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\PageInfo;
  27 use Friendica\Content\Text\HTML;
  28 use Friendica\Core\Logger;
  29 use Friendica\Core\Protocol;
  30 use Friendica\Database\DBA;
  31 use Friendica\DI;
  32 use Friendica\Model\Item;
  33 use Friendica\Model\Tag;
  34 use Friendica\Util\Network;
  35 use Friendica\Util\ParseUrl;
  36 use Friendica\Util\XML;
  37
  38 /**
  39  * This class contain functions to import feeds (RSS/RDF/Atom)
  40  */
  41 class Feed
  42 {
  43         /**
  44          * consume - process atom feed and update anything/everything we might need to update
  45          *
  46          * $xml = the (atom) feed to consume - RSS isn't as fully supported but may work for simple feeds.
  47          *
  48          * $importer = the contact_record (joined to user_record) of the local user who owns this relationship.
  49          *             It is this person's stuff that is going to be updated.
  50          * $contact =  the person who is sending us stuff. If not set, we MAY be processing a "follow" activity
  51          *             from an external network and MAY create an appropriate contact record. Otherwise, we MUST
  52          *             have a contact record.
  53          * $hub = should we find a hub declation in the feed, pass it back to our calling process, who might (or
  54          *        might not) try and subscribe to it.
  55          * $datedir sorts in reverse order
  56          * $pass - by default ($pass = 0) we cannot guarantee that a parent item has been
  57          *      imported prior to its children being seen in the stream unless we are certain
  58          *      of how the feed is arranged/ordered.
  59          * With $pass = 1, we only pull parent items out of the stream.
  60          * With $pass = 2, we only pull children (comments/likes).
  61          *
  62          * So running this twice, first with pass 1 and then with pass 2 will do the right
  63          * thing regardless of feed ordering. This won't be adequate in a fully-threaded
  64          * model where comments can have sub-threads. That would require some massive sorting
  65          * to get all the feed items into a mostly linear ordering, and might still require
  66          * recursion.
  67          *
  68          * @param       $xml
  69          * @param array $importer
  70          * @param array $contact
  71          * @param       $hub
  72          * @throws ImagickException
  73          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
  74          */
  75         public static function consume($xml, array $importer, array $contact, &$hub)
  76         {
  77                 if ($contact['network'] === Protocol::OSTATUS) {
  78                         Logger::info('Consume OStatus messages');
  79                         OStatus::import($xml, $importer, $contact, $hub);
  80
  81                         return;
  82                 }
  83
  84                 if ($contact['network'] === Protocol::FEED) {
  85                         Logger::info('Consume feeds');
  86                         self::import($xml, $importer, $contact);
  87
  88                         return;
  89                 }
  90
  91                 if ($contact['network'] === Protocol::DFRN) {
  92                         Logger::info('Consume DFRN messages');
  93                         $dfrn_importer = DFRN::getImporter($contact['id'], $importer['uid']);
  94                         if (!empty($dfrn_importer)) {
  95                                 Logger::info('Now import the DFRN feed');
  96                                 DFRN::import($xml, $dfrn_importer, true);
  97                                 return;
  98                         }
  99                 }
 100         }
 101
 102         /**
 103          * Read a RSS/RDF/Atom feed and create an item entry for it
 104          *
 105          * @param string $xml      The feed data
 106          * @param array  $importer The user record of the importer
 107          * @param array  $contact  The contact record of the feed
 108          *
 109          * @return array Returns the header and the first item in dry run mode
 110          * @throws \Friendica\Network\HTTPException\InternalServerErrorException
 111          */
 112         public static function import($xml, array $importer = [], array $contact = [])
 113         {
 114                 $dryRun = empty($importer) && empty($contact);
 115
 116                 if ($dryRun) {
 117                         Logger::info("Test Atom/RSS feed");
 118                 } else {
 119                         Logger::info("Import Atom/RSS feed '" . $contact["name"] . "' (Contact " . $contact["id"] . ") for user " . $importer["uid"]);
 120                 }
 121
 122                 if (empty($xml)) {
 123                         Logger::info('XML is empty.');
 124                         return [];
 125                 }
 126
 127                 if (!empty($contact['poll'])) {
 128                         $basepath = $contact['poll'];
 129                 } elseif (!empty($contact['url'])) {
 130                         $basepath = $contact['url'];
 131                 } else {
 132                         $basepath = '';
 133                 }
 134
 135                 $doc = new DOMDocument();
 136                 @$doc->loadXML(trim($xml));
 137                 $xpath = new DOMXPath($doc);
 138                 $xpath->registerNamespace('atom', ActivityNamespace::ATOM1);
 139                 $xpath->registerNamespace('dc', "http://purl.org/dc/elements/1.1/");
 140                 $xpath->registerNamespace('content', "http://purl.org/rss/1.0/modules/content/");
 141                 $xpath->registerNamespace('rdf', "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
 142                 $xpath->registerNamespace('rss', "http://purl.org/rss/1.0/");
 143                 $xpath->registerNamespace('media', "http://search.yahoo.com/mrss/");
 144                 $xpath->registerNamespace('poco', ActivityNamespace::POCO);
 145
 146                 $author = [];
 147                 $entries = null;
 148
 149                 // Is it RDF?
