src/Protocol/Feed.php

   1 <?php
   2 /**
   3  * @copyright Copyright (C) 2020, Friendica
   4  *
   5  * @license GNU AGPL version 3 or any later version
   6  *
   7  * This program is free software: you can redistribute it and/or modify
   8  * it under the terms of the GNU Affero General Public License as
   9  * published by the Free Software Foundation, either version 3 of the
  10  * License, or (at your option) any later version.
  11  *
  12  * This program is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15  * GNU Affero General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU Affero General Public License
  18  * along with this program.  If not, see <https://www.gnu.org/licenses/>.
  19  *
  20  */
  21
  22 namespace Friendica\Protocol;
  23
  24 use DOMDocument;
  25 use DOMXPath;
  26 use Friendica\Content\Text\HTML;
  27 use Friendica\Core\Logger;
  28 use Friendica\Core\Protocol;
  29 use Friendica\Database\DBA;
  30 use Friendica\DI;
  31 use Friendica\Model\Item;
  32 use Friendica\Model\Tag;
  33 use Friendica\Util\Network;
  34 use Friendica\Util\ParseUrl;
  35 use Friendica\Util\XML;
  36
  37 /**
  38  * This class contain functions to import feeds (RSS/RDF/Atom)
  39  */
  40 class Feed
  41 {
  42         /**
  43          * consume - process atom feed and update anything/everything we might need to update
  44          *
  45          * $xml = the (atom) feed to consume - RSS isn't as fully supported but may work for simple feeds.
  46          *
  47          * $importer = the contact_record (joined to user_record) of the local user who owns this relationship.
  48          *             It is this person's stuff that is going to be updated.
  49          * $contact =  the person who is sending us stuff. If not set, we MAY be processing a "follow" activity
  50          *             from an external network and MAY create an appropriate contact record. Otherwise, we MUST
  51          *             have a contact record.
  52          * $hub = should we find a hub declation in the feed, pass it back to our calling process, who might (or
  53          *        might not) try and subscribe to it.
  54          * $datedir sorts in reverse order
  55          * $pass - by default ($pass = 0) we cannot guarantee that a parent item has been
  56          *      imported prior to its children being seen in the stream unless we are certain
  57          *      of how the feed is arranged/ordered.
  58          * With $pass = 1, we only pull parent items out of the stream.
  59          * With $pass = 2, we only pull children (comments/likes).
  60          *
  61          * So running this twice, first with pass 1 and then with pass 2 will do the right
  62          * thing regardless of feed ordering. This won't be adequate in a fully-threaded
  63          * model where comments can have sub-threads. That would require some massive sorting
  64          * to get all the feed items into a mostly linear ordering, and might still require
  65          * recursion.
  66          *
  67          * @param       $xml
  68          * @param array $importer
  69          * @param array $contact
  70          * @param       $hub
  71          * @throws ImagickException
  72          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
  73          */
  74         public static function consume($xml, array $importer, array $contact, &$hub)
  75         {
  76                 if ($contact['network'] === Protocol::OSTATUS) {
  77                         Logger::info('Consume OStatus messages');
  78                         OStatus::import($xml, $importer, $contact, $hub);
  79
  80                         return;
  81                 }
  82
  83                 if ($contact['network'] === Protocol::FEED) {
  84                         Logger::info('Consume feeds');
  85                         self::import($xml, $importer, $contact);
  86
  87                         return;
  88                 }
  89
  90                 if ($contact['network'] === Protocol::DFRN) {
  91                         Logger::info('Consume DFRN messages');
  92                         $dfrn_importer = DFRN::getImporter($contact['id'], $importer['uid']);
  93                         if (!empty($dfrn_importer)) {
  94                                 Logger::info('Now import the DFRN feed');
  95                                 DFRN::import($xml, $dfrn_importer, true);
  96                                 return;
  97                         }
  98                 }
  99         }
 100
 101         /**
 102          * Read a RSS/RDF/Atom feed and create an item entry for it
 103          *
 104          * @param string $xml      The feed data
 105          * @param array  $importer The user record of the importer
 106          * @param array  $contact  The contact record of the feed
 107          *
 108          * @return array Returns the header and the first item in dry run mode
 109          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
 110          */
 111         public static function import($xml, array $importer = [], array $contact = [])
 112         {
 113                 $dryRun = empty($importer) && empty($contact);
 114
 115                 if ($dryRun) {
 116                         Logger::info("Test Atom/RSS feed");
 117                 } else {
 118                         Logger::info("Import Atom/RSS feed '" . $contact["name"] . "' (Contact " . $contact["id"] . ") for user " . $importer["uid"]);
 119                 }
 120
 121                 if (empty($xml)) {
 122                         Logger::info('XML is empty.');
 123                         return [];
 124                 }
 125
 126                 if (!empty($contact['poll'])) {
 127                         $basepath = $contact['poll'];
 128                 } elseif (!empty($contact['url'])) {
 129                         $basepath = $contact['url'];
 130                 } else {
 131                         $basepath = '';
 132                 }
 133
 134                 $doc = new DOMDocument();
 135                 @$doc->loadXML(trim($xml));
