src/Content/Text/BBCode.php

   1 <?php
   2 /**
   3  * @file src/Content/Text/BBCode.php
   4  */
   5 namespace Friendica\Content\Text;
   6
   7 use DOMDocument;
   8 use Exception;
   9 use Friendica\Content\OEmbed;
  10 use Friendica\Content\Smilies;
  11 use Friendica\Content\Text\Plaintext;
  12 use Friendica\Core\Addon;
  13 use Friendica\Core\Cache;
  14 use Friendica\Core\Config;
  15 use Friendica\Core\L10n;
  16 use Friendica\Core\PConfig;
  17 use Friendica\Core\System;
  18 use Friendica\Model\Contact;
  19 use Friendica\Object\Image;
  20 use Friendica\Util\Map;
  21 use Friendica\Util\Network;
  22 use Friendica\Util\ParseUrl;
  23
  24 require_once "include/bbcode.php";
  25 require_once "include/html2plain.php";
  26
  27 class BBCode
  28 {
  29         /**
  30          * @brief Fetches attachment data that were generated the old way
  31          *
  32          * @param string $body Message body
  33          * @return array
  34          * 'type' -> Message type ("link", "video", "photo")
  35          * 'text' -> Text before the shared message
  36          * 'after' -> Text after the shared message
  37          * 'image' -> Preview image of the message
  38          * 'url' -> Url to the attached message
  39          * 'title' -> Title of the attachment
  40          * 'description' -> Description of the attachment
  41          */
  42         private static function getOldAttachmentData($body)
  43         {
  44                 $post = [];
  45
  46                 // Simplify image codes
  47                 $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
  48
  49                 if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
  50                         foreach ($attached as $data) {
  51                                 if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
  52                                         continue;
  53                                 }
  54
  55                                 $post["type"] = substr($data[1], 5);
  56
  57                                 $pos = strpos($body, $data[0]);
  58                                 if ($pos > 0) {
  59                                         $post["text"] = trim(substr($body, 0, $pos));
  60                                         $post["after"] = trim(substr($body, $pos + strlen($data[0])));
  61                                 } else {
  62                                         $post["text"] = trim(str_replace($data[0], "", $body));
  63                                 }
  64
  65                                 $attacheddata = $data[2];
  66
  67                                 $URLSearchString = "^\[\]";
  68
  69                                 if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
  70
  71                                         $picturedata = Image::getInfoFromURL($matches[1]);
  72
  73                                         if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
  74                                                 $post["image"] = $matches[1];
  75                                         } else {
  76                                                 $post["preview"] = $matches[1];
  77                                         }
  78                                 }
  79
  80                                 if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
  81                                         $post["url"] = $matches[1];
  82                                         $post["title"] = $matches[2];
  83                                 }
  84                                 if (($post["url"] == "") && (in_array($post["type"], ["link", "video"]))
  85                                         && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
  86                                         $post["url"] = $matches[1];
  87                                 }
  88
  89                                 // Search for description
  90                                 if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
  91                                         $post["description"] = $matches[1];
  92                                 }
  93                         }
  94                 }
  95                 return $post;
  96         }
  97
  98         /**
  99          * @brief Fetches attachment data that were generated with the "attachment" element
 100          *
 101          * @param string $body Message body
 102          * @return array
 103          * 'type' -> Message type ("link", "video", "photo")
 104          * 'text' -> Text before the shared message
 105          * 'after' -> Text after the shared message
 106          * 'image' -> Preview image of the message
 107          * 'url' -> Url to the attached message
 108          * 'title' -> Title of the attachment
 109          * 'description' -> Description of the attachment
 110          */
 111         public static function getAttachmentData($body)
 112         {
 113                 $data = [];
 114
 115                 if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
 116                         return self::getOldAttachmentData($body);
 117                 }
 118
 119                 $attributes = $match[2];
 120
 121                 $data["text"] = trim($match[1]);
 122
 123                 $type = "";
 124                 preg_match("/type='(.*?)'/ism", $attributes, $matches);
 125                 if (x($matches, 1)) {
 126                         $type = strtolower($matches[1]);
 127                 }
 128
 129                 preg_match('/type="(.*?)"/ism', $attributes, $matches);
 130                 if (x($matches, 1)) {
 131                         $type = strtolower($matches[1]);
 132                 }
 133
 134                 if ($type == "") {
 135                         return [];
 136                 }
 137
 138                 if (!in_array($type, ["link", "audio", "photo", "video"])) {
 139                         return [];
 140                 }
 141
 142                 if ($type != "") {
 143                         $data["type"] = $type;
 144                 }
 145
 146                 $url = "";
 147                 preg_match("/url='(.*?)'/ism", $attributes, $matches);
 148                 if (x($matches, 1)) {
 149                         $url = $matches[1];
 150                 }
 151
 152                 preg_match('/url="(.*?)"/ism', $attributes, $matches);
 153                 if (x($matches, 1)) {
 154                         $url = $matches[1];
 155                 }
 156
 157                 if ($url != "") {
 158                         $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
 159                 }
 160
 161                 $title = "";
 162                 preg_match("/title='(.*?)'/ism", $attributes, $matches);
 163                 if (x($matches, 1)) {
 164                         $title = $matches[1];
 165                 }
 166
 167                 preg_match('/title="(.*?)"/ism', $attributes, $matches);
 168                 if (x($matches, 1)) {
 169                         $title = $matches[1];
 170                 }
 171
 172                 if ($title != "") {
 173                         $title = bbcode(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, false, true);
 174                         $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
 175                         $title = str_replace(["[", "]"], ["&#91;", "&#93;"], $title);
 176                         $data["title"] = $title;
 177                 }
 178
 179                 $image = "";
 180                 preg_match("/image='(.*?)'/ism", $attributes, $matches);
 181                 if (x($matches, 1)) {
 182                         $image = $matches[1];
 183                 }
 184
 185                 preg_match('/image="(.*?)"/ism', $attributes, $matches);
 186                 if (x($matches, 1)) {
 187                         $image = $matches[1];
 188                 }
 189
 190                 if ($image != "") {
 191                         $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
 192                 }
 193
 194                 $preview = "";
 195                 preg_match("/preview='(.*?)'/ism", $attributes, $matches);
 196                 if (x($matches, 1)) {
 197                         $preview = $matches[1];
 198                 }
 199
 200                 preg_match('/preview="(.*?)"/ism', $attributes, $matches);
 201                 if (x($matches, 1)) {
 202                         $preview = $matches[1];
 203                 }
 204
 205                 if ($preview != "") {
 206                         $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
 207                 }
 208
 209                 $data["description"] = trim($match[3]);
 210
 211                 $data["after"] = trim($match[4]);
 212
 213                 return $data;
 214         }
 215
 216         public static function getAttachedData($body, $item = [])
 217         {
 218                 /*
 219                 - text:
 220                 - type: link, video, photo
 221                 - title:
 222                 - url:
 223                 - image:
 224                 - description:
 225                 - (thumbnail)
 226                 */
 227
 228                 $has_title = !empty($item['title']);
 229                 $plink = (!empty($item['plink']) ? $item['plink'] : '');
 230                 $post = self::getAttachmentData($body);
 231
 232                 // if nothing is found, it maybe having an image.
 233                 if (!isset($post["type"])) {
 234                         // Simplify image codes
 235                         $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
 236
 237                         $URLSearchString = "^\[\]";
 238                         if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
 239                                 if ((count($pictures) == 1) && !$has_title) {
 240                                         // Checking, if the link goes to a picture
 241                                         $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
 242
 243                                         // Workaround:
 244                                         // Sometimes photo posts to the own album are not detected at the start.
 245                                         // So we seem to cannot use the cache for these cases. That's strange.