 150                 if ($xpath->query('/rdf:RDF/rss:channel')->length > 0) {
 151                         $author["author-link"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:link/text()');
 152                         $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:title/text()');
 153
 154                         if (empty($author["author-name"])) {
 155                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:description/text()');
 156                         }
 157                         $entries = $xpath->query('/rdf:RDF/rss:item');
 158                 }
 159
 160                 // Is it Atom?
 161                 if ($xpath->query('/atom:feed')->length > 0) {
 162                         $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']");
 163                         if (is_object($alternate)) {
 164                                 foreach ($alternate AS $attribute) {
 165                                         if ($attribute->name == "href") {
 166                                                 $author["author-link"] = $attribute->textContent;
 167                                         }
 168                                 }
 169                         }
 170
 171                         if (empty($author["author-link"])) {
 172                                 $self = XML::getFirstAttributes($xpath, "atom:link[@rel='self']");
 173                                 if (is_object($self)) {
 174                                         foreach ($self AS $attribute) {
 175                                                 if ($attribute->name == "href") {
 176                                                         $author["author-link"] = $attribute->textContent;
 177                                                 }
 178                                         }
 179                                 }
 180                         }
 181
 182                         if (empty($author["author-link"])) {
 183                                 $author["author-link"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:id/text()');
 184                         }
 185                         $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:logo/text()');
 186
 187                         $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:title/text()');
 188
 189                         if (empty($author["author-name"])) {
 190                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:subtitle/text()');
 191                         }
 192
 193                         if (empty($author["author-name"])) {
 194                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:name/text()');
 195                         }
 196
 197                         $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:displayName/text()');
 198                         if ($value != "") {
 199                                 $author["author-name"] = $value;
 200                         }
 201
 202                         if ($dryRun) {
 203                                 $author["author-id"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:id/text()');
 204
 205                                 // See https://tools.ietf.org/html/rfc4287#section-3.2.2
 206                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/atom:uri/text()');
 207                                 if ($value != "") {
 208                                         $author["author-link"] = $value;
 209                                 }
 210
 211                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:preferredUsername/text()');
 212                                 if ($value != "") {
 213                                         $author["author-nick"] = $value;
 214                                 }
 215
 216                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:address/poco:formatted/text()');
 217                                 if ($value != "") {
 218                                         $author["author-location"] = $value;
 219                                 }
 220
 221                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:note/text()');
 222                                 if ($value != "") {
 223                                         $author["author-about"] = $value;
 224                                 }
 225
 226                                 $avatar = XML::getFirstAttributes($xpath, "atom:author/atom:link[@rel='avatar']");
 227                                 if (is_object($avatar)) {
 228                                         foreach ($avatar AS $attribute) {
 229                                                 if ($attribute->name == "href") {
 230                                                         $author["author-avatar"] = $attribute->textContent;
 231                                                 }
 232                                         }
 233                                 }
 234                         }
 235
 236                         $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:updated/text()');
 237
 238                         $author["app"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:generator/text()');
 239
 240                         $entries = $xpath->query('/atom:feed/atom:entry');
 241                 }
 242
 243                 // Is it RSS?