 136                 $xpath = new DOMXPath($doc);
 137                 $xpath->registerNamespace('atom', ActivityNamespace::ATOM1);
 138                 $xpath->registerNamespace('dc', "http://purl.org/dc/elements/1.1/");
 139                 $xpath->registerNamespace('content', "http://purl.org/rss/1.0/modules/content/");
 140                 $xpath->registerNamespace('rdf', "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
 141                 $xpath->registerNamespace('rss', "http://purl.org/rss/1.0/");
 142                 $xpath->registerNamespace('media', "http://search.yahoo.com/mrss/");
 143                 $xpath->registerNamespace('poco', ActivityNamespace::POCO);
 144
 145                 $author = [];
 146                 $entries = null;
 147
 148                 // Is it RDF?
 149                 if ($xpath->query('/rdf:RDF/rss:channel')->length > 0) {
 150                         $author["author-link"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:link/text()');
 151                         $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:title/text()');
 152
 153                         if (empty($author["author-name"])) {
 154                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:description/text()');
 155                         }
 156                         $entries = $xpath->query('/rdf:RDF/rss:item');
 157                 }
 158
 159                 // Is it Atom?
 160                 if ($xpath->query('/atom:feed')->length > 0) {
 161                         $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']");
 162                         if (is_object($alternate)) {
 163                                 foreach ($alternate AS $attribute) {
 164                                         if ($attribute->name == "href") {
 165                                                 $author["author-link"] = $attribute->textContent;
 166                                         }
 167                                 }
 168                         }
 169
 170                         if (empty($author["author-link"])) {
 171                                 $self = XML::getFirstAttributes($xpath, "atom:link[@rel='self']");
 172                                 if (is_object($self)) {
 173                                         foreach ($self AS $attribute) {
 174                                                 if ($attribute->name == "href") {
 175                                                         $author["author-link"] = $attribute->textContent;
 176                                                 }
 177                                         }
 178                                 }
 179                         }
 180
 181                         if (empty($author["author-link"])) {
 182                                 $author["author-link"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:id/text()');
 183                         }
 184                         $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:logo/text()');
 185
 186                         $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:title/text()');
 187
 188                         if (empty($author["author-name"])) {
 189                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:subtitle/text()');
 190                         }
 191
 192                         if (empty($author["author-name"])) {
 193                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:name/text()');
 194                         }
 195
 196                         $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:displayName/text()');
 197                         if ($value != "") {
 198                                 $author["author-name"] = $value;
 199                         }
 200
 201                         if ($dryRun) {
 202                                 $author["author-id"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:id/text()');
 203
 204                                 // See https://tools.ietf.org/html/rfc4287#section-3.2.2
 205                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/atom:uri/text()');
 206                                 if ($value != "") {
 207                                         $author["author-link"] = $value;
 208                                 }
 209
 210                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:preferredUsername/text()');
 211                                 if ($value != "") {
 212                                         $author["author-nick"] = $value;
 213                                 }
 214
 215                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:address/poco:formatted/text()');
 216                                 if ($value != "") {
 217                                         $author["author-location"] = $value;
 218                                 }
 219
 220                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:note/text()');
 221                                 if ($value != "") {
 222                                         $author["author-about"] = $value;
 223                                 }
 224
 225                                 $avatar = XML::getFirstAttributes($xpath, "atom:author/atom:link[@rel='avatar']");
 226                                 if (is_object($avatar)) {
 227                                         foreach ($avatar AS $attribute) {
 228                                                 if ($attribute->name == "href") {
 229                                                         $author["author-avatar"] = $attribute->textContent;
 230                                                 }
 231                                         }
 232                                 }
 233                         }
 234
 235                         $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:updated/text()');
 236
 237                         $author["app"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:generator/text()');
 238
 239                         $entries = $xpath->query('/atom:feed/atom:entry');
 240                 }
 241
 242                 // Is it RSS?