 246                                         if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
 247                                                 $data = ParseUrl::getSiteinfo($pictures[0][1], true);
 248                                         }
 249
 250                                         if ($data["type"] == "photo") {
 251                                                 $post["type"] = "photo";
 252                                                 if (isset($data["images"][0])) {
 253                                                         $post["image"] = $data["images"][0]["src"];
 254                                                         $post["url"] = $data["url"];
 255                                                 } else {
 256                                                         $post["image"] = $data["url"];
 257                                                 }
 258
 259                                                 $post["preview"] = $pictures[0][2];
 260                                                 $post["text"] = str_replace($pictures[0][0], "", $body);
 261                                         } else {
 262                                                 $imgdata = Image::getInfoFromURL($pictures[0][1]);
 263                                                 if (substr($imgdata["mime"], 0, 6) == "image/") {
 264                                                         $post["type"] = "photo";
 265                                                         $post["image"] = $pictures[0][1];
 266                                                         $post["preview"] = $pictures[0][2];
 267                                                         $post["text"] = str_replace($pictures[0][0], "", $body);
 268                                                 }
 269                                         }
 270                                 } elseif (count($pictures) > 0) {
 271                                         $post["type"] = "link";
 272                                         $post["url"] = $plink;
 273                                         $post["image"] = $pictures[0][2];
 274                                         $post["text"] = $body;
 275                                 }
 276                         } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
 277                                 if ((count($pictures) == 1) && !$has_title) {
 278                                         $post["type"] = "photo";
 279                                         $post["image"] = $pictures[0][1];
 280                                         $post["text"] = str_replace($pictures[0][0], "", $body);
 281                                 } elseif (count($pictures) > 0) {
 282                                         $post["type"] = "link";
 283                                         $post["url"] = $plink;
 284                                         $post["image"] = $pictures[0][1];
 285                                         $post["text"] = $body;
 286                                 }
 287                         }
 288
 289                         // Test for the external links
 290                         preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
 291                         preg_match_all("(\[url\=([$URLSearchString]*)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
 292
 293                         $links = array_merge($links1, $links2);
 294
 295                         // If there is only a single one, then use it.
 296                         // This should cover link posts via API.
 297                         if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
 298                                 $post["type"] = "link";
 299                                 $post["text"] = trim($body);
 300                                 $post["url"] = $links[0][1];
 301                         }
 302
 303                         // Now count the number of external media links
 304                         preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
 305                         preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
 306                         preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
 307                         preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
 308
 309                         // Add them to the other external links
 310                         $links = array_merge($links, $links1, $links2, $links3, $links4);
 311
 312                         // Are there more than one?
 313                         if (count($links) > 1) {
 314                                 // The post will be the type "text", which means a blog post
 315                                 unset($post["type"]);
 316                                 $post["url"] = $plink;
 317                         }
 318
 319                         if (!isset($post["type"])) {
 320                                 $post["type"] = "text";
 321                                 $post["text"] = trim($body);
 322                         }
 323                 } elseif (isset($post["url"]) && ($post["type"] == "video")) {
 324                         $data = ParseUrl::getSiteinfoCached($post["url"], true);
 325
 326                         if (isset($data["images"][0])) {
 327                                 $post["image"] = $data["images"][0]["src"];
 328                         }
 329                 }
 330
 331                 return $post;
 332         }
 333
 334         /**
 335          * @brief Convert a message into plaintext for connectors to other networks
 336          *
 337          * @param array $b The message array that is about to be posted
 338          * @param int $limit The maximum number of characters when posting to that network
 339          * @param bool $includedlinks Has an attached link to be included into the message?
 340          * @param int $htmlmode This triggers the behaviour of the bbcode conversion
 341          * @param string $target_network Name of the network where the post should go to.
 342          *
 343          * @return string The converted message
 344          */
 345         public static function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "")
 346         {
 347                 // Remove the hash tags
 348                 $URLSearchString = "^\[\]";
 349                 $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]);
 350
 351                 // Add an URL element if the text contains a raw link
 352                 $body = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2[/url]', $body);
 353
 354                 // Remove the abstract
 355                 $body = remove_abstract($body);
 356
 357                 // At first look at data that is attached via "type-..." stuff
 358                 // This will hopefully replaced with a dedicated bbcode later
 359                 //$post = self::getAttachedData($b["body"]);
 360                 $post = self::getAttachedData($body, $b);
 361
 362                 if (($b["title"] != "") && ($post["text"] != "")) {
 363                         $post["text"] = trim($b["title"]."\n\n".$post["text"]);
 364                 } elseif ($b["title"] != "") {
 365                         $post["text"] = trim($b["title"]);
 366                 }
 367
 368                 $abstract = "";
 369
 370                 // Fetch the abstract from the given target network
 371                 if ($target_network != "") {
 372                         $default_abstract = fetch_abstract($b["body"]);
 373                         $abstract = fetch_abstract($b["body"], $target_network);
 374
 375                         // If we post to a network with no limit we only fetch
 376                         // an abstract exactly for this network
 377                         if (($limit == 0) && ($abstract == $default_abstract)) {
 378                                 $abstract = "";
 379                         }
 380                 } else {// Try to guess the correct target network
 381                         switch ($htmlmode) {
 382                                 case 8:
 383                                         $abstract = fetch_abstract($b["body"], NETWORK_TWITTER);
 384                                         break;
 385                                 case 7:
 386                                         $abstract = fetch_abstract($b["body"], NETWORK_STATUSNET);
 387                                         break;
 388                                 case 6:
 389                                         $abstract = fetch_abstract($b["body"], NETWORK_APPNET);
 390                                         break;
 391                                 default: // We don't know the exact target.
 392                                         // We fetch an abstract since there is a posting limit.
 393                                         if ($limit > 0) {
 394                                                 $abstract = fetch_abstract($b["body"]);
 395                                         }
 396                         }
 397                 }
 398
 399                 if ($abstract != "") {
 400                         $post["text"] = $abstract;
 401
 402                         if ($post["type"] == "text") {
 403                                 $post["type"] = "link";
 404                                 $post["url"] = $b["plink"];
 405                         }
 406                 }
 407
 408                 $html = bbcode($post["text"].$post["after"], false, false, $htmlmode);
 409                 $msg = html2plain($html, 0, true);
 410                 $msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8'));
 411
 412                 $link = "";
 413                 if ($includedlinks) {
 414                         if ($post["type"] == "link") {
 415                                 $link = $post["url"];
 416                         } elseif ($post["type"] == "text") {
 417                                 $link = $post["url"];
 418                         } elseif ($post["type"] == "video") {
 419                                 $link = $post["url"];
 420                         } elseif ($post["type"] == "photo") {
 421                                 $link = $post["image"];
 422                         }
 423
 424                         if (($msg == "") && isset($post["title"])) {
 425                                 $msg = trim($post["title"]);
 426                         }
 427
 428                         if (($msg == "") && isset($post["description"])) {
 429                                 $msg = trim($post["description"]);
 430                         }
 431
 432                         // If the link is already contained in the post, then it neeedn't to be added again
 433                         // But: if the link is beyond the limit, then it has to be added.
 434                         if (($link != "") && strstr($msg, $link)) {
 435                                 $pos = strpos($msg, $link);
 436
 437                                 // Will the text be shortened in the link?
 438                                 // Or is the link the last item in the post?
 439                                 if (($limit > 0) && ($pos < $limit) && (($pos + 23 > $limit) || ($pos + strlen($link) == strlen($msg)))) {
 440                                         $msg = trim(str_replace($link, "", $msg));
 441                                 } elseif (($limit == 0) || ($pos < $limit)) {
 442                                         // The limit has to be increased since it will be shortened - but not now
 443                                         // Only do it with Twitter (htmlmode = 8)
 444                                         if (($limit > 0) && (strlen($link) > 23) && ($htmlmode == 8)) {
 445                                                 $limit = $limit - 23 + strlen($link);
 446                                         }
 447
 448                                         $link = "";
 449
 450                                         if ($post["type"] == "text") {
 451                                                 unset($post["url"]);
 452                                         }
 453                                 }
 454                         }
 455                 }
 456
 457                 if ($limit > 0) {
 458                         // Reduce multiple spaces
 459                         // When posted to a network with limited space, we try to gain space where possible
 460                         while (strpos($msg, "  ") !== false) {
 461                                 $msg = str_replace("  ", " ", $msg);
 462                         }
 463
 464                         // Twitter is using its own limiter, so we always assume that shortened links will have this length
 465                         if (iconv_strlen($link, "UTF-8") > 0) {
 466                                 $limit = $limit - 23;
 467                         }
 468
 469                         if (iconv_strlen($msg, "UTF-8") > $limit) {
 470                                 if (($post["type"] == "text") && isset($post["url"])) {
 471                                         $post["url"] = $b["plink"];
 472                                 } elseif (!isset($post["url"])) {
 473                                         $limit = $limit - 23;
 474                                         $post["url"] = $b["plink"];
 475                                 // Which purpose has this line? It is now uncommented, but left as a reminder
 476                                 //} elseif (strpos($b["body"], "[share") !== false) {
 477                                 //      $post["url"] = $b["plink"];
 478                                 } elseif (PConfig::get($b["uid"], "system", "no_intelligent_shortening")) {
 479                                         $post["url"] = $b["plink"];
 480                                 }
 481                                 $msg = Plaintext::shorten($msg, $limit);
 482                         }
 483                 }
 484
 485                 $post["text"] = trim($msg);
 486
 487                 return($post);
 488         }
 489
 490         public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
 491         {
 492                 // Suppress "view full size"
 493                 if (intval(Config::get('system', 'no_view_full_size'))) {
 494                         $include_link = false;
 495                 }
 496
 497                 // Picture addresses can contain special characters
 498                 $s = htmlspecialchars_decode($srctext);
 499
 500                 $matches = null;
 501                 $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
 502                 if ($c) {
 503                         foreach ($matches as $mtch) {
 504                                 logger('scale_external_image: ' . $mtch[1]);
 505
 506                                 $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
 507                                 if (stristr($mtch[1], $hostname)) {
 508                                         continue;
 509                                 }
 510
 511                                 // $scale_replace, if passed, is an array of two elements. The
 512                                 // first is the name of the full-size image. The second is the
 513                                 // name of a remote, scaled-down version of the full size image.