 244                 if ($xpath->query('/rss/channel')->length > 0) {
 245                         $author["author-link"] = XML::getFirstNodeValue($xpath, '/rss/channel/link/text()');
 246
 247                         $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/title/text()');
 248                         $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/rss/channel/image/url/text()');
 249
 250                         if (empty($author["author-name"])) {
 251                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/copyright/text()');
 252                         }
 253
 254                         if (empty($author["author-name"])) {
 255                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/description/text()');
 256                         }
 257
 258                         $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/rss/channel/pubDate/text()');
 259
 260                         $author["app"] = XML::getFirstNodeValue($xpath, '/rss/channel/generator/text()');
 261
 262                         $entries = $xpath->query('/rss/channel/item');
 263                 }
 264
 265                 if (!$dryRun) {
 266                         $author["author-link"] = $contact["url"];
 267
 268                         if (empty($author["author-name"])) {
 269                                 $author["author-name"] = $contact["name"];
 270                         }
 271
 272                         $author["author-avatar"] = $contact["thumb"];
 273
 274                         $author["owner-link"] = $contact["url"];
 275                         $author["owner-name"] = $contact["name"];
 276                         $author["owner-avatar"] = $contact["thumb"];
 277                 }
 278
 279                 $header = [];
 280                 $header["uid"] = $importer["uid"] ?? 0;
 281                 $header["network"] = Protocol::FEED;
 282                 $header["wall"] = 0;
 283                 $header["origin"] = 0;
 284                 $header["gravity"] = GRAVITY_PARENT;
 285                 $header["private"] = Item::PUBLIC;
 286                 $header["verb"] = Activity::POST;
 287                 $header["object-type"] = Activity\ObjectType::NOTE;
 288
 289                 $header["contact-id"] = $contact["id"] ?? 0;
 290
 291                 if (!is_object($entries)) {
 292                         Logger::info("There are no entries in this feed.");
 293                         return [];
 294                 }
 295
 296                 $items = [];
 297
 298                 // Limit the number of items that are about to be fetched
 299                 $total_items = ($entries->length - 1);
 300                 $max_items = DI::config()->get('system', 'max_feed_items');
 301                 if (($max_items > 0) && ($total_items > $max_items)) {
 302                         $total_items = $max_items;
 303                 }
 304
 305                 // Importing older entries first
 306                 for ($i = $total_items; $i >= 0; --$i) {
 307                         $entry = $entries->item($i);
 308
 309                         $item = array_merge($header, $author);
 310
 311                         $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']", $entry);
 312                         if (!is_object($alternate)) {
 313                                 $alternate = XML::getFirstAttributes($xpath, "atom:link", $entry);
 314                         }
 315                         if (is_object($alternate)) {
 316                                 foreach ($alternate AS $attribute) {
 317                                         if ($attribute->name == "href") {
 318                                                 $item["plink"] = $attribute->textContent;
 319                                         }
 320                                 }
 321                         }
 322
 323                         if (empty($item["plink"])) {
 324                                 $item["plink"] = XML::getFirstNodeValue($xpath, 'link/text()', $entry);
 325                         }
 326
 327                         if (empty($item["plink"])) {
 328                                 $item["plink"] = XML::getFirstNodeValue($xpath, 'rss:link/text()', $entry);
 329                         }
 330
 331                         $item["uri"] = XML::getFirstNodeValue($xpath, 'atom:id/text()', $entry);
