src/Protocol/Feed.php

   1 <?php
   2 /**
   3  * @file src/Protocol/Feed.php
   4  * @brief Imports RSS/RDF/Atom feeds
   5  *
   6  */
   7 namespace Friendica\Protocol;
   8
   9 use DOMDocument;
  10 use DOMXPath;
  11 use Friendica\Content\Text\HTML;
  12 use Friendica\Core\Logger;
  13 use Friendica\Core\Protocol;
  14 use Friendica\Core\System;
  15 use Friendica\Database\DBA;
  16 use Friendica\Model\Item;
  17 use Friendica\Util\Network;
  18 use Friendica\Util\XML;
  19
  20 /**
  21  * @brief This class contain functions to import feeds
  22  *
  23  */
  24 class Feed {
  25         /**
  26          * @brief Read a RSS/RDF/Atom feed and create an item entry for it
  27          *
  28          * @param string $xml The feed data
  29          * @param array $importer The user record of the importer
  30          * @param array $contact The contact record of the feed
  31          * @param string $hub Unused dummy value for compatibility reasons
  32          * @param bool $simulate If enabled, no data is imported
  33          *
  34          * @return array In simulation mode it returns the header and the first item
  35          */
  36         public static function import($xml, $importer, &$contact, &$hub, $simulate = false) {
  37
  38                 $a = get_app();
  39
  40                 if (!$simulate) {
  41                         Logger::log("Import Atom/RSS feed '".$contact["name"]."' (Contact ".$contact["id"].") for user ".$importer["uid"], Logger::DEBUG);
  42                 } else {
  43                         Logger::log("Test Atom/RSS feed", Logger::DEBUG);
  44                 }
  45                 if (empty($xml)) {
  46                         Logger::log('XML is empty.', Logger::DEBUG);
  47                         return;
  48                 }
  49
  50                 if (!empty($contact['poll'])) {
  51                         $basepath = $contact['poll'];
  52                 } elseif (!empty($contact['url'])) {
  53                         $basepath = $contact['url'];
  54                 } else {
  55                         $basepath = '';
  56                 }
  57
  58                 $doc = new DOMDocument();
  59                 @$doc->loadXML(trim($xml));
  60                 $xpath = new DOMXPath($doc);
  61                 $xpath->registerNamespace('atom', NAMESPACE_ATOM1);
  62                 $xpath->registerNamespace('dc', "http://purl.org/dc/elements/1.1/");
  63                 $xpath->registerNamespace('content', "http://purl.org/rss/1.0/modules/content/");
  64                 $xpath->registerNamespace('rdf', "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
  65                 $xpath->registerNamespace('rss', "http://purl.org/rss/1.0/");
  66                 $xpath->registerNamespace('media', "http://search.yahoo.com/mrss/");
  67                 $xpath->registerNamespace('poco', NAMESPACE_POCO);
  68
  69                 $author = [];
  70                 $entries = null;
  71
  72                 // Is it RDF?
  73                 if ($xpath->query('/rdf:RDF/rss:channel')->length > 0) {
  74                         $author["author-link"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:link/text()');
  75                         $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:title/text()');
  76
  77                         if (empty($author["author-name"])) {
  78                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:description/text()');
  79                         }
  80                         $entries = $xpath->query('/rdf:RDF/rss:item');
  81                 }
  82
  83                 // Is it Atom?