 514                                 // This allows Friendica to display the smaller remote image if
 515                                 // one exists, while still linking to the full-size image
 516                                 if ($scale_replace) {
 517                                         $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
 518                                 } else {
 519                                         $scaled = $mtch[1];
 520                                 }
 521                                 $i = Network::fetchUrl($scaled);
 522                                 if (!$i) {
 523                                         return $srctext;
 524                                 }
 525
 526                                 // guess mimetype from headers or filename
 527                                 $type = Image::guessType($mtch[1], true);
 528
 529                                 if ($i) {
 530                                         $Image = new Image($i, $type);
 531                                         if ($Image->isValid()) {
 532                                                 $orig_width = $Image->getWidth();
 533                                                 $orig_height = $Image->getHeight();
 534
 535                                                 if ($orig_width > 640 || $orig_height > 640) {
 536                                                         $Image->scaleDown(640);
 537                                                         $new_width = $Image->getWidth();
 538                                                         $new_height = $Image->getHeight();
 539                                                         logger('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], LOGGER_DEBUG);
 540                                                         $s = str_replace(
 541                                                                 $mtch[0],
 542                                                                 '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
 543                                                                 . "\n" . (($include_link)
 544                                                                         ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
 545                                                                         : ''),
 546                                                                 $s
 547                                                         );
 548                                                         logger('scale_external_images: new string: ' . $s, LOGGER_DEBUG);
 549                                                 }
 550                                         }
 551                                 }
 552                         }
 553                 }
 554
 555                 // replace the special char encoding
 556                 $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
 557                 return $s;
 558         }
 559
 560         /**
 561          * The purpose of this function is to apply system message length limits to
 562          * imported messages without including any embedded photos in the length
 563          *
 564          * @brief Truncates imported message body string length to max_import_size
 565          * @param string $body
 566          * @return string
 567          */
 568         public static function limitBodySize($body)
 569         {
 570                 $maxlen = get_max_import_size();
 571
 572                 // If the length of the body, including the embedded images, is smaller
 573                 // than the maximum, then don't waste time looking for the images
 574                 if ($maxlen && (strlen($body) > $maxlen)) {
 575
 576                         logger('the total body length exceeds the limit', LOGGER_DEBUG);
 577
 578                         $orig_body = $body;
 579                         $new_body = '';
 580                         $textlen = 0;
 581
 582                         $img_start = strpos($orig_body, '[img');
 583                         $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
 584                         $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
 585                         while (($img_st_close !== false) && ($img_end !== false)) {
 586
 587                                 $img_st_close++; // make it point to AFTER the closing bracket
 588                                 $img_end += $img_start;
 589                                 $img_end += strlen('[/img]');
 590
 591                                 if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
 592                                         // This is an embedded image
 593
 594                                         if (($textlen + $img_start) > $maxlen) {
 595                                                 if ($textlen < $maxlen) {
 596                                                         logger('the limit happens before an embedded image', LOGGER_DEBUG);
 597                                                         $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
 598                                                         $textlen = $maxlen;
 599                                                 }
 600                                         } else {
 601                                                 $new_body = $new_body . substr($orig_body, 0, $img_start);
 602                                                 $textlen += $img_start;
 603                                         }
 604
 605                                         $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
 606                                 } else {
 607
 608                                         if (($textlen + $img_end) > $maxlen) {
 609                                                 if ($textlen < $maxlen) {
 610                                                         logger('the limit happens before the end of a non-embedded image', LOGGER_DEBUG);
 611                                                         $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
 612                                                         $textlen = $maxlen;
 613                                                 }
 614                                         } else {
 615                                                 $new_body = $new_body . substr($orig_body, 0, $img_end);
 616                                                 $textlen += $img_end;
 617                                         }
 618                                 }
 619                                 $orig_body = substr($orig_body, $img_end);
 620
 621                                 if ($orig_body === false) {
 622                                         // in case the body ends on a closing image tag
 623                                         $orig_body = '';
 624                                 }
 625
 626                                 $img_start = strpos($orig_body, '[img');
 627                                 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
 628                                 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
 629                         }
 630
 631                         if (($textlen + strlen($orig_body)) > $maxlen) {
 632                                 if ($textlen < $maxlen) {
 633                                         logger('the limit happens after the end of the last image', LOGGER_DEBUG);
 634                                         $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
 635                                 }
 636                         } else {
 637                                 logger('the text size with embedded images extracted did not violate the limit', LOGGER_DEBUG);
 638                                 $new_body = $new_body . $orig_body;
 639                         }
 640
 641                         return $new_body;
 642                 } else {
 643                         return $body;
 644                 }
 645         }
 646
 647         /**
 648          * Processes [attachment] tags
 649          *
 650          * Note: Can produce a [bookmark] tag in the returned string
 651          *
 652          * @brief Processes [attachment] tags
 653          * @param string $return
 654          * @param bool|int $simplehtml
 655          * @param bool $tryoembed
 656          * @return string
 657          */
 658         private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
 659         {
 660                 $data = self::getAttachmentData($return);
 661                 if (!$data) {
 662                         return $return;
 663                 }
 664
 665                 if (isset($data["title"])) {
 666                         $data["title"] = strip_tags($data["title"]);
 667                         $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
 668                 }
 669
 670                 if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) {
 671                         $data["preview"] = $data["image"];
 672                         $data["image"] = "";
 673                 }
 674
 675                 $return = '';
 676                 if ($simplehtml == 7) {
 677                         $return = self::convertUrlForMastodon($data["url"]);
 678                 } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
 679                         $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data["url"], $data["title"]);
 680                 } else {
 681                         try {
 682                                 if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
 683                                         $return = OEmbed::getHTML($data['url'], $data['title']);
 684                                 } else {
 685                                         throw new Exception('OEmbed is disabled for this attachment.');
 686                                 }
 687                         } catch (Exception $e) {
 688                                 if ($simplehtml != 4) {
 689                                         $return = sprintf('<div class="type-%s">', $data["type"]);
 690                                 }
 691
 692                                 if ($data["image"] != "") {
 693                                         $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], proxy_url($data["image"]), $data["title"]);
 694                                 } elseif ($data["preview"] != "") {
 695                                         $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], proxy_url($data["preview"]), $data["title"]);
 696                                 }
 697
 698                                 if (($data["type"] == "photo") && ($data["url"] != "") && ($data["image"] != "")) {
 699                                         $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], proxy_url($data["image"]), $data["title"]);
 700                                 } else {
 701                                         $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
 702                                 }
 703
 704                                 if ($data["description"] != "" && $data["description"] != $data["title"]) {
