3 * @copyright Copyright (C) 2020, Friendica
5 * @license GNU AGPL version 3 or any later version
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.
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.
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/>.
22 namespace Friendica\Protocol;
26 use Friendica\Content\PageInfo;
27 use Friendica\Content\Text\BBCode;
28 use Friendica\Content\Text\HTML;
29 use Friendica\Core\Cache\Duration;
30 use Friendica\Core\Logger;
31 use Friendica\Core\Protocol;
32 use Friendica\Core\Worker;
33 use Friendica\Database\DBA;
35 use Friendica\Model\Contact;
36 use Friendica\Model\Conversation;
37 use Friendica\Model\Item;
38 use Friendica\Model\Post;
39 use Friendica\Model\Tag;
40 use Friendica\Model\User;
41 use Friendica\Util\DateTimeFormat;
42 use Friendica\Util\Network;
43 use Friendica\Util\ParseUrl;
44 use Friendica\Util\Strings;
45 use Friendica\Util\XML;
48 * This class contain functions to import feeds (RSS/RDF/Atom)
53 * Read a RSS/RDF/Atom feed and create an item entry for it
55 * @param string $xml The feed data
56 * @param array $importer The user record of the importer
57 * @param array $contact The contact record of the feed
59 * @return array Returns the header and the first item in dry run mode
60 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
62 public static function import($xml, array $importer = [], array $contact = [])
64 $dryRun = empty($importer) && empty($contact);
67 Logger::info("Test Atom/RSS feed");
69 Logger::info("Import Atom/RSS feed '" . $contact["name"] . "' (Contact " . $contact["id"] . ") for user " . $importer["uid"]);
73 Logger::info('XML is empty.');
77 if (!empty($contact['poll'])) {
78 $basepath = $contact['poll'];
79 } elseif (!empty($contact['url'])) {
80 $basepath = $contact['url'];
85 $doc = new DOMDocument();
86 @$doc->loadXML(trim($xml));
87 $xpath = new DOMXPath($doc);
88 $xpath->registerNamespace('atom', ActivityNamespace::ATOM1);
89 $xpath->registerNamespace('dc', "http://purl.org/dc/elements/1.1/");
90 $xpath->registerNamespace('content', "http://purl.org/rss/1.0/modules/content/");
91 $xpath->registerNamespace('rdf', "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
92 $xpath->registerNamespace('rss', "http://purl.org/rss/1.0/");
93 $xpath->registerNamespace('media', "http://search.yahoo.com/mrss/");
94 $xpath->registerNamespace('poco', ActivityNamespace::POCO);
100 if ($xpath->query('/rdf:RDF/rss:channel')->length > 0) {
101 $author["author-link"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:link/text()');
102 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:title/text()');
104 if (empty($author["author-name"])) {
105 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:description/text()');
107 $entries = $xpath->query('/rdf:RDF/rss:item');
111 if ($xpath->query('/atom:feed')->length > 0) {
112 $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']");
113 if (is_object($alternate)) {
114 foreach ($alternate AS $attribute) {
115 if ($attribute->name == "href") {
116 $author["author-link"] = $attribute->textContent;
121 if (empty($author["author-link"])) {
122 $self = XML::getFirstAttributes($xpath, "atom:link[@rel='self']");
123 if (is_object($self)) {
124 foreach ($self AS $attribute) {
125 if ($attribute->name == "href") {
126 $author["author-link"] = $attribute->textContent;
132 if (empty($author["author-link"])) {
133 $author["author-link"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:id/text()');
135 $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:logo/text()');
137 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:title/text()');
139 if (empty($author["author-name"])) {
140 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:subtitle/text()');
143 if (empty($author["author-name"])) {
144 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:name/text()');