  84                 if ($xpath->query('/atom:feed')->length > 0) {
  85                         $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']");
  86                         if (is_object($alternate)) {
  87                                 foreach ($alternate AS $attribute) {
  88                                         if ($attribute->name == "href") {
  89                                                 $author["author-link"] = $attribute->textContent;
  90                                         }
  91                                 }
  92                         }
  93
  94                         if (empty($author["author-link"])) {
  95                                 $self = XML::getFirstAttributes($xpath, "atom:link[@rel='self']");
  96                                 if (is_object($self)) {
  97                                         foreach ($self AS $attribute) {
  98                                                 if ($attribute->name == "href") {
  99                                                         $author["author-link"] = $attribute->textContent;
 100                                                 }
 101                                         }
 102                                 }
 103                         }
 104
 105                         if (empty($author["author-link"])) {
 106                                 $author["author-link"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:id/text()');
 107                         }
 108                         $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:logo/text()');
 109
 110                         $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:title/text()');
 111
 112                         if (empty($author["author-name"])) {
 113                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:subtitle/text()');
 114                         }
 115                         if (empty($author["author-name"])) {
 116                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:name/text()');
 117                         }
 118                         $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:displayName/text()');
 119                         if ($value != "") {
 120                                 $author["author-name"] = $value;
 121                         }
 122                         if ($simulate) {
 123                                 $author["author-id"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:id/text()');
 124
 125                                 // See https://tools.ietf.org/html/rfc4287#section-3.2.2
 126                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/atom:uri/text()');
 127                                 if ($value != "") {
 128                                         $author["author-link"] = $value;
 129                                 }
 130
 131                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:preferredUsername/text()');
 132                                 if ($value != "") {
 133                                         $author["author-nick"] = $value;
 134                                 }
 135                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:address/poco:formatted/text()');
 136                                 if ($value != "") {
 137                                         $author["author-location"] = $value;
 138                                 }
 139                                 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:note/text()');
 140                                 if ($value != "") {
 141                                         $author["author-about"] = $value;
 142                                 }
 143                                 $avatar = XML::getFirstAttributes($xpath, "atom:author/atom:link[@rel='avatar']");
 144                                 if (is_object($avatar)) {
 145                                         foreach ($avatar AS $attribute) {
 146                                                 if ($attribute->name == "href") {
 147                                                         $author["author-avatar"] = $attribute->textContent;
 148                                                 }
 149                                         }
 150                                 }
 151                         }
 152
 153                         $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:updated/text()');
 154
 155                         $author["app"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:generator/text()');
 156
 157                         $entries = $xpath->query('/atom:feed/atom:entry');
 158                 }
 159
 160                 // Is it RSS?
 161                 if ($xpath->query('/rss/channel')->length > 0) {
 162                         $author["author-link"] = XML::getFirstNodeValue($xpath, '/rss/channel/link/text()');
 163
 164                         $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/title/text()');
 165                         $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/rss/channel/image/url/text()');
 166
 167                         if (empty($author["author-name"])) {
 168                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/copyright/text()');
 169                         }
 170                         if (empty($author["author-name"])) {
 171                                 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/description/text()');
 172                         }
 173                         $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/rss/channel/pubDate/text()');
 174
 175                         $author["app"] = XML::getFirstNodeValue($xpath, '/rss/channel/generator/text()');
 176
 177                         $entries = $xpath->query('/rss/channel/item');
 178                 }
 179
 180                 if (!$simulate) {
 181                         $author["author-link"] = $contact["url"];
 182
 183                         if (empty($author["author-name"])) {
 184                                 $author["author-name"] = $contact["name"];
 185                         }
 186                         $author["author-avatar"] = $contact["thumb"];
 187
 188                         $author["owner-link"] = $contact["url"];
 189                         $author["owner-name"] = $contact["name"];