 705                                         $return .= sprintf('<blockquote>%s</blockquote>', trim(bbcode($data["description"])));
 706                                 }
 707
 708                                 if ($data["type"] == "link") {
 709                                         $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
 710                                 }
 711
 712                                 if ($simplehtml != 4) {
 713                                         $return .= '</div>';
 714                                 }
 715                         }
 716                 }
 717
 718                 return trim($data["text"] . ' ' . $return . ' ' . $data["after"]);
 719         }
 720
 721         public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
 722         {
 723                 $data = self::getAttachmentData($Text);
 724
 725                 if (!$data) {
 726                         return $Text;
 727                 } elseif ($nolink) {
 728                         return $data["text"] . $data["after"];
 729                 }
 730
 731                 $title = htmlentities($data["title"], ENT_QUOTES, 'UTF-8', false);
 732                 $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
 733                 if ($plaintext || (($title != "") && strstr($text, $title))) {
 734                         $data["title"] = $data["url"];
 735                 } elseif (($text != "") && strstr($title, $text)) {
 736                         $data["text"] = $data["title"];
 737                         $data["title"] = $data["url"];
 738                 }
 739
 740                 if (($data["text"] == "") && ($data["title"] != "") && ($data["url"] == "")) {
 741                         return $data["title"] . $data["after"];
 742                 }
 743
 744                 // If the link already is included in the post, don't add it again
 745                 if (($data["url"] != "") && strpos($data["text"], $data["url"])) {
 746                         return $data["text"] . $data["after"];
 747                 }
 748
 749                 $text = $data["text"];
 750
 751                 if (($data["url"] != "") && ($data["title"] != "")) {
 752                         $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
 753                 } elseif (($data["url"] != "")) {
 754                         $text .= "\n" . $data["url"];
 755                 }
 756
 757                 return $text . "\n" . $data["after"];
 758         }
 759
 760         private static function cleanCss($input)
 761         {
 762                 $cleaned = "";
 763
 764                 $input = strtolower($input);
 765
 766                 for ($i = 0; $i < strlen($input); $i++) {
 767                         $char = substr($input, $i, 1);
 768
 769                         if (($char >= "a") && ($char <= "z")) {
 770                                 $cleaned .= $char;
 771                         }
 772
 773                         if (!(strpos(" #;:0123456789-_.%", $char) === false)) {
 774                                 $cleaned .= $char;
 775                         }
 776                 }
 777
 778                 return $cleaned;
 779         }
 780
 781         /**
 782          * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
 783          *
 784          * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
 785          * @param array $match Array with the matching values
 786          * @return string reformatted link including HTML codes
 787          */
 788         private static function convertUrlForMastodonCallback($match)
 789         {
 790                 $url = $match[1];
 791
 792                 if (isset($match[2]) && ($match[1] != $match[2])) {
 793                         return $match[0];
 794                 }
 795
 796                 $parts = parse_url($url);
 797                 if (!isset($parts['scheme'])) {
 798                         return $match[0];
 799                 }
 800
 801                 return self::convertUrlForMastodon($url);
 802         }
 803
 804         /**
 805          * @brief Converts [url] BBCodes in a format that looks fine on Mastodon and GNU Social.
 806          * @param string $url URL that is about to be reformatted
 807          * @return string reformatted link including HTML codes
 808          */
 809         private static function convertUrlForMastodon($url)
 810         {
 811                 $parts = parse_url($url);
 812                 $scheme = $parts['scheme'] . '://';
 813                 $styled_url = str_replace($scheme, '', $url);
 814
 815                 $html = '<a href="%s" class="attachment" rel="nofollow noopener" target="_blank">' .
 816                         '<span class="invisible">%s</span>';
 817
 818                 if (strlen($styled_url) > 30) {
 819                         $html .= '<span class="ellipsis">%s</span>' .
 820                                 '<span class="invisible">%s</span></a>';
 821
 822                         $ellipsis = substr($styled_url, 0, 30);
 823                         $rest = substr($styled_url, 30);
 824                         return sprintf($html, $url, $scheme, $ellipsis, $rest);
 825                 } else {
 826                         $html .= '%s</a>';
 827                         return sprintf($html, $url, $scheme, $styled_url);
 828                 }
 829         }
 830
 831         /*
 832          * [noparse][i]italic[/i][/noparse] turns into
 833          * [noparse][ i ]italic[ /i ][/noparse],
 834          * to hide them from parser.
 835          */
 836         private static function escapeNoparseCallback($match)
 837         {
 838                 $whole_match = $match[0];
 839                 $captured = $match[1];
 840                 $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured);
 841                 $new_str = str_replace($captured, $spacefied, $whole_match);
 842                 return $new_str;
 843         }
 844
 845         /*
 846          * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
 847          * now turns back and the [noparse] tags are trimed
 848          * returning [i]italic[/i]
 849          */
 850         private static function unescapeNoparseCallback($match)
 851         {
 852                 $captured = $match[1];
 853                 $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured);
 854                 return $unspacefied;
 855         }
 856
 857         /**
 858          * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first
 859          * occurrences
 860          *
 861          * @param string $text        Text to search
 862          * @param string $name        Tag name
 863          * @param int    $occurrences Number of first occurrences to skip
 864          * @return boolean|array
 865          */
 866         public static function getTagPosition($text, $name, $occurrences = 0)
 867         {
 868                 if ($occurrences < 0) {
 869                         $occurrences = 0;
 870                 }
 871
 872                 $start_open = -1;
 873                 for ($i = 0; $i <= $occurrences; $i++) {
 874                         if ($start_open !== false) {
 875                                 $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags
 876                         }
 877                 }
 878
 879                 if ($start_open === false) {
 880                         return false;
 881                 }
 882
 883                 $start_equal = strpos($text, '=', $start_open);
 884                 $start_close = strpos($text, ']', $start_open);
 885
 886                 if ($start_close === false) {
 887                         return false;
 888                 }
 889
 890                 $start_close++;
 891
 892                 $end_open = strpos($text, '[/' . $name . ']', $start_close);
 893
 894                 if ($end_open === false) {
 895                         return false;
 896                 }
 897
 898                 $res = [
 899                         'start' => [
 900                                 'open' => $start_open,
 901                                 'close' => $start_close
 902                         ],
 903                         'end' => [
 904                                 'open' => $end_open,
 905                                 'close' => $end_open + strlen('[/' . $name . ']')
 906                         ],
 907                 ];
 908
 909                 if ($start_equal !== false) {
 910                         $res['start']['equal'] = $start_equal + 1;
 911                 }
 912
 913                 return $res;
 914         }
 915
 916         /**
 917          * Performs a preg_replace within the boundaries of all named BBCode tags in a text
 918          *
 919          * @param type $pattern Preg pattern string
 920          * @param type $replace Preg replace string
 921          * @param type $name    BBCode tag name
 922          * @param type $text    Text to search
 923          * @return string
 924          */
 925         public static function pregReplaceInTag($pattern, $replace, $name, $text)
 926         {
 927                 $occurence = 1;
 928                 $pos = get_bb_tag_pos($text, $name, $occurence);
 929                 while ($pos !== false && $occurence < 1000) {
 930                         $start = substr($text, 0, $pos['start']['open']);
 931                         $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']);
 932                         $end = substr($text, $pos['end']['close']);
 933                         if ($end === false) {
 934                                 $end = '';
 935                         }
 936
 937                         $subject = preg_replace($pattern, $replace, $subject);
 938                         $text = $start . $subject . $end;
 939
 940                         $occurence++;
 941                         $pos = get_bb_tag_pos($text, $name, $occurence);
 942                 }
 943
 944                 return $text;
 945         }
 946
 947         public static function extractImagesFromItemBody($body)
 948         {
 949                 $saved_image = [];
 950                 $orig_body = $body;
 951                 $new_body = '';
 952
 953                 $cnt = 0;
 954                 $img_start = strpos($orig_body, '[img');
 955                 $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
 956                 $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
 957                 while (($img_st_close !== false) && ($img_end !== false)) {
 958                         $img_st_close++; // make it point to AFTER the closing bracket
 959                         $img_end += $img_start;
 960
 961                         if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
 962                                 // This is an embedded image
 963                                 $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close));
 964                                 $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]';
 965
 966                                 $cnt++;
 967                         } else {