147 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:displayName/text()');
149 $author["author-name"] = $value;
153 $author["author-id"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:id/text()');
155 // See https://tools.ietf.org/html/rfc4287#section-3.2.2
156 $value = XML::getFirstNodeValue($xpath, 'atom:author/atom:uri/text()');
158 $author["author-link"] = $value;
161 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:preferredUsername/text()');
163 $author["author-nick"] = $value;
166 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:address/poco:formatted/text()');
168 $author["author-location"] = $value;
171 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:note/text()');
173 $author["author-about"] = $value;
176 $avatar = XML::getFirstAttributes($xpath, "atom:author/atom:link[@rel='avatar']");
177 if (is_object($avatar)) {
178 foreach ($avatar AS $attribute) {
179 if ($attribute->name == "href") {
180 $author["author-avatar"] = $attribute->textContent;
186 $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:updated/text()');
188 $author["app"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:generator/text()');
190 $entries = $xpath->query('/atom:feed/atom:entry');
194 if ($xpath->query('/rss/channel')->length > 0) {
195 $author["author-link"] = XML::getFirstNodeValue($xpath, '/rss/channel/link/text()');
197 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/title/text()');
198 $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/rss/channel/image/url/text()');
200 if (empty($author["author-name"])) {
201 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/copyright/text()');
204 if (empty($author["author-name"])) {
205 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/description/text()');
208 $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/rss/channel/pubDate/text()');
210 $author["app"] = XML::getFirstNodeValue($xpath, '/rss/channel/generator/text()');
212 $entries = $xpath->query('/rss/channel/item');
216 $author["author-link"] = $contact["url"];
218 if (empty($author["author-name"])) {
219 $author["author-name"] = $contact["name"];
222 $author["author-avatar"] = $contact["thumb"];
224 $author["owner-link"] = $contact["url"];
225 $author["owner-name"] = $contact["name"];
226 $author["owner-avatar"] = $contact["thumb"];
230 $header["uid"] = $importer["uid"] ?? 0;
231 $header["network"] = Protocol::FEED;
233 $header["origin"] = 0;
234 $header["gravity"] = GRAVITY_PARENT;
235 $header["private"] = Item::PUBLIC;
236 $header["verb"] = Activity::POST;
237 $header["object-type"] = Activity\ObjectType::NOTE;
239 $header["contact-id"] = $contact["id"] ?? 0;
241 if (!is_object($entries)) {
242 Logger::info("There are no entries in this feed.");
247 $creation_dates = [];
249 // Limit the number of items that are about to be fetched
250 $total_items = ($entries->length - 1);
251 $max_items = DI::config()->get('system', 'max_feed_items');
252 if (($max_items > 0) && ($total_items > $max_items)) {
253 $total_items = $max_items;
258 // Importing older entries first
259 for ($i = $total_items; $i >= 0; --$i) {
260 $entry = $entries->item($i);
262 $item = array_merge($header, $author);
264 $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']", $entry);
265 if (!is_object($alternate)) {
266 $alternate = XML::getFirstAttributes($xpath, "atom:link", $entry);
268 if (is_object($alternate)) {
269 foreach ($alternate AS $attribute) {
270 if ($attribute->name == "href") {
271 $item["plink"] = $attribute->textContent;
276 if (empty($item["plink"])) {
277 $item["plink"] = XML::getFirstNodeValue($xpath, 'link/text()', $entry);
280 if (empty($item["plink"])) {
281 $item["plink"] = XML::getFirstNodeValue($xpath, 'rss:link/text()', $entry);
284 $item["uri"] = XML::getFirstNodeValue($xpath, 'atom:id/text()', $entry);
286 if (empty($item["uri"])) {
287 $item["uri"] = XML::getFirstNodeValue($xpath, 'guid/text()', $entry);