 332
 333                         if (empty($item["uri"])) {
 334                                 $item["uri"] = XML::getFirstNodeValue($xpath, 'guid/text()', $entry);
 335                         }
 336
 337                         if (empty($item["uri"])) {
 338                                 $item["uri"] = $item["plink"];
 339                         }
 340
 341                         // Add the base path if missing
 342                         $item["uri"] = Network::addBasePath($item["uri"], $basepath);
 343                         $item["plink"] = Network::addBasePath($item["plink"], $basepath);
 344
 345                         $orig_plink = $item["plink"];
 346
 347                         $item["plink"] = Network::finalUrl($item["plink"]);
 348
 349                         $item["parent-uri"] = $item["uri"];
 350
 351                         if (!$dryRun) {
 352                                 $condition = ["`uid` = ? AND `uri` = ? AND `network` IN (?, ?)",
 353                                         $importer["uid"], $item["uri"], Protocol::FEED, Protocol::DFRN];
 354                                 $previous = Item::selectFirst(['id'], $condition);
 355                                 if (DBA::isResult($previous)) {
 356                                         Logger::info("Item with uri " . $item["uri"] . " for user " . $importer["uid"] . " already existed under id " . $previous["id"]);
 357                                         continue;
 358                                 }
 359                         }
 360
 361                         $item["title"] = XML::getFirstNodeValue($xpath, 'atom:title/text()', $entry);
 362
 363                         if (empty($item["title"])) {
 364                                 $item["title"] = XML::getFirstNodeValue($xpath, 'title/text()', $entry);
 365                         }
 366                         if (empty($item["title"])) {
 367                                 $item["title"] = XML::getFirstNodeValue($xpath, 'rss:title/text()', $entry);
 368                         }
 369
 370                         $item["title"] = html_entity_decode($item["title"], ENT_QUOTES, 'UTF-8');
 371
 372                         $published = XML::getFirstNodeValue($xpath, 'atom:published/text()', $entry);
 373
 374                         if (empty($published)) {
 375                                 $published = XML::getFirstNodeValue($xpath, 'pubDate/text()', $entry);
 376                         }
 377
 378                         if (empty($published)) {
 379                                 $published = XML::getFirstNodeValue($xpath, 'dc:date/text()', $entry);
 380                         }
 381
 382                         $updated = XML::getFirstNodeValue($xpath, 'atom:updated/text()', $entry);
 383
 384                         if (empty($updated) && !empty($published)) {
 385                                 $updated = $published;
 386                         }
 387
 388                         if (empty($published) && !empty($updated)) {
 389                                 $published = $updated;
 390                         }
 391
 392                         if ($published != "") {
 393                                 $item["created"] = $published;
 394                         }
 395
 396                         if ($updated != "") {
 397                                 $item["edited"] = $updated;
 398                         }
 399
 400                         $creator = XML::getFirstNodeValue($xpath, 'author/text()', $entry);
 401
 402                         if (empty($creator)) {
 403                                 $creator = XML::getFirstNodeValue($xpath, 'atom:author/atom:name/text()', $entry);
 404                         }
 405
 406                         if (empty($creator)) {
 407                                 $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
 408                         }
 409
 410                         if ($creator != "") {
 411                                 $item["author-name"] = $creator;
 412                         }
 413
 414                         $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
 415
 416                         if ($creator != "") {
 417                                 $item["author-name"] = $creator;
 418                         }
 419
 420                         /// @TODO ?