 968                                 $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]'));
 969                         }
 970
 971                         $orig_body = substr($orig_body, $img_end + strlen('[/img]'));
 972
 973                         if ($orig_body === false) {
 974                                 // in case the body ends on a closing image tag
 975                                 $orig_body = '';
 976                         }
 977
 978                         $img_start = strpos($orig_body, '[img');
 979                         $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
 980                         $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
 981                 }
 982
 983                 $new_body = $new_body . $orig_body;
 984
 985                 return ['body' => $new_body, 'images' => $saved_image];
 986         }
 987
 988         public static function interpolateSavedImagesIntoItemBody($body, array $images)
 989         {
 990                 $newbody = $body;
 991
 992                 $cnt = 0;
 993                 foreach ($images as $image) {
 994                         // We're depending on the property of 'foreach' (specified on the PHP website) that
 995                         // it loops over the array starting from the first element and going sequentially
 996                         // to the last element
 997                         $newbody = str_replace('[$#saved_image' . $cnt . '#$]',
 998                                 '<img src="' . proxy_url($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
 999                         $cnt++;
1000                 }
1001
1002                 return $newbody;
1003         }
1004
1005         /**
1006          * Processes [share] tags
1007          *
1008          * Note: Can produce a [bookmark] tag in the output
1009          *
1010          * @brief Processes [share] tags
1011          * @param array    $share      preg_match_callback result array
1012          * @param bool|int $simplehtml
1013          * @return string
1014          */
1015         public static function convertShare($share, $simplehtml)
1016         {
1017                 $attributes = $share[2];
1018
1019                 $author = "";
1020                 preg_match("/author='(.*?)'/ism", $attributes, $matches);
1021                 if (x($matches, 1)) {
1022                         $author = html_entity_decode($matches[1], ENT_QUOTES, 'UTF-8');
1023                 }
1024
1025                 preg_match('/author="(.*?)"/ism', $attributes, $matches);
1026                 if (x($matches, 1)) {
1027                         $author = $matches[1];
1028                 }
1029
1030                 $profile = "";
1031                 preg_match("/profile='(.*?)'/ism", $attributes, $matches);
1032                 if (x($matches, 1)) {
1033                         $profile = $matches[1];
1034                 }
1035
1036                 preg_match('/profile="(.*?)"/ism', $attributes, $matches);
1037                 if (x($matches, 1)) {
1038                         $profile = $matches[1];
1039                 }
1040
1041                 $avatar = "";
1042                 preg_match("/avatar='(.*?)'/ism", $attributes, $matches);
1043                 if (x($matches, 1)) {
1044                         $avatar = $matches[1];
1045                 }
1046
1047                 preg_match('/avatar="(.*?)"/ism', $attributes, $matches);
1048                 if (x($matches, 1)) {
1049                         $avatar = $matches[1];
1050                 }
1051
1052                 $link = "";
1053                 preg_match("/link='(.*?)'/ism", $attributes, $matches);
1054                 if (x($matches, 1)) {
1055                         $link = $matches[1];
1056                 }
1057
1058                 preg_match('/link="(.*?)"/ism', $attributes, $matches);
1059                 if (x($matches, 1)) {
1060                         $link = $matches[1];
1061                 }
1062
1063                 $posted = "";
1064
1065                 preg_match("/posted='(.*?)'/ism", $attributes, $matches);
1066                 if (x($matches, 1)) {
1067                         $posted = $matches[1];
1068                 }
1069
1070                 preg_match('/posted="(.*?)"/ism', $attributes, $matches);
1071                 if (x($matches, 1)) {
1072                         $posted = $matches[1];
1073                 }
1074
1075                 // We only call this so that a previously unknown contact can be added.
1076                 // This is important for the function "get_contact_details_by_url".
1077                 // This function then can fetch an entry from the contact table.
1078                 Contact::getIdForURL($profile, 0);
1079
1080                 $data = Contact::getDetailsByURL($profile);
1081
1082                 if (x($data, "name") && x($data, "addr")) {
1083                         $userid_compact = $data["name"] . " (" . $data["addr"] . ")";
1084                 } else {
1085                         $userid_compact = GetProfileUsername($profile, $author, true);
1086                 }
1087
1088                 if (x($data, "addr")) {
1089                         $userid = $data["addr"];
1090                 } else {
1091                         $userid = GetProfileUsername($profile, $author, false);
1092                 }
1093
1094                 if (x($data, "name")) {
1095                         $author = $data["name"];
1096                 }
1097
1098                 if (x($data, "micro")) {
1099                         $avatar = $data["micro"];
1100                 }
1101
1102                 $preshare = trim($share[1]);
1103                 if ($preshare != "") {
1104                         $preshare .= "<br />";
1105                 }
1106
1107                 switch ($simplehtml) {
1108                         case 1:
1109                                 $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . ' <a href="' . $profile . '">' . $userid . "</a>: <br />»" . $share[3] . "«";
1110                                 break;
1111                         case 2:
1112                                 $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1113                                 break;
1114                         case 3: // Diaspora
1115                                 $headline .= '<b>' . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . $userid . ':</b><br />';
1116
1117                                 $text = trim($share[1]);
1118
1119                                 if ($text != "") {
1120                                         $text .= "<hr />";
1121                                 }
1122
1123                                 if (stripos(normalise_link($link), 'http://twitter.com/') === 0) {
1124                                         $text .= $headline . '<blockquote>' . trim($share[3]) . "</blockquote><br />";
1125
1126                                         if ($link != "") {
1127                                                 $text .= '<br /><a href="' . $link . '">[l]</a>';
1128                                         }
1129                                 } else {
1130                                         $text .= '<br /><a href="' . $link . '">' . $link . '</a>';
1131                                 }
1132
1133                                 break;
1134                         case 4:
1135                                 $headline .= '<br /><b>' . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8');
1136                                 $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $link, $userid, $posted);
1137                                 $headline .= ":</b><br />";
1138
1139                                 $text = trim($share[1]);
1140
1141                                 if ($text != "") {
1142                                         $text .= "<hr />";
1143                                 }
1144
1145                                 $text .= $headline . '<blockquote class="shared_content">' . trim($share[3]) . "</blockquote><br />";
1146
1147                                 break;
1148                         case 5:
1149                                 $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1150                                 break;
1151                         case 6: // app.net
1152                                 $text = $preshare . "&gt;&gt; @" . $userid_compact . ": <br />" . $share[3];
1153                                 break;
1154                         case 7: // statusnet/GNU Social
1155                                 $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3];
1156                                 break;
1157                         case 8: // twitter
1158                                 $text = $preshare . "RT @" . $userid_compact . ": " . $share[3];
1159                                 break;
1160                         case 9: // Google+/Facebook
1161                                 $text = $preshare . html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
1162
1163                                 if ($link != "") {
1164                                         $text .= "<br /><br />" . $link;
1165                                 }
1166                                 break;
1167                         default:
1168                                 // Transforms quoted tweets in rich attachments to avoid nested tweets
1169                                 if (stripos(normalise_link($link), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($link)) {
1170                                         try {
1171                                                 $oembed = OEmbed::getHTML($link, $preshare);
1172                                         } catch (Exception $e) {
1173                                                 $oembed = sprintf('[bookmark=%s]%s[/bookmark]', $link, $preshare);
1174                                         }
1175
1176                                         $text = $preshare . $oembed;
1177                                 } else {
1178                                         $text = trim($share[1]) . "\n";
1179
1180                                         $avatar = proxy_url($avatar, false, PROXY_SIZE_THUMB);
1181
1182                                         $tpl = get_markup_template('shared_content.tpl');
1183                                         $text .= replace_macros($tpl, [
1184                                                 '$profile' => $profile,
1185                                                 '$avatar' => $avatar,
1186                                                 '$author' => $author,
1187                                                 '$link' => $link,
1188                                                 '$posted' => $posted,
1189                                                 '$content' => trim($share[3])
1190                                         ]);
1191                                 }
1192                                 break;
1193                 }
1194
1195                 return $text;
1196         }
1197
1198         private static function removePictureLinksCallback($match)
1199         {
1200                 $text = Cache::get($match[1]);
1201