 243                 if ($xpath->query('/rss/channel')->length > 0) {
 244                         $author["author-link"] = XML::getFirstNodeValue($xpath, '/rss/channel/link/text()');
 245
 246                         $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/title/text()');
 247                         $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/rss/channel/image/url/text()');
 248
 249                         if (empty($author["author-name"])) {
 250                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/copyright/text()');
 251                         }
 252
 253                         if (empty($author["author-name"])) {
 254                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/description/text()');
 255                         }
 256
 257                         $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/rss/channel/pubDate/text()');
 258
 259                         $author["app"] = XML::getFirstNodeValue($xpath, '/rss/channel/generator/text()');
 260
 261                         $entries = $xpath->query('/rss/channel/item');
 262                 }
 263
 264                 if (!$dryRun) {
 265                         $author["author-link"] = $contact["url"];
 266
 267                         if (empty($author["author-name"])) {
 268                                 $author["author-name"] = $contact["name"];
 269                         }
 270
 271                         $author["author-avatar"] = $contact["thumb"];
 272
 273                         $author["owner-link"] = $contact["url"];
 274                         $author["owner-name"] = $contact["name"];
 275                         $author["owner-avatar"] = $contact["thumb"];
 276                 }
 277
 278                 $header = [];
 279                 $header["uid"] = $importer["uid"] ?? 0;
 280                 $header["network"] = Protocol::FEED;
 281                 $header["wall"] = 0;
 282                 $header["origin"] = 0;
 283                 $header["gravity"] = GRAVITY_PARENT;
 284                 $header["private"] = Item::PUBLIC;
 285                 $header["verb"] = Activity::POST;
 286                 $header["object-type"] = Activity\ObjectType::NOTE;
 287
 288                 $header["contact-id"] = $contact["id"] ?? 0;
 289
 290                 if (!is_object($entries)) {
 291                         Logger::info("There are no entries in this feed.");
 292                         return [];
 293                 }
 294
 295                 $items = [];
 296
 297                 // Limit the number of items that are about to be fetched
 298                 $total_items = ($entries->length - 1);
 299                 $max_items = DI::config()->get('system', 'max_feed_items');
 300                 if (($max_items > 0) && ($total_items > $max_items)) {
 301                         $total_items = $max_items;
 302                 }
 303
 304                 // Importing older entries first
 305                 for ($i = $total_items; $i >= 0; --$i) {
 306                         $entry = $entries->item($i);
 307
 308                         $item = array_merge($header, $author);
 309
 310                         $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']", $entry);
 311                         if (!is_object($alternate)) {
 312                                 $alternate = XML::getFirstAttributes($xpath, "atom:link", $entry);
 313                         }
 314                         if (is_object($alternate)) {
 315                                 foreach ($alternate AS $attribute) {
 316                                         if ($attribute->name == "href") {
 317                                                 $item["plink"] = $attribute->textContent;
 318                                         }
 319                                 }
 320                         }
 321
 322                         if (empty($item["plink"])) {
 323                                 $item["plink"] = XML::getFirstNodeValue($xpath, 'link/text()', $entry);
 324                         }
 325
 326                         if (empty($item["plink"])) {
 327                                 $item["plink"] = XML::getFirstNodeValue($xpath, 'rss:link/text()', $entry);