290 if (empty($item["uri"])) {
291 $item["uri"] = $item["plink"];
294 // Add the base path if missing
295 $item["uri"] = Network::addBasePath($item["uri"], $basepath);
296 $item["plink"] = Network::addBasePath($item["plink"], $basepath);
298 $orig_plink = $item["plink"];
300 $item["plink"] = DI::httpRequest()->finalUrl($item["plink"]);
302 $item["title"] = XML::getFirstNodeValue($xpath, 'atom:title/text()', $entry);
304 if (empty($item["title"])) {
305 $item["title"] = XML::getFirstNodeValue($xpath, 'title/text()', $entry);
307 if (empty($item["title"])) {
308 $item["title"] = XML::getFirstNodeValue($xpath, 'rss:title/text()', $entry);
311 $item["title"] = html_entity_decode($item["title"], ENT_QUOTES, 'UTF-8');
313 $published = XML::getFirstNodeValue($xpath, 'atom:published/text()', $entry);
315 if (empty($published)) {
316 $published = XML::getFirstNodeValue($xpath, 'pubDate/text()', $entry);
319 if (empty($published)) {
320 $published = XML::getFirstNodeValue($xpath, 'dc:date/text()', $entry);
323 $updated = XML::getFirstNodeValue($xpath, 'atom:updated/text()', $entry);
325 if (empty($updated) && !empty($published)) {
326 $updated = $published;
329 if (empty($published) && !empty($updated)) {
330 $published = $updated;
333 if ($published != "") {
334 $item["created"] = $published;
337 if ($updated != "") {
338 $item["edited"] = $updated;
342 $condition = ["`uid` = ? AND `uri` = ? AND `network` IN (?, ?)",
343 $importer["uid"], $item["uri"], Protocol::FEED, Protocol::DFRN];
344 $previous = Post::selectFirst(['id', 'created'], $condition);
345 if (DBA::isResult($previous)) {
346 // Use the creation date when the post had been stored. It can happen this date changes in the feed.
347 $creation_dates[] = $previous['created'];
348 Logger::info("Item with uri " . $item["uri"] . " for user " . $importer["uid"] . " already existed under id " . $previous["id"]);
351 $creation_dates[] = DateTimeFormat::utc($item['created']);
354 $creator = XML::getFirstNodeValue($xpath, 'author/text()', $entry);
356 if (empty($creator)) {
357 $creator = XML::getFirstNodeValue($xpath, 'atom:author/atom:name/text()', $entry);
360 if (empty($creator)) {
361 $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
364 if ($creator != "") {
365 $item["author-name"] = $creator;
368 $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
370 if ($creator != "") {
371 $item["author-name"] = $creator;
375 // <category>Ausland</category>
376 // <media:thumbnail width="152" height="76" url="http://www.taz.de/picture/667875/192/14388767.jpg"/>
380 $enclosures = $xpath->query("enclosure|atom:link[@rel='enclosure']", $entry);
381 foreach ($enclosures AS $enclosure) {
386 foreach ($enclosure->attributes AS $attribute) {
387 if (in_array($attribute->name, ["url", "href"])) {
388 $href = $attribute->textContent;
389 } elseif ($attribute->name == "length") {
390 $length = (int)$attribute->textContent;
391 } elseif ($attribute->name == "type") {
392 $type = $attribute->textContent;
397 $attachments[] = ['type' => Post\Media::DOCUMENT, 'url' => $href, 'mimetype' => $type, 'size' => $length];
402 $categories = $xpath->query("category", $entry);
403 foreach ($categories AS $category) {
404 $taglist[] = $category->nodeValue;
407 $body = trim(XML::getFirstNodeValue($xpath, 'atom:content/text()', $entry));
410 $body = trim(XML::getFirstNodeValue($xpath, 'content:encoded/text()', $entry));
413 $summary = trim(XML::getFirstNodeValue($xpath, 'atom:summary/text()', $entry));
415 if (empty($summary)) {
416 $summary = trim(XML::getFirstNodeValue($xpath, 'description/text()', $entry));
424 if ($body == $summary) {
428 // remove the content of the title if it is identically to the body
429 // This helps with auto generated titles e.g. from tumblr
430 if (self::titleIsBody($item["title"], $body)) {
433 $item["body"] = HTML::toBBCode($body, $basepath);