1202                 if (is_null($text)) {
1203                         $a = get_app();
1204
1205                         $stamp1 = microtime(true);
1206
1207                         $ch = @curl_init($match[1]);
1208                         @curl_setopt($ch, CURLOPT_NOBODY, true);
1209                         @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1210                         @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
1211                         @curl_exec($ch);
1212                         $curl_info = @curl_getinfo($ch);
1213
1214                         $a->save_timestamp($stamp1, "network");
1215
1216                         if (substr($curl_info["content_type"], 0, 6) == "image/") {
1217                                 $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
1218                         } else {
1219                                 $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
1220
1221                                 // if its not a picture then look if its a page that contains a picture link
1222                                 $body = Network::fetchUrl($match[1]);
1223
1224                                 $doc = new DOMDocument();
1225                                 @$doc->loadHTML($body);
1226                                 $xpath = new DomXPath($doc);
1227                                 $list = $xpath->query("//meta[@name]");
1228                                 foreach ($list as $node) {
1229                                         $attr = [];
1230
1231                                         if ($node->attributes->length) {
1232                                                 foreach ($node->attributes as $attribute) {
1233                                                         $attr[$attribute->name] = $attribute->value;
1234                                                 }
1235                                         }
1236
1237                                         if (strtolower($attr["name"]) == "twitter:image") {
1238                                                 $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
1239                                         }
1240                                 }
1241                         }
1242                         Cache::set($match[1], $text);
1243                 }
1244
1245                 return $text;
1246         }
1247
1248         private static function expandLinksCallback($match)
1249         {
1250                 if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
1251                         return ($match[1] . "[url]" . $match[2] . "[/url]");
1252                 } else {
1253                         return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]");
1254                 }
1255         }
1256
1257         private static function cleanPictureLinksCallback($match)
1258         {
1259                 $text = Cache::get($match[1]);
1260
1261                 if (is_null($text)) {
1262                         $a = get_app();
1263
1264                         $stamp1 = microtime(true);
1265
1266                         $ch = @curl_init($match[1]);
1267                         @curl_setopt($ch, CURLOPT_NOBODY, true);
1268                         @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
1269                         @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
1270                         @curl_exec($ch);
1271                         $curl_info = @curl_getinfo($ch);
1272
1273                         $a->save_timestamp($stamp1, "network");
1274
1275                         // if its a link to a picture then embed this picture
1276                         if (substr($curl_info["content_type"], 0, 6) == "image/") {
1277                                 $text = "[img]" . $match[1] . "[/img]";
1278                         } else {
1279                                 $text = "[img]" . $match[2] . "[/img]";
1280
1281                                 // if its not a picture then look if its a page that contains a picture link
1282                                 $body = Network::fetchUrl($match[1]);
1283
1284                                 $doc = new DOMDocument();
1285                                 @$doc->loadHTML($body);
1286                                 $xpath = new DomXPath($doc);
1287                                 $list = $xpath->query("//meta[@name]");
1288                                 foreach ($list as $node) {
1289                                         $attr = [];
1290                                         if ($node->attributes->length) {
1291                                                 foreach ($node->attributes as $attribute) {
1292                                                         $attr[$attribute->name] = $attribute->value;
1293                                                 }
1294                                         }
1295
1296                                         if (strtolower($attr["name"]) == "twitter:image") {
1297                                                 $text = "[img]" . $attr["content"] . "[/img]";
1298                                         }
1299                                 }
1300                         }
1301                         Cache::set($match[1], $text);
1302                 }
1303
1304                 return $text;
1305         }
1306
1307         public static function cleanPictureLinks($text)
1308         {
1309                 $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
1310                 return $return;
1311         }
1312
1313         private static function textHighlightCallback($match)
1314         {
1315                 if (in_array(strtolower($match[1]),
1316                                 ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby',
1317                                 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh'])
1318                 ) {
1319                         return text_highlight($match[2], strtolower($match[1]));
1320                 }
1321                 return $match[0];
1322         }
1323
1324         /**
1325          * @brief Converts a BBCode message to HTML message
1326          *
1327          * BBcode 2 HTML was written by WAY2WEB.net
1328          * extended to work with Mistpark/Friendica - Mike Macgirvin
1329          *
1330          * Simple HTML values meaning:
1331          * - 0: Friendica display
1332          * - 1: Unused
1333          * - 2: Used for Facebook, Google+, Windows Phone push, Friendica API
1334          * - 3: Used before converting to Markdown in bb2diaspora.php
1335          * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
1336          * - 5: Unused
1337          * - 6: Used for Appnet
1338          * - 7: Used for dfrn, OStatus
1339          * - 8: Used for WP backlink text setting
1340          *
1341          * @param string $text
1342          * @param bool   $preserve_nl
1343          * @param bool   $try_oembed
1344          * @param int    $simple_html
1345          * @param bool   $for_plaintext
1346          * @return string
1347          */
1348         public static function convert($text, $preserve_nl = false, $try_oembed = true, $simple_html = false, $for_plaintext = false)
1349         {
1350                 $a = get_app();
1351
1352                 /*
1353                  * preg_match_callback function to replace potential Oembed tags with Oembed content
1354                  *
1355                  * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
1356                  * $match[1] = $url
1357                  * $match[2] = $title or absent
1358                  */
1359                 $try_oembed_callback = function ($match)
1360                 {
1361                         $url = $match[1];
1362                         $title = defaults($match, 2, null);
1363
1364                         try {
1365                                 $return = OEmbed::getHTML($url, $title);
1366                         } catch (Exception $ex) {
1367                                 $return = $match[0];
1368                         }
1369
1370                         return $return;
1371                 };
1372
1373                 // Hide all [noparse] contained bbtags by spacefying them
1374                 // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
1375
1376                 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
1377                 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
1378                 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
1379
1380                 // Remove the abstract element. It is a non visible element.
1381                 $text = remove_abstract($text);
1382
1383                 // Move all spaces out of the tags
1384                 $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text);
1385                 $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text);
1386
1387                 // Extract the private images which use data urls since preg has issues with
1388                 // large data sizes. Stash them away while we do bbcode conversion, and then put them back
1389                 // in after we've done all the regex matching. We cannot use any preg functions to do this.
1390
1391                 $extracted = bb_extract_images($text);
1392                 $text = $extracted['body'];
1393                 $saved_image = $extracted['images'];
1394
1395                 // If we find any event code, turn it into an event.
1396                 // After we're finished processing the bbcode we'll
1397                 // replace all of the event code with a reformatted version.
1398
1399                 $ev = bbtoevent($text);
1400
1401                 // Replace any html brackets with HTML Entities to prevent executing HTML or script
1402                 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
1403
1404                 $text = str_replace("<", "&lt;", $text);
1405                 $text = str_replace(">", "&gt;", $text);
1406
1407                 // remove some newlines before the general conversion
1408                 $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text);
1409                 $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text);
1410
1411                 $text = preg_replace("/\n\[code\]/ism", "[code]", $text);
1412                 $text = preg_replace("/\[\/code\]\n/ism", "[/code]", $text);
1413
1414                 // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
1415                 if (!$try_oembed) {
1416                         $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
1417                 }
1418
1419                 // Check for [code] text here, before the linefeeds are messed with.
1420                 // The highlighter will unescape and re-escape the content.
1421                 if (strpos($text, '[code=') !== false) {
1422                         $text = preg_replace_callback("/\[code=(.*?)\](.*?)\[\/code\]/ism", 'self::textHighlightCallback', $text);
1423                 }
1424                 // Convert new line chars to html <br /> tags
1425
1426                 // nlbr seems to be hopelessly messed up
1427                 //      $Text = nl2br($Text);
1428
1429                 // We'll emulate it.