 190                         $author["owner-avatar"] = $contact["thumb"];
 191                 }
 192
 193                 $header = [];
 194                 $header["uid"] = $importer["uid"];
 195                 $header["network"] = Protocol::FEED;
 196                 $header["wall"] = 0;
 197                 $header["origin"] = 0;
 198                 $header["gravity"] = GRAVITY_PARENT;
 199                 $header["private"] = 2;
 200                 $header["verb"] = ACTIVITY_POST;
 201                 $header["object-type"] = ACTIVITY_OBJ_NOTE;
 202
 203                 $header["contact-id"] = $contact["id"];
 204
 205                 if (!is_object($entries)) {
 206                         Logger::log("There are no entries in this feed.", Logger::DEBUG);
 207                         return;
 208                 }
 209
 210                 $items = [];
 211                 // Importing older entries first
 212                 for($i = $entries->length - 1; $i >= 0;--$i) {
 213                         $entry = $entries->item($i);
 214
 215                         $item = array_merge($header, $author);
 216
 217                         $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']", $entry);
 218                         if (!is_object($alternate)) {
 219                                 $alternate = XML::getFirstAttributes($xpath, "atom:link", $entry);
 220                         }
 221                         if (is_object($alternate)) {
 222                                 foreach ($alternate AS $attribute) {
 223                                         if ($attribute->name == "href") {
 224                                                 $item["plink"] = $attribute->textContent;
 225                                         }
 226                                 }
 227                         }
 228                         if (empty($item["plink"])) {
 229                                 $item["plink"] = XML::getFirstNodeValue($xpath, 'link/text()', $entry);
 230                         }
 231                         if (empty($item["plink"])) {
 232                                 $item["plink"] = XML::getFirstNodeValue($xpath, 'rss:link/text()', $entry);
 233                         }
 234
 235                         $item["uri"] = XML::getFirstNodeValue($xpath, 'atom:id/text()', $entry);
 236
 237                         if (empty($item["uri"])) {
 238                                 $item["uri"] = XML::getFirstNodeValue($xpath, 'guid/text()', $entry);
 239                         }
 240                         if (empty($item["uri"])) {
 241                                 $item["uri"] = $item["plink"];
 242                         }
 243
 244                         $orig_plink = $item["plink"];
 245
 246                         $item["plink"] = Network::finalUrl($item["plink"]);
 247
 248                         $item["parent-uri"] = $item["uri"];
 249
 250                         if (!$simulate) {
 251                                 $condition = ["`uid` = ? AND `uri` = ? AND `network` IN (?, ?)",
 252                                         $importer["uid"], $item["uri"], Protocol::FEED, Protocol::DFRN];
 253                                 $previous = Item::selectFirst(['id'], $condition);
 254                                 if (DBA::isResult($previous)) {
 255                                         Logger::log("Item with uri ".$item["uri"]." for user ".$importer["uid"]." already existed under id ".$previous["id"], Logger::DEBUG);
 256                                         continue;
 257                                 }
 258                         }
 259
 260                         $item["title"] = XML::getFirstNodeValue($xpath, 'atom:title/text()', $entry);
 261
 262                         if (empty($item["title"])) {
 263                                 $item["title"] = XML::getFirstNodeValue($xpath, 'title/text()', $entry);
 264                         }
 265                         if (empty($item["title"])) {
 266                                 $item["title"] = XML::getFirstNodeValue($xpath, 'rss:title/text()', $entry);
 267                         }
 268                         $published = XML::getFirstNodeValue($xpath, 'atom:published/text()', $entry);
 269
 270                         if (empty($published)) {
 271                                 $published = XML::getFirstNodeValue($xpath, 'pubDate/text()', $entry);
 272                         }
 273                         if (empty($published)) {
 274                                 $published = XML::getFirstNodeValue($xpath, 'dc:date/text()', $entry);
 275                         }
 276                         $updated = XML::getFirstNodeValue($xpath, 'atom:updated/text()', $entry);
 277
 278                         if (empty($updated) && !empty($published)) {
 279                                 $updated = $published;
 280                         }
 281
 282                         if (empty($published) && !empty($updated)) {
 283                                 $published = $updated;
 284                         }
 285
 286                         if ($published != "") {
 287                                 $item["created"] = $published;
 288                         }
 289                         if ($updated != "") {
 290                                 $item["edited"] = $updated;
 291                         }
 292                         $creator = XML::getFirstNodeValue($xpath, 'author/text()', $entry);
 293
 294                         if (empty($creator)) {
 295                                 $creator = XML::getFirstNodeValue($xpath, 'atom:author/atom:name/text()', $entry);
 296                         }
 297                         if (empty($creator)) {
 298                                 $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
 299                         }
 300                         if ($creator != "") {
 301                                 $item["author-name"] = $creator;
 302                         }
 303                         $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
 304
 305                         if ($creator != "") {
 306                                 $item["author-name"] = $creator;
 307                         }
 308
 309                         /// @TODO ?