 421                         // <category>Ausland</category>
 422                         // <media:thumbnail width="152" height="76" url="http://www.taz.de/picture/667875/192/14388767.jpg"/>
 423
 424                         $attachments = [];
 425
 426                         $enclosures = $xpath->query("enclosure|atom:link[@rel='enclosure']", $entry);
 427                         foreach ($enclosures AS $enclosure) {
 428                                 $href = "";
 429                                 $length = "";
 430                                 $type = "";
 431
 432                                 foreach ($enclosure->attributes AS $attribute) {
 433                                         if (in_array($attribute->name, ["url", "href"])) {
 434                                                 $href = $attribute->textContent;
 435                                         } elseif ($attribute->name == "length") {
 436                                                 $length = $attribute->textContent;
 437                                         } elseif ($attribute->name == "type") {
 438                                                 $type = $attribute->textContent;
 439                                         }
 440                                 }
 441
 442                                 if (!empty($item["attach"])) {
 443                                         $item["attach"] .= ',';
 444                                 } else {
 445                                         $item["attach"] = '';
 446                                 }
 447
 448                                 $attachments[] = ["link" => $href, "type" => $type, "length" => $length];
 449
 450                                 $item["attach"] .= '[attach]href="' . $href . '" length="' . $length . '" type="' . $type . '"[/attach]';
 451                         }
 452
 453                         $taglist = [];
 454                         $categories = $xpath->query("category", $entry);
 455                         foreach ($categories AS $category) {
 456                                 $taglist[] = $category->nodeValue;
 457                         }
 458
 459                         $body = trim(XML::getFirstNodeValue($xpath, 'atom:content/text()', $entry));
 460
 461                         if (empty($body)) {
 462                                 $body = trim(XML::getFirstNodeValue($xpath, 'content:encoded/text()', $entry));
 463                         }
 464
 465                         $summary = trim(XML::getFirstNodeValue($xpath, 'atom:summary/text()', $entry));
 466
 467                         if (empty($summary)) {
 468                                 $summary = trim(XML::getFirstNodeValue($xpath, 'description/text()', $entry));
 469                         }
 470
 471                         if (empty($body)) {
 472                                 $body = $summary;
 473                                 $summary = '';
 474                         }
 475
 476                         if ($body == $summary) {
 477                                 $summary = '';
 478                         }
 479
 480                         // remove the content of the title if it is identically to the body
 481                         // This helps with auto generated titles e.g. from tumblr
 482                         if (self::titleIsBody($item["title"], $body)) {
 483                                 $item["title"] = "";
 484                         }
 485                         $item["body"] = HTML::toBBCode($body, $basepath);
 486
 487                         if (($item["body"] == '') && ($item["title"] != '')) {
 488                                 $item["body"] = $item["title"];
 489                                 $item["title"] = '';
 490                         }
 491
 492                         $preview = '';
 493                         if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] < 3)) {
 494                                 // Handle enclosures and treat them as preview picture
 495                                 foreach ($attachments AS $attachment) {
 496                                         if ($attachment["type"] == "image/jpeg") {
 497                                                 $preview = $attachment["link"];
 498                                         }
 499                                 }
 500
 501                                 // Remove a possible link to the item itself
 502                                 $item["body"] = str_replace($item["plink"], '', $item["body"]);
 503                                 $item["body"] = trim(preg_replace('/\[url\=\](\w+.*?)\[\/url\]/i', '', $item["body"]));
 504
 505                                 // Replace the content when the title is longer than the body
 506                                 $replace = (strlen($item["title"]) > strlen($item["body"]));
 507
 508                                 // Replace it, when there is an image in the body
 509                                 if (strstr($item["body"], '[/img]')) {
 510                                         $replace = true;
 511                                 }
 512
 513                                 // Replace it, when there is a link in the body
 514                                 if (strstr($item["body"], '[/url]')) {
 515                                         $replace = true;
 516                                 }
 517
 518                                 if ($replace) {
 519                                         $item["body"] = trim($item["title"]);
 520                                 }
 521
 522                                 $data = ParseUrl::getSiteinfoCached($item['plink'], true);
 523                                 if (!empty($data['text']) && !empty($data['title']) && (mb_strlen($item['body']) < mb_strlen($data['text']))) {
 524                                         // When the fetched page info text is longer than the body, we do try to enhance the body
 525                                         if (!empty($item['body']) && (strpos($data['title'], $item['body']) === false) && (strpos($data['text'], $item['body']) === false)) {
 526                                                 // The body is not part of the fetched page info title or page info text. So we add the text to the body