1430
1431                 $text = trim($text);
1432                 $text = str_replace("\r\n", "\n", $text);
1433
1434                 // removing multiplicated newlines
1435                 if (Config::get("system", "remove_multiplicated_lines")) {
1436                         $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
1437                                         "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
1438                         $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
1439                                         "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
1440                         do {
1441                                 $oldtext = $text;
1442                                 $text = str_replace($search, $replace, $text);
1443                         } while ($oldtext != $text);
1444                 }
1445
1446                 // Set up the parameters for a URL search string
1447                 $URLSearchString = "^\[\]";
1448                 // Set up the parameters for a MAIL search string
1449                 $MAILSearchString = $URLSearchString;
1450
1451                 // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text
1452                 if (!$for_plaintext) {
1453                         // Autolink feature (thanks to http://code.seebz.net/p/autolink-php/)
1454                         // Currently disabled, since the function is too greedy
1455                         // $autolink_regex = "`([^\]\=\"']|^)(https?\://[^\s<]+[^\s<\.\)])`ism";
1456                         $autolink_regex = "/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism";
1457                         $text = preg_replace($autolink_regex, '$1[url]$2[/url]', $text);
1458                         if ($simple_html == 7) {
1459                                 $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForMastodonCallback', $text);
1460                                 $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForMastodonCallback', $text);
1461                         }
1462                 } else {
1463                         $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text);
1464                         $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
1465                 }
1466
1467
1468                 // Handle attached links or videos
1469                 $text = self::convertAttachment($text, $simple_html, $try_oembed);
1470
1471                 $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
1472
1473                 if ($preserve_nl) {
1474                         $text = str_replace(["\n", "\r"], ['', ''], $text);
1475                 }
1476
1477                 // Remove all hashtag addresses
1478                 if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7])) {
1479                         $text = preg_replace("/([#@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $text);
1480                 } elseif ($simple_html == 3) {
1481                         // The ! is converted to @ since Diaspora only understands the @
1482                         $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1483                                 '@<a href="$2">$3</a>',
1484                                 $text);
1485                 } elseif ($simple_html == 7) {
1486                         $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1487                                 '$1<span class="vcard"><a href="$2" class="url" title="$3"><span class="fn nickname mention">$3</span></a></span>',
1488                                 $text);
1489                 } elseif (!$simple_html) {
1490                         $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1491                                 '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
1492                                 $text);
1493                 }
1494
1495                 // Bookmarks in red - will be converted to bookmarks in friendica
1496                 $text = preg_replace("/#\^\[url\]([$URLSearchString]*)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
1497                 $text = preg_replace("/#\^\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
1498                 $text = preg_replace("/#\[url\=[$URLSearchString]*\]\^\[\/url\]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i",
1499                                         "[bookmark=$1]$2[/bookmark]", $text);
1500
1501                 if (in_array($simple_html, [2, 6, 7, 8, 9])) {
1502                         $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
1503                         //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
1504                         $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
1505                 }
1506
1507                 if ($simple_html == 5) {
1508                         $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
1509                 }
1510
1511                 // Perform URL Search
1512                 if ($try_oembed) {
1513                         $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
1514                 }
1515
1516                 if ($simple_html == 5) {
1517                         $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text);
1518                 } else {
1519                         $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
1520                 }
1521
1522                 // Handle Diaspora posts
1523                 $text = preg_replace_callback(
1524                         "&\[url=/posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
1525                         function ($match) {
1526                                 return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
1527                         }, $text
1528                 );
1529
1530                 // Server independent link to posts and comments
1531                 // See issue: https://github.com/diaspora/diaspora_federation/issues/75
1532                 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
1533                 $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
1534
1535                 $text = preg_replace("/([#])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
1536                                         '$1<a href="' . System::baseUrl() . '/search?tag=$3" class="tag" title="$3">$3</a>', $text);
1537
1538                 $text = preg_replace("/\[url\=([$URLSearchString]*)\]#(.*?)\[\/url\]/ism",
1539                                         '#<a href="' . System::baseUrl() . '/search?tag=$2" class="tag" title="$2">$2</a>', $text);
1540
1541                 $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
1542                 $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1543                 //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $Text);
1544
1545                 // Red compatibility, though the link can't be authenticated on Friendica
1546                 $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
1547
1548
1549                 // we may need to restrict this further if it picks up too many strays
1550                 // link acct:user@host to a webfinger profile redirector
1551
1552                 $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', '<a href="' . System::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
1553
1554                 // Perform MAIL Search
1555                 $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
1556                 $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
1557
1558                 // leave open the posibility of [map=something]
1559                 // this is replaced in prepare_body() which has knowledge of the item location
1560
1561                 if (strpos($text, '[/map]') !== false) {
1562                         $text = preg_replace_callback(
1563                                 "/\[map\](.*?)\[\/map\]/ism",
1564                                 function ($match) {
1565                                         // the extra space in the following line is intentional
1566                                         // Whyyy? - @MrPetovan
1567                                         return str_replace($match[0], '<div class="map"  >' . Map::byLocation($match[1]) . '</div>', $match[0]);
1568                                 },
1569                                 $text
1570                         );
1571                 }
1572                 if (strpos($text, '[map=') !== false) {
1573                         $text = preg_replace_callback(
1574                                 "/\[map=(.*?)\]/ism",
1575                                 function ($match) {
1576                                         // the extra space in the following line is intentional
1577                                         // Whyyy? - @MrPetovan
1578                                         return str_replace($match[0], '<div class="map"  >' . Map::byCoordinates(str_replace('/', ' ', $match[1])) . '</div>', $match[0]);
1579                                 },
1580                                 $text
1581                         );
1582                 }
1583                 if (strpos($text, '[map]') !== false) {
1584                         $text = preg_replace("/\[map\]/", '<div class="map"></div>', $text);
1585                 }
1586
1587                 // Check for headers
1588                 $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
1589                 $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
1590                 $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
1591                 $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
1592                 $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
1593                 $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
1594
1595                 // Check for paragraph
1596                 $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
1597
1598                 // Check for bold text
1599                 $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
1600
1601                 // Check for Italics text
1602                 $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
1603
1604                 // Check for Underline text
1605                 $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
1606
1607                 // Check for strike-through text
1608                 $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<strike>$1</strike>', $text);
1609
1610                 // Check for over-line text
1611                 $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
1612
1613                 // Check for colored text
1614                 $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
1615
1616                 // Check for sized text
1617                 // [size=50] --> font-size: 50px (with the unit).
1618                 $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
1619                 $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
1620
1621                 // Check for centered text
1622                 $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
1623
1624                 // Check for list text
1625                 $text = str_replace("[*]", "<li>", $text);
1626
1627                 // Check for style sheet commands
1628                 $text = preg_replace_callback(
1629                         "(\[style=(.*?)\](.*?)\[\/style\])ism",
1630                         function ($match) {
1631                                 return "<span style=\"" . self::cleanCss($match[1]) . ";\">" . $match[2] . "</span>";
1632                         },
1633                         $text
1634                 );
1635
1636                 // Check for CSS classes
1637                 $text = preg_replace_callback(
1638                         "(\[class=(.*?)\](.*?)\[\/class\])ism",
1639                         function ($match) {
1640                                 return "<span class=\"" . self::cleanCss($match[1]) . "\">" . $match[2] . "</span>";
1641                         },
1642                         $text
1643                 );
1644
1645                 // handle nested lists
1646                 $endlessloop = 0;
1647
1648                 while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) ||
1649                            ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
1650                            ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
1651                            ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
1652                         $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1653                         $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
1654                         $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1655                         $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
1656                         $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
1657                         $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
1658                         $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
1659                         $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
1660                         $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
1661                         $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
1662                 }
1663
1664                 $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
1665                 $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
1666                 $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
1667                 $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
1668
1669                 $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
1670                 $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
1671
1672                 $text = str_replace('[hr]', '<hr />', $text);
1673
1674                 // This is actually executed in prepare_body()
1675
1676                 $text = str_replace('[nosmile]', '', $text);
1677
1678                 // Check for font change text
1679                 $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
1680
1681                 // Declare the format for [code] layout
1682
1683                 $CodeLayout = '<code>$1</code>';
1684                 // Check for [code] text
1685                 $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text);
1686
1687                 // Declare the format for [spoiler] layout
1688                 $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
1689
1690                 // Check for [spoiler] text
1691                 // handle nested quotes
1692                 $endlessloop = 0;
1693                 while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
1694                         $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
1695                 }