 310                         // <category>Ausland</category>
 311                         // <media:thumbnail width="152" height="76" url="http://www.taz.de/picture/667875/192/14388767.jpg"/>
 312
 313                         $attachments = [];
 314
 315                         $enclosures = $xpath->query("enclosure|atom:link[@rel='enclosure']", $entry);
 316                         foreach ($enclosures AS $enclosure) {
 317                                 $href = "";
 318                                 $length = "";
 319                                 $type = "";
 320                                 $title = "";
 321
 322                                 foreach ($enclosure->attributes AS $attribute) {
 323                                         if (in_array($attribute->name, ["url", "href"])) {
 324                                                 $href = $attribute->textContent;
 325                                         } elseif ($attribute->name == "length") {
 326                                                 $length = $attribute->textContent;
 327                                         } elseif ($attribute->name == "type") {
 328                                                 $type = $attribute->textContent;
 329                                         }
 330                                 }
 331                                 if (!empty($item["attach"])) {
 332                                         $item["attach"] .= ',';
 333                                 } else {
 334                                         $item["attach"] = '';
 335                                 }
 336
 337                                 $attachments[] = ["link" => $href, "type" => $type, "length" => $length];
 338
 339                                 $item["attach"] .= '[attach]href="'.$href.'" length="'.$length.'" type="'.$type.'"[/attach]';
 340                         }
 341
 342                         $tags = '';
 343                         $categories = $xpath->query("category", $entry);
 344                         foreach ($categories AS $category) {
 345                                 $hashtag = $category->nodeValue;
 346                                 if ($tags != '') {
 347                                         $tags .= ', ';
 348                                 }
 349
 350                                 $taglink = "#[url=" . System::baseUrl() . "/search?tag=" . $hashtag . "]" . $hashtag . "[/url]";
 351                                 $tags .= $taglink;
 352                         }
 353
 354                         $body = trim(XML::getFirstNodeValue($xpath, 'atom:content/text()', $entry));
 355
 356                         if (empty($body)) {
 357                                 $body = trim(XML::getFirstNodeValue($xpath, 'content:encoded/text()', $entry));
 358                         }
 359                         if (empty($body)) {
 360                                 $body = trim(XML::getFirstNodeValue($xpath, 'description/text()', $entry));
 361                         }
 362                         if (empty($body)) {
 363                                 $body = trim(XML::getFirstNodeValue($xpath, 'atom:summary/text()', $entry));
 364                         }
 365
 366                         // remove the content of the title if it is identically to the body
 367                         // This helps with auto generated titles e.g. from tumblr
 368                         if (self::titleIsBody($item["title"], $body)) {
 369                                 $item["title"] = "";
 370                         }
 371                         $item["body"] = HTML::toBBCode($body, $basepath);
 372
 373                         if (($item["body"] == '') && ($item["title"] != '')) {
 374                                 $item["body"] = $item["title"];
 375                                 $item["title"] = '';
 376                         }
 377
 378                         $preview = '';
 379                         if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] < 3)) {
 380                                 // Handle enclosures and treat them as preview picture
 381                                 foreach ($attachments AS $attachment) {
 382                                         if ($attachment["type"] == "image/jpeg") {
 383                                                 $preview = $attachment["link"];
 384                                         }
 385                                 }
 386
 387                                 // Remove a possible link to the item itself
 388                                 $item["body"] = str_replace($item["plink"], '', $item["body"]);
 389                                 $item["body"] = preg_replace('/\[url\=\](\w+.*?)\[\/url\]/i', '', $item["body"]);
 390
 391                                 // Replace the content when the title is longer than the body
 392                                 $replace = (strlen($item["title"]) > strlen($item["body"]));
 393
 394                                 // Replace it, when there is an image in the body
 395                                 if (strstr($item["body"], '[/img]')) {