 328                         }
 329
 330                         $item["uri"] = XML::getFirstNodeValue($xpath, 'atom:id/text()', $entry);
 331
 332                         if (empty($item["uri"])) {
 333                                 $item["uri"] = XML::getFirstNodeValue($xpath, 'guid/text()', $entry);
 334                         }
 335
 336                         if (empty($item["uri"])) {
 337                                 $item["uri"] = $item["plink"];
 338                         }
 339
 340                         $orig_plink = $item["plink"];
 341
 342                         $item["plink"] = Network::finalUrl($item["plink"]);
 343
 344                         $item["parent-uri"] = $item["uri"];
 345
 346                         if (!$dryRun) {
 347                                 $condition = ["`uid` = ? AND `uri` = ? AND `network` IN (?, ?)",
 348                                         $importer["uid"], $item["uri"], Protocol::FEED, Protocol::DFRN];
 349                                 $previous = Item::selectFirst(['id'], $condition);
 350                                 if (DBA::isResult($previous)) {
 351                                         Logger::info("Item with uri " . $item["uri"] . " for user " . $importer["uid"] . " already existed under id " . $previous["id"]);
 352                                         continue;
 353                                 }
 354                         }
 355
 356                         $item["title"] = XML::getFirstNodeValue($xpath, 'atom:title/text()', $entry);
 357
 358                         if (empty($item["title"])) {
 359                                 $item["title"] = XML::getFirstNodeValue($xpath, 'title/text()', $entry);
 360                         }
 361                         if (empty($item["title"])) {
 362                                 $item["title"] = XML::getFirstNodeValue($xpath, 'rss:title/text()', $entry);
 363                         }
 364
 365                         $item["title"] = html_entity_decode($item["title"], ENT_QUOTES, 'UTF-8');
 366
 367                         $published = XML::getFirstNodeValue($xpath, 'atom:published/text()', $entry);
 368
 369                         if (empty($published)) {
 370                                 $published = XML::getFirstNodeValue($xpath, 'pubDate/text()', $entry);
 371                         }
 372
 373                         if (empty($published)) {
 374                                 $published = XML::getFirstNodeValue($xpath, 'dc:date/text()', $entry);
 375                         }
 376
 377                         $updated = XML::getFirstNodeValue($xpath, 'atom:updated/text()', $entry);
 378
 379                         if (empty($updated) && !empty($published)) {
 380                                 $updated = $published;
 381                         }
 382
 383                         if (empty($published) && !empty($updated)) {
 384                                 $published = $updated;
 385                         }
 386
 387                         if ($published != "") {
 388                                 $item["created"] = $published;
 389                         }
 390
 391                         if ($updated != "") {
 392                                 $item["edited"] = $updated;
 393                         }
 394
 395                         $creator = XML::getFirstNodeValue($xpath, 'author/text()', $entry);
 396
 397                         if (empty($creator)) {
 398                                 $creator = XML::getFirstNodeValue($xpath, 'atom:author/atom:name/text()', $entry);
 399                         }
 400
 401                         if (empty($creator)) {
 402                                 $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
 403                         }
 404
 405                         if ($creator != "") {
 406                                 $item["author-name"] = $creator;
 407                         }
 408
 409                         $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
 410
 411                         if ($creator != "") {
 412                                 $item["author-name"] = $creator;
 413                         }
 414
 415                         /// @TODO ?