1696
1697                 // Check for [spoiler=Author] text
1698
1699                 $t_wrote = L10n::t('$1 wrote:');
1700
1701                 // handle nested quotes
1702                 $endlessloop = 0;
1703                 while ((strpos($text, "[/spoiler]")!== false)  && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
1704                         $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
1705                                                  "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
1706                                                  $text);
1707                 }
1708
1709                 // Declare the format for [quote] layout
1710                 $QuoteLayout = '<blockquote>$1</blockquote>';
1711
1712                 // Check for [quote] text
1713                 // handle nested quotes
1714                 $endlessloop = 0;
1715                 while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
1716                         $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
1717                 }
1718
1719                 // Check for [quote=Author] text
1720
1721                 $t_wrote = L10n::t('$1 wrote:');
1722
1723                 // handle nested quotes
1724                 $endlessloop = 0;
1725                 while ((strpos($text, "[/quote]")!== false)  && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
1726                         $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
1727                                                  "<br /><strong class=".'"author"'.">" . $t_wrote . "</strong><blockquote>$2</blockquote>",
1728                                                  $text);
1729                 }
1730
1731
1732                 // [img=widthxheight]image source[/img]
1733                 $text = preg_replace_callback(
1734                         "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
1735                         function ($matches) {
1736                                 if (strpos($matches[3], "data:image/") === 0) {
1737                                         return $matches[0];
1738                                 }
1739
1740                                 $matches[3] = proxy_url($matches[3]);
1741                                 return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
1742                         },
1743                         $text
1744                 );
1745
1746                 $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
1747                 $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
1748
1749                 // Images
1750                 // [img]pathtoimage[/img]
1751                 $text = preg_replace_callback(
1752                         "/\[img\](.*?)\[\/img\]/ism",
1753                         function ($matches) {
1754                                 if (strpos($matches[1], "data:image/") === 0) {
1755                                         return $matches[0];
1756                                 }
1757
1758                                 $matches[1] = proxy_url($matches[1]);
1759                                 return "[img]" . $matches[1] . "[/img]";
1760                         },
1761                         $text
1762                 );
1763
1764                 $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1765                 $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
1766
1767                 // Shared content
1768                 $text = preg_replace_callback("/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
1769                         function ($match) use ($simple_html) {
1770                                 return bb_ShareAttributes($match, $simple_html);
1771                         }, $text);
1772
1773                 $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . L10n::t('Encrypted content') . '" /><br />', $text);
1774                 $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $text);
1775                 //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $Text);
1776
1777                 // Try to Oembed
1778                 if ($try_oembed) {
1779                         $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4))\[\/video\]/ism", '<video src="$1" controls="controls" width="' . $a->videowidth . '" height="' . $a->videoheight . '" loop="true"><a href="$1">$1</a></video>', $text);
1780                         $text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3))\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
1781
1782                         $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
1783                         $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
1784                 } else {
1785                         $text = preg_replace("/\[video\](.*?)\[\/video\]/",
1786                                                 '<a href="$1" target="_blank">$1</a>', $text);
1787                         $text = preg_replace("/\[audio\](.*?)\[\/audio\]/",
1788                                                 '<a href="$1" target="_blank">$1</a>', $text);
1789                 }
1790
1791                 // html5 video and audio
1792
1793
1794                 if ($try_oembed) {
1795                         $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
1796                 } else {
1797                         $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
1798                 }
1799
1800                 // Youtube extensions
1801                 if ($try_oembed) {
1802                         $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1803                         $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1804                         $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
1805                 }
1806
1807                 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1808                 $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1809                 $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
1810
1811                 if ($try_oembed) {
1812                         $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);
1813                 } else {
1814                         $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
1815                                                 '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
1816                 }
1817
1818                 if ($try_oembed) {
1819                         $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1820                         $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
1821                 }
1822
1823                 $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1824                 $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
1825
1826                 if ($try_oembed) {
1827                         $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);
1828                 } else {
1829                         $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
1830                                                 '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
1831                 }
1832
1833                 // oembed tag
1834                 $text = OEmbed::BBCode2HTML($text);
1835
1836                 // Avoid triple linefeeds through oembed
1837                 $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
1838
1839                 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
1840                 // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
1841                 // Summary (e.g. title) is required, earlier revisions only required description (in addition to
1842                 // start which is always required). Allow desc with a missing summary for compatibility.
1843
1844                 if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) {
1845                         $sub = format_event_html($ev, $simple_html);
1846
1847                         $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
1848                         $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
1849                         $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text);
1850                         $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text);
1851                         $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text);
1852                         $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text);
1853                         $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text);
1854                 }
1855
1856                 // Replace non graphical smilies for external posts
1857                 if ($simple_html) {
1858                         $text = Smilies::replace($text, false, true);
1859                 }
1860
1861                 // Replace inline code blocks
1862                 $text = preg_replace_callback("|(?!<br[^>]*>)<code>([^<]*)</code>(?!<br[^>]*>)|ism",
1863                         function ($match) use ($simple_html) {
1864                                 $return = '<key>' . $match[1] . '</key>';
1865                                 // Use <code> for Diaspora inline code blocks
1866                                 if ($simple_html === 3) {
1867                                         $return = '<code>' . $match[1] . '</code>';
1868                                 }
1869                                 return $return;
1870                         }
1871                 , $text);
1872
1873                 // Unhide all [noparse] contained bbtags unspacefying them
1874                 // and triming the [noparse] tag.
1875
1876                 $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text);
1877                 $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
1878                 $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
1879
1880
1881                 $text = preg_replace('/\[\&amp\;([#a-z0-9]+)\;\]/', '&$1;', $text);
1882                 $text = preg_replace('/\&\#039\;/', '\'', $text);
1883                 $text = preg_replace('/\&quot\;/', '"', $text);
1884
1885                 // fix any escaped ampersands that may have been converted into links
1886                 $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&amp\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
1887
1888                 // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
1889                 $allowed_src_protocols = ['http', 'redir', 'cid'];
1890                 $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
1891                                          '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text);
1892
1893                 // sanitize href attributes (only whitelisted protocols URLs)
1894                 // default value for backward compatibility
1895                 $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']);
1896
1897                 // Always allowed protocol even if config isn't set or not including it
1898                 $allowed_link_protocols[] = 'http';
1899                 $allowed_link_protocols[] = 'redir/';
1900
1901                 $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
1902                 $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text);
1903
1904                 if ($saved_image) {
1905                         $text = bb_replace_images($text, $saved_image);
1906                 }
1907
1908                 // Clean up the HTML by loading and saving the HTML with the DOM.
1909                 // Bad structured html can break a whole page.
1910                 // For performance reasons do it only with ativated item cache or at export.
1911                 if (!$try_oembed || (get_itemcachepath() != "")) {
1912                         $doc = new DOMDocument();
1913                         $doc->preserveWhiteSpace = false;
1914
1915                         $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8");
1916
1917                         $doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
1918                         $encoding = '<?xml encoding="UTF-8">';
1919                         @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
1920                         $doc->encoding = 'UTF-8';
1921                         $text = $doc->saveHTML();
1922                         $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
1923
1924                         $text = str_replace('<br></li>', '</li>', $text);
1925
1926                         //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
1927                 }
1928
1929                 // Clean up some useless linebreaks in lists
1930                 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1931                 //$Text = str_replace('</ul><br />', '</ul>', $Text);
1932                 //$Text = str_replace('</li><br />', '</li>', $Text);
1933                 //$Text = str_replace('<br /><li>', '<li>', $Text);
1934                 //$Text = str_replace('<br /><ul', '<ul ', $Text);
1935
1936                 Addon::callHooks('bbcode', $text);
1937
1938                 return trim($text);
1939         }
1940
1941         /**
1942          * @brief Removes the "abstract" element from the text
1943          *
1944          * @param string $text The text with BBCode
1945          * @return string The same text - but without "abstract" element
1946          */
1947         public static function removeAbstract($text)
1948         {
1949                 $text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", '', $text);
1950                 $text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", '', $text);
1951
1952                 return $text;
1953         }
1954
1955         /**
1956          * @brief Returns the value of the "abstract" element
1957          *
1958          * @param string $text The text that maybe contains the element
1959          * @param string $addon The addon for which the abstract is meant for
1960          * @return string The abstract
1961          */
1962         public static function getAbstract($text, $addon = "")
1963         {
1964                 $abstract = "";
1965                 $abstracts = [];
1966                 $addon = strtolower($addon);
1967
1968                 if (preg_match_all("/\[abstract=(.*?)\](.*?)\[\/abstract\]/ism", $text, $results, PREG_SET_ORDER)) {
1969                         foreach ($results AS $result) {
1970                                 $abstracts[strtolower($result[1])] = $result[2];
1971                         }
1972                 }
1973
1974                 if (isset($abstracts[$addon])) {
1975                         $abstract = $abstracts[$addon];
1976                 }
1977
1978                 if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
1979                         $abstract = $result[1];
1980                 }
1981
1982                 return $abstract;
1983         }
1984 }