435 // Remove tracking pixels
436 $item["body"] = preg_replace("/\[img=1x1\]([^\[\]]*)\[\/img\]/Usi", '', $item["body"]);
438 if (($item["body"] == '') && ($item["title"] != '')) {
439 $item["body"] = $item["title"];
446 } elseif (!Item::isValid($item)) {
447 Logger::info('Feed item is invalid', ['created' => $item['created'], 'uid' => $item['uid'], 'uri' => $item['uri']]);
449 } elseif (Item::isTooOld($item)) {
450 Logger::info('Feed is too old', ['created' => $item['created'], 'uid' => $item['uid'], 'uri' => $item['uri']]);
455 if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] < 3)) {
456 // Handle enclosures and treat them as preview picture
457 foreach ($attachments AS $attachment) {
458 if ($attachment["mimetype"] == "image/jpeg") {
459 $preview = $attachment["url"];
463 // Remove a possible link to the item itself
464 $item["body"] = str_replace($item["plink"], '', $item["body"]);
465 $item["body"] = trim(preg_replace('/\[url\=\](\w+.*?)\[\/url\]/i', '', $item["body"]));
467 // Replace the content when the title is longer than the body
468 $replace = (strlen($item["title"]) > strlen($item["body"]));
470 // Replace it, when there is an image in the body
471 if (strstr($item["body"], '[/img]')) {
475 // Replace it, when there is a link in the body
476 if (strstr($item["body"], '[/url]')) {
480 $saved_body = $item["body"];
481 $saved_title = $item["title"];
484 $item["body"] = trim($item["title"]);
487 $data = ParseUrl::getSiteinfoCached($item['plink'], true);
488 if (!empty($data['text']) && !empty($data['title']) && (mb_strlen($item['body']) < mb_strlen($data['text']))) {
489 // When the fetched page info text is longer than the body, we do try to enhance the body
490 if (!empty($item['body']) && (strpos($data['title'], $item['body']) === false) && (strpos($data['text'], $item['body']) === false)) {
491 // The body is not part of the fetched page info title or page info text. So we add the text to the body
492 $item['body'] .= "\n\n" . $data['text'];
494 // Else we replace the body with the page info text
495 $item['body'] = $data['text'];
499 $data = PageInfo::queryUrl($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_denylist"] ?? '');
502 // Take the data that was provided by the feed if the query is empty
503 if (($data['type'] == 'link') && empty($data['title']) && empty($data['text'])) {
504 $data['title'] = $saved_title;
505 $item["body"] = $saved_body;
508 $data_text = strip_tags(trim($data['text'] ?? ''));
509 $item_body = strip_tags(trim($item['body'] ?? ''));
511 if (!empty($data_text) && (($data_text == $item_body) || strstr($item_body, $data_text))) {
515 // We always strip the title since it will be added in the page information
517 $item["body"] = $item["body"] . "\n" . PageInfo::getFooterFromData($data, false);
518 $taglist = $contact["fetch_further_information"] == 2 ? PageInfo::getTagsFromUrl($item["plink"], $preview, $contact["ffi_keyword_denylist"] ?? '') : [];
519 $item["object-type"] = Activity\ObjectType::BOOKMARK;
523 if (!empty($summary)) {
524 $item["body"] = '[abstract]' . HTML::toBBCode($summary, $basepath) . "[/abstract]\n" . $item["body"];
527 if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] == 3)) {
528 if (empty($taglist)) {
529 $taglist = PageInfo::getTagsFromUrl($item["plink"], $preview, $contact["ffi_keyword_denylist"] ?? '');
531 $item["body"] .= "\n" . self::tagToString($taglist);
536 // Add the link to the original feed entry if not present in feed
537 if (($item['plink'] != '') && !strstr($item["body"], $item['plink'])) {
538 $item["body"] .= "[hr][url]" . $item['plink'] . "[/url]";
542 Logger::info('Stored feed', ['item' => $item]);
544 $notify = Item::isRemoteSelf($contact, $item);
546 // Distributed items should have a well formatted URI.
547 // Additionally we have to avoid conflicts with identical URI between imported feeds and these items.