 416                         // <category>Ausland</category>
 417                         // <media:thumbnail width="152" height="76" url="http://www.taz.de/picture/667875/192/14388767.jpg"/>
 418
 419                         $attachments = [];
 420
 421                         $enclosures = $xpath->query("enclosure|atom:link[@rel='enclosure']", $entry);
 422                         foreach ($enclosures AS $enclosure) {
 423                                 $href = "";
 424                                 $length = "";
 425                                 $type = "";
 426
 427                                 foreach ($enclosure->attributes AS $attribute) {
 428                                         if (in_array($attribute->name, ["url", "href"])) {
 429                                                 $href = $attribute->textContent;
 430                                         } elseif ($attribute->name == "length") {
 431                                                 $length = $attribute->textContent;
 432                                         } elseif ($attribute->name == "type") {
 433                                                 $type = $attribute->textContent;
 434                                         }
 435                                 }
 436
 437                                 if (!empty($item["attach"])) {
 438                                         $item["attach"] .= ',';
 439                                 } else {
 440                                         $item["attach"] = '';
 441                                 }
 442
 443                                 $attachments[] = ["link" => $href, "type" => $type, "length" => $length];
 444
 445                                 $item["attach"] .= '[attach]href="' . $href . '" length="' . $length . '" type="' . $type . '"[/attach]';
 446                         }
 447
 448                         $taglist = [];
 449                         $categories = $xpath->query("category", $entry);
 450                         foreach ($categories AS $category) {
 451                                 $taglist[] = $category->nodeValue;
 452                         }
 453
 454                         $body = trim(XML::getFirstNodeValue($xpath, 'atom:content/text()', $entry));
 455
 456                         if (empty($body)) {
 457                                 $body = trim(XML::getFirstNodeValue($xpath, 'content:encoded/text()', $entry));
 458                         }
 459
 460                         $summary = trim(XML::getFirstNodeValue($xpath, 'atom:summary/text()', $entry));
 461
 462                         if (empty($summary)) {
 463                                 $summary = trim(XML::getFirstNodeValue($xpath, 'description/text()', $entry));
 464                         }
 465
 466                         if (empty($body)) {
 467                                 $body = $summary;
 468                                 $summary = '';
 469                         }
 470
 471                         if ($body == $summary) {
 472                                 $summary = '';
 473                         }
 474
 475                         // remove the content of the title if it is identically to the body
 476                         // This helps with auto generated titles e.g. from tumblr
 477                         if (self::titleIsBody($item["title"], $body)) {
 478                                 $item["title"] = "";
 479                         }
 480                         $item["body"] = HTML::toBBCode($body, $basepath);
 481
 482                         if (($item["body"] == '') && ($item["title"] != '')) {
 483                                 $item["body"] = $item["title"];
 484                                 $item["title"] = '';
 485                         }
 486
 487                         $preview = '';
 488                         if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] < 3)) {
 489                                 // Handle enclosures and treat them as preview picture
 490                                 foreach ($attachments AS $attachment) {
 491                                         if ($attachment["type"] == "image/jpeg") {
 492                                                 $preview = $attachment["link"];
 493                                         }
 494                                 }
 495
 496                                 // Remove a possible link to the item itself
 497                                 $item["body"] = str_replace($item["plink"], '', $item["body"]);
 498                                 $item["body"] = trim(preg_replace('/\[url\=\](\w+.*?)\[\/url\]/i', '', $item["body"]));
 499
 500                                 // Replace the content when the title is longer than the body
 501                                 $replace = (strlen($item["title"]) > strlen($item["body"]));
 502
 503                                 // Replace it, when there is an image in the body
 504                                 if (strstr($item["body"], '[/img]')) {
 505                                         $replace = true;
 506                                 }
 507
 508                                 // Replace it, when there is a link in the body
 509                                 if (strstr($item["body"], '[/url]')) {
 510                                         $replace = true;
 511                                 }
 512
 513                                 if ($replace) {
 514                                         $item["body"] = trim($item["title"]);
 515                                 }
 516
 517                                 $data = ParseUrl::getSiteinfoCached($item['plink'], true);
 518                                 if (!empty($data['text']) && !empty($data['title']) && (mb_strlen($item['body']) < mb_strlen($data['text']))) {
 519                                         // When the fetched page info text is longer than the body, we do try to enhance the body
 520                                         if (!empty($item['body']) && (strpos($data['title'], $item['body']) === false) && (strpos($data['text'], $item['body']) === false)) {
 521                                                 // The body is not part of the fetched page info title or page info text. So we add the text to the body