 396                                         $replace = true;
 397                                 }
 398
 399                                 // Replace it, when there is a link in the body
 400                                 if (strstr($item["body"], '[/url]')) {
 401                                         $replace = true;
 402                                 }
 403
 404                                 if ($replace) {
 405                                         $item["body"] = $item["title"];
 406                                 }
 407                                 // We always strip the title since it will be added in the page information
 408                                 $item["title"] = "";
 409                                 $item["body"] = $item["body"].add_page_info($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]);
 410                                 $item["tag"] = add_page_keywords($item["plink"], $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]);
 411                                 $item["object-type"] = ACTIVITY_OBJ_BOOKMARK;
 412                                 unset($item["attach"]);
 413                         } else {
 414                                 if ($contact["fetch_further_information"] == 3) {
 415                                         if (!empty($tags)) {
 416                                                 $item["tag"] = $tags;
 417                                         } else {
 418                                                 // @todo $preview is never set in this case, is it intended? - @MrPetovan 2018-02-13
 419                                                 $item["tag"] = add_page_keywords($item["plink"], $preview, true, $contact["ffi_keyword_blacklist"]);
 420                                         }
 421                                         $item["body"] .= "\n".$item['tag'];
 422                                 }
 423                                 // Add the link to the original feed entry if not present in feed
 424                                 if (($item['plink'] != '') && !strstr($item["body"], $item['plink'])) {
 425                                         $item["body"] .= "[hr][url]".$item['plink']."[/url]";
 426                                 }
 427                         }
 428
 429                         if (!$simulate) {
 430                                 Logger::log("Stored feed: ".print_r($item, true), Logger::DEBUG);
 431
 432                                 $notify = Item::isRemoteSelf($contact, $item);
 433
 434                                 // Distributed items should have a well formatted URI.
 435                                 // Additionally we have to avoid conflicts with identical URI between imported feeds and these items.
 436                                 if ($notify) {
 437                                         $item['guid'] = Item::guidFromUri($orig_plink, $a->getHostName());
 438                                         unset($item['uri']);
 439                                         unset($item['parent-uri']);
 440
 441                                         // Set the delivery priority for "remote self" to "medium"
 442                                         $notify = PRIORITY_MEDIUM;
 443                                 }
 444
 445                                 $id = Item::insert($item, false, $notify);
 446
 447                                 Logger::log("Feed for contact ".$contact["url"]." stored under id ".$id);
 448                         } else {
 449                                 $items[] = $item;
 450                         }
 451                         if ($simulate) {
 452                                 break;
 453                         }
 454                 }
 455
 456                 if ($simulate) {
 457                         return ["header" => $author, "items" => $items];
 458                 }
 459         }
 460
 461         private static function titleIsBody($title, $body)
 462         {
 463                 $title = strip_tags($title);
 464                 $title = trim($title);
 465                 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
 466                 $title = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $title);
 467
 468                 $body = strip_tags($body);
 469                 $body = trim($body);
 470                 $body = html_entity_decode($body, ENT_QUOTES, 'UTF-8');
 471                 $body = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $body);
 472
 473                 if (strlen($title) < strlen($body)) {
 474                         $body = substr($body, 0, strlen($title));
 475                 }
 476
 477                 if (($title != $body) && (substr($title, -3) == "...")) {
 478                         $pos = strrpos($title, "...");
 479                         if ($pos > 0) {
 480                                 $title = substr($title, 0, $pos);
 481                                 $body = substr($body, 0, $pos);
 482                         }
 483                 }
 484                 return ($title == $body);
 485         }
 486 }