 'video', 'audio'])) {
2301                         switch ($siteinfo['type']) {
2302                                 case 'video':
2303                                         $bbcode = "\n" . '[video]' . $url . '[/video]' . "\n";
2304                                         break;
2305                                 case 'audio':
2306                                         $bbcode = "\n" . '[audio]' . $url . '[/audio]' . "\n";
2307                                         break;
2308                                 default:
2309                                         $bbcode = "\n" . '[img]' . $url . '[/img]' . "\n";
2310                                         break;
2311                         }
2312
2313                         return $bbcode;
2314                 }
2315
2316                 unset($siteinfo['keywords']);
2317
2318                 // Bypass attachment if parse url for a comment
2319                 if (!$tryAttachment) {
2320                         return "\n" . '[url=' . $url . ']' . $siteinfo['title'] . '[/url]';
2321                 }
2322
2323                 // Format it as BBCode attachment
2324                 return "\n" . PageInfo::getFooterFromData($siteinfo);
2325         }
2326 }