 527                                                 $item['body'] .= "\n\n" . $data['text'];
 528                                         } else {
 529                                                 // Else we replace the body with the page info text
 530                                                 $item['body'] = $data['text'];
 531                                         }
 532                                 }
 533
 534                                 // We always strip the title since it will be added in the page information
 535                                 $item["title"] = "";
 536                                 $item["body"] = $item["body"] . "\n" . PageInfo::getFooterFromUrl($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_denylist"] ?? '');
 537                                 $taglist = $contact["fetch_further_information"] == 2 ? PageInfo::getTagsFromUrl($item["plink"], $preview, $contact["ffi_keyword_denylist"] ?? '') : [];
 538                                 $item["object-type"] = Activity\ObjectType::BOOKMARK;
 539                                 unset($item["attach"]);
 540                         } else {
 541                                 if (!empty($summary)) {
 542                                         $item["body"] = '[abstract]' . HTML::toBBCode($summary, $basepath) . "[/abstract]\n" . $item["body"];
 543                                 }
 544
 545                                 if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] == 3)) {
 546                                         if (empty($taglist)) {
 547                                                 $taglist = PageInfo::getTagsFromUrl($item["plink"], $preview, $contact["ffi_keyword_denylist"]);
 548                                         }
 549                                         $item["body"] .= "\n" . self::tagToString($taglist);
 550                                 } else {
 551                                         $taglist = [];
 552                                 }
 553
 554                                 // Add the link to the original feed entry if not present in feed
 555                                 if (($item['plink'] != '') && !strstr($item["body"], $item['plink'])) {
 556                                         $item["body"] .= "[hr][url]" . $item['plink'] . "[/url]";
 557                                 }
 558                         }
 559
 560                         if ($dryRun) {
 561                                 $items[] = $item;
 562                                 break;
 563                         } else {
 564                                 Logger::info('Stored feed', ['item' => $item]);
 565
 566                                 $notify = Item::isRemoteSelf($contact, $item);
 567
 568                                 // Distributed items should have a well formatted URI.
 569                                 // Additionally we have to avoid conflicts with identical URI between imported feeds and these items.
 570                                 if ($notify) {
 571                                         $item['guid'] = Item::guidFromUri($orig_plink, DI::baseUrl()->getHostname());
 572                                         unset($item['uri']);
 573                                         unset($item['parent-uri']);
 574
 575                                         // Set the delivery priority for "remote self" to "medium"
 576                                         $notify = PRIORITY_MEDIUM;
 577                                 }
 578
 579                                 $id = Item::insert($item, $notify);
 580
 581                                 Logger::info("Feed for contact " . $contact["url"] . " stored under id " . $id);
 582
 583                                 if (!empty($id) && !empty($taglist)) {
 584                                         $feeditem = Item::selectFirst(['uri-id'], ['id' => $id]);
 585                                         foreach ($taglist as $tag) {
 586                                                 Tag::store($feeditem['uri-id'], Tag::HASHTAG, $tag);
 587                                         }
 588                                 }
 589                         }
 590                 }
 591
 592                 return ["header" => $author, "items" => $items];
 593         }
 594
 595         /**
 596          * Convert a tag array to a tag string
 597          *
 598          * @param array $tags
 599          * @return string tag string
 600          */
 601         private static function tagToString(array $tags)
 602         {
 603                 $tagstr = '';
 604
 605                 foreach ($tags as $tag) {
 606                         if ($tagstr != "") {
 607                                 $tagstr .= ", ";
 608                         }
 609
 610                         $tagstr .= "#[url=" . DI::baseUrl() . "/search?tag=" . urlencode($tag) . "]" . $tag . "[/url]";
 611                 }
 612
 613                 return $tagstr;
 614         }
 615
 616         private static function titleIsBody($title, $body)
 617         {
 618                 $title = strip_tags($title);
 619                 $title = trim($title);
 620                 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
 621                 $title = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $title);
 622
 623                 $body = strip_tags($body);
 624                 $body = trim($body);
 625                 $body = html_entity_decode($body, ENT_QUOTES, 'UTF-8');
 626                 $body = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $body);
 627
 628                 if (strlen($title) < strlen($body)) {
 629                         $body = substr($body, 0, strlen($title));
 630                 }
 631
 632                 if (($title != $body) && (substr($title, -3) == "...")) {
 633                         $pos = strrpos($title, "...");
 634                         if ($pos > 0) {
 635                                 $title = substr($title, 0, $pos);
 636                                 $body = substr($body, 0, $pos);
 637                         }
 638                 }
 639                 return ($title == $body);
 640         }
 641 }