549 $item['guid'] = Item::guidFromUri($orig_plink, DI::baseUrl()->getHostname());
550 $item['uri'] = Item::newURI($item['uid'], $item['guid']);
551 unset($item['thr-parent']);
552 unset($item['parent-uri']);
554 // Set the delivery priority for "remote self" to "medium"
555 $notify = PRIORITY_MEDIUM;
558 $condition = ['uid' => $item['uid'], 'uri' => $item['uri']];
559 if (!Post::exists($condition) && !Post\Delayed::exists($item["uri"], $item['uid'])) {
561 Post\Delayed::publish($item, $notify, $taglist, $attachments);
563 $postings[] = ['item' => $item, 'notify' => $notify,
564 'taglist' => $taglist, 'attachments' => $attachments];
567 Logger::info('Post already created or exists in the delayed posts queue', ['uid' => $item['uid'], 'uri' => $item["uri"]]);
571 if (!empty($postings)) {
572 $min_posting = DI::config()->get('system', 'minimum_posting_interval', 0);
573 $total = count($postings);
575 // Posts shouldn't be delayed more than a day
576 $interval = min(1440, self::getPollInterval($contact));
577 $delay = max(round(($interval * 60) / $total), 60 * $min_posting);
578 Logger::info('Got posting delay', ['delay' => $delay, 'interval' => $interval, 'items' => $total, 'cid' => $contact['id'], 'url' => $contact['url']]);
585 foreach ($postings as $posting) {
587 $publish_time = time() + $post_delay;
588 $post_delay += $delay;
590 $publish_time = time();
593 $last_publish = DI::pConfig()->get($posting['item']['uid'], 'system', 'last_publish', 0, true);
594 $next_publish = max($last_publish + (60 * $min_posting), time());
595 if ($publish_time < $next_publish) {
596 $publish_time = $next_publish;
598 $publish_at = date(DateTimeFormat::MYSQL, $publish_time);
600 Post\Delayed::add($posting['item']['uri'], $posting['item'], $posting['notify'], false, $publish_at, $posting['taglist'], $posting['attachments']);
604 if (!$dryRun && DI::config()->get('system', 'adjust_poll_frequency')) {
605 self::adjustPollFrequency($contact, $creation_dates);
608 return ["header" => $author, "items" => $items];
612 * Automatically adjust the poll frequency according to the post frequency
614 * @param array $contact
615 * @param array $creation_dates
618 private static function adjustPollFrequency(array $contact, array $creation_dates)
620 if ($contact['network'] != Protocol::FEED) {
621 Logger::info('Contact is no feed, skip.', ['id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url'], 'network' => $contact['network']]);
625 if (!empty($creation_dates)) {
626 // Count the post frequency and the earliest and latest post date
630 $oldest_date = $newest_date = '';
632 foreach ($creation_dates as $date) {
633 $timestamp = strtotime($date);
634 $day = intdiv($timestamp, 86400);
635 $hour = $timestamp % 86400;
637 // Only have a look at values from the last seven days
638 if (((time() / 86400) - $day) < 7) {
639 if (empty($frequency[$day])) {
640 $frequency[$day] = ['count' => 1, 'low' => $hour, 'high' => $hour];
642 ++$frequency[$day]['count'];
643 if ($frequency[$day]['low'] > $hour) {
644 $frequency[$day]['low'] = $hour;
646 if ($frequency[$day]['high'] < $hour) {
647 $frequency[$day]['high'] = $hour;
651 if ($oldest > $day) {
653 $oldest_date = $date;
656 if ($newest < $day) {
658 $newest_date = $date;
662 if (count($creation_dates) == 1) {
663 Logger::info('Feed had posted a single time, switching to daily polling', ['newest' => $newest_date, 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
664 $priority = 8; // Poll once a day
667 if (empty($priority) && (((time() / 86400) - $newest) > 730)) {
668 Logger::info('Feed had not posted for two years, switching to monthly polling', ['newest' => $newest_date, 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
669 $priority = 10; // Poll every month
672 if (empty($priority) && (((time() / 86400) - $newest) > 365)) {
673 Logger::info('Feed had not posted for a year, switching to weekly polling', ['newest' => $newest_date, 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