 522                                                 $item['body'] .= "\n\n" . $data['text'];
 523                                         } else {
 524                                                 // Else we replace the body with the page info text
 525                                                 $item['body'] = $data['text'];
 526                                         }
 527                                 }
 528
 529                                 // We always strip the title since it will be added in the page information
 530                                 $item["title"] = "";
 531                                 $item["body"] = $item["body"] . add_page_info($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_denylist"] ?? '');
 532                                 $taglist = get_page_keywords($item["plink"], $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_denylist"]);
 533                                 $item["object-type"] = Activity\ObjectType::BOOKMARK;
 534                                 unset($item["attach"]);
 535                         } else {
 536                                 if (!empty($summary)) {
 537                                         $item["body"] = '[abstract]' . HTML::toBBCode($summary, $basepath) . "[/abstract]\n" . $item["body"];
 538                                 }
 539
 540                                 if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] == 3)) {
 541                                         if (empty($taglist)) {
 542                                                 $taglist = get_page_keywords($item["plink"], $preview, true, $contact["ffi_keyword_denylist"]);
 543                                         }
 544                                         $item["body"] .= "\n" . self::tagToString($taglist);
 545                                 } else {
 546                                         $taglist = [];
 547                                 }
 548
 549                                 // Add the link to the original feed entry if not present in feed
 550                                 if (($item['plink'] != '') && !strstr($item["body"], $item['plink'])) {
 551                                         $item["body"] .= "[hr][url]" . $item['plink'] . "[/url]";
 552                                 }
 553                         }
 554
 555                         if ($dryRun) {
 556                                 $items[] = $item;
 557                                 break;
 558                         } else {
 559                                 Logger::info('Stored feed', ['item' => $item]);
 560
 561                                 $notify = Item::isRemoteSelf($contact, $item);
 562
 563                                 // Distributed items should have a well formatted URI.
 564                                 // Additionally we have to avoid conflicts with identical URI between imported feeds and these items.
 565                                 if ($notify) {
 566                                         $item['guid'] = Item::guidFromUri($orig_plink, DI::baseUrl()->getHostname());
 567                                         unset($item['uri']);
 568                                         unset($item['parent-uri']);
 569
 570                                         // Set the delivery priority for "remote self" to "medium"
 571                                         $notify = PRIORITY_MEDIUM;
 572                                 }
 573
 574                                 $id = Item::insert($item, $notify);
 575
 576                                 Logger::info("Feed for contact " . $contact["url"] . " stored under id " . $id);
 577
 578                                 if (!empty($id) && !empty($taglist)) {
 579                                         $feeditem = Item::selectFirst(['uri-id'], ['id' => $id]);
 580                                         foreach ($taglist as $tag) {
 581                                                 Tag::store($feeditem['uri-id'], Tag::HASHTAG, $tag);
 582                                         }
 583                                 }
 584                         }
 585                 }
 586
 587                 return ["header" => $author, "items" => $items];
 588         }
 589
 590         /**
 591          * Convert a tag array to a tag string
 592          *
 593          * @param array $tags
 594          * @return string tag string
 595          */
 596         private static function tagToString(array $tags)
 597         {
 598                 $tagstr = '';
 599
 600                 foreach ($tags as $tag) {
 601                         if ($tagstr != "") {
 602                                 $tagstr .= ", ";
 603                         }
 604
 605                         $tagstr .= "#[url=" . DI::baseUrl() . "/search?tag=" . urlencode($tag) . "]" . $tag . "[/url]";
 606                 }
 607
 608                 return $tagstr;
 609         }
 610
 611         private static function titleIsBody($title, $body)
 612         {
 613                 $title = strip_tags($title);
 614                 $title = trim($title);
 615                 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
 616                 $title = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $title);
 617
 618                 $body = strip_tags($body);
 619                 $body = trim($body);
 620                 $body = html_entity_decode($body, ENT_QUOTES, 'UTF-8');
 621                 $body = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $body);
 622
 623                 if (strlen($title) < strlen($body)) {
 624                         $body = substr($body, 0, strlen($title));
 625                 }
 626
 627                 if (($title != $body) && (substr($title, -3) == "...")) {
 628                         $pos = strrpos($title, "...");
 629                         if ($pos > 0) {
 630                                 $title = substr($title, 0, $pos);
 631                                 $body = substr($body, 0, $pos);
 632                         }
 633                 }
 634                 return ($title == $body);
 635         }
 636 }