674 $priority = 9; // Poll every week
677 if (empty($priority) && empty($frequency)) {
678 Logger::info('Feed had not posted for at least a week, switching to daily polling', ['newest' => $newest_date, 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
679 $priority = 8; // Poll once a day
682 if (empty($priority)) {
683 // Calculate the highest "posts per day" value
685 foreach ($frequency as $entry) {
686 if (($entry['count'] == 1) || ($entry['high'] == $entry['low'])) {
690 // We take the earliest and latest post day and interpolate the number of post per day
691 // that would had been created with this post frequency
693 // Assume at least four hours between oldest and newest post per day - should be okay for news outlets
694 $duration = max($entry['high'] - $entry['low'], 14400);
695 $ppd = (86400 / $duration) * $entry['count'];
701 $priority = 1; // Poll every quarter hour
702 } elseif ($max > 24) {
703 $priority = 2; // Poll half an hour
704 } elseif ($max > 12) {
705 $priority = 3; // Poll hourly
706 } elseif ($max > 8) {
707 $priority = 4; // Poll every two hours
708 } elseif ($max > 4) {
709 $priority = 5; // Poll every three hours
710 } elseif ($max > 2) {
711 $priority = 6; // Poll every six hours
713 $priority = 7; // Poll twice a day
715 Logger::info('Calculated priority by the posts per day', ['priority' => $priority, 'max' => round($max, 2), 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
718 Logger::info('No posts, switching to daily polling', ['id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
719 $priority = 8; // Poll once a day
722 if ($contact['rating'] != $priority) {
723 Logger::notice('Adjusting priority', ['old' => $contact['rating'], 'new' => $priority, 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
724 DBA::update('contact', ['rating' => $priority], ['id' => $contact['id']]);
729 * Get the poll interval for the given contact array
731 * @param array $contact
732 * @return int Poll interval in minutes
734 public static function getPollInterval(array $contact)
736 if (in_array($contact['network'], [Protocol::MAIL, Protocol::FEED])) {
737 $ratings = [0, 3, 7, 8, 9, 10];
738 if (DI::config()->get('system', 'adjust_poll_frequency') && ($contact['network'] == Protocol::FEED)) {
739 $rating = $contact['rating'];
740 } elseif (array_key_exists($contact['priority'], $ratings)) {
741 $rating = $ratings[$contact['priority']];
746 // Check once a week per default for all other networks
750 // Friendica and OStatus are checked once a day
751 if (in_array($contact['network'], [Protocol::DFRN, Protocol::OSTATUS])) {
755 // Check archived contacts or contacts with unsupported protocols once a month
756 if ($contact['archive'] || in_array($contact['network'], [Protocol::ZOT, Protocol::PHANTOM])) {
764 * Based on $contact['priority'], should we poll this site now? Or later?
767 $min_poll_interval = max(1, DI::config()->get('system', 'min_poll_interval'));
769 $poll_intervals = [$min_poll_interval, 15, 30, 60, 120, 180, 360, 720 ,1440, 10080, 43200];
771 //$poll_intervals = [$min_poll_interval . ' minute', '15 minute', '30 minute',
772 // '1 hour', '2 hour', '3 hour', '6 hour', '12 hour' ,'1 day', '1 week', '1 month'];
774 return $poll_intervals[$rating];
778 * Convert a tag array to a tag string
781 * @return string tag string
783 private static function tagToString(array $tags)
787 foreach ($tags as $tag) {
792 $tagstr .= "#[url=" . DI::baseUrl() . "/search?tag=" . urlencode($tag) . "]" . $tag . "[/url]";
798 private static function titleIsBody($title, $body)
800 $title = strip_tags($title);
801 $title = trim($title);
802 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
803 $title = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $title);
805 $body = strip_tags($body);
807 $body = html_entity_decode($body, ENT_QUOTES, 'UTF-8');
808 $body = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $body);
810 if (strlen($title) < strlen($body)) {
811 $body = substr($body, 0, strlen($title));
814 if (($title != $body) && (substr($title, -3) == "...")) {
815 $pos = strrpos($title, "...");
817 $title = substr($title, 0, $pos);
818 $body = substr($body, 0, $pos);
821 return ($title == $body);
825 * Creates the Atom feed for a given nickname
828 * - activity (default): all the public posts
829 * - posts: all the public top-level posts
830 * - comments: all the public replies
832 * Updates the provided last_update parameter if the result comes from the
833 * cache or it is empty
835 * @param string $owner_nick Nickname of the feed owner
836 * @param string $last_update Date of the last update
837 * @param integer $max_items Number of maximum items to fetch
838 * @param string $filter Feed items filter (activity, posts or comments)
839 * @param boolean $nocache Wether to bypass caching
841 * @return string Atom feed
842 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
843 * @throws \ImagickException
845 public static function atom($owner_nick, $last_update, $max_items = 300, $filter = 'activity', $nocache = false)
847 $stamp = microtime(true);
849 $owner = User::getOwnerDataByNick($owner_nick);
854 $cachekey = "feed:feed:" . $owner_nick . ":" . $filter . ":" . $last_update;
856 $previous_created = $last_update;
858 // Don't cache when the last item was posted less then 15 minutes ago (Cache duration)
859 if ((time() - strtotime($owner['last-item'])) < 15*60) {
860 $result = DI::cache()->get($cachekey);
861 if (!$nocache && !is_null($result)) {
862 Logger::info('Cached feed duration', ['seconds' => number_format(microtime(true) - $stamp, 3), 'nick' => $owner_nick, 'filter' => $filter, 'created' => $previous_created]);
863 return $result['feed'];
867 $check_date = empty($last_update) ? '' : DateTimeFormat::utc($last_update);
868 $authorid = Contact::getIdForURL($owner["url"]);
870 $condition = ["`uid` = ? AND `received` > ? AND NOT `deleted` AND `gravity` IN (?, ?)
871 AND `private` != ? AND `visible` AND `wall` AND `parent-network` IN (?, ?, ?, ?)",
872 $owner["uid"], $check_date, GRAVITY_PARENT, GRAVITY_COMMENT,
873 Item::PRIVATE, Protocol::ACTIVITYPUB,
874 Protocol::OSTATUS, Protocol::DFRN, Protocol::DIASPORA];
876 if ($filter === 'comments') {
877 $condition[0] .= " AND `object-type` = ? ";
878 $condition[] = Activity\ObjectType::COMMENT;
881 if ($owner['account-type'] != User::ACCOUNT_TYPE_COMMUNITY) {
882 $condition[0] .= " AND `contact-id` = ? AND `author-id` = ?";
883 $condition[] = $owner["id"];
884 $condition[] = $authorid;
887 $params = ['order' => ['received' => true], 'limit' => $max_items];
889 if ($filter === 'posts') {
890 $ret = Item::selectThread([], $condition, $params);
892 $ret = Item::select([], $condition, $params);
895 $items = Item::inArray($ret);
897 $doc = new DOMDocument('1.0', 'utf-8');
898 $doc->formatOutput = true;
900 $root = self::addHeader($doc, $owner, $filter);
902 foreach ($items as $item) {
903 $entry = self::entry($doc, $item, $owner);
904 $root->appendChild($entry);
906 if ($last_update < $item['created']) {
907 $last_update = $item['created'];
911 $feeddata = trim($doc->saveXML());
913 $msg = ['feed' => $feeddata, 'last_update' => $last_update];
914 DI::cache()->set($cachekey, $msg, Duration::QUARTER_HOUR);
916 Logger::info('Feed duration', ['seconds' => number_format(microtime(true) - $stamp, 3), 'nick' => $owner_nick, 'filter' => $filter, 'created' => $previous_created]);
922 * Adds the header elements to the XML document
924 * @param DOMDocument $doc XML document
925 * @param array $owner Contact data of the poster
926 * @param string $filter The related feed filter (activity, posts or comments)
928 * @return object header root element
929 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
931 private static function addHeader(DOMDocument $doc, array $owner, $filter)
933 $root = $doc->createElementNS(ActivityNamespace::ATOM1, 'feed');
934 $doc->appendChild($root);
937 $selfUri = '/feed/' . $owner["nick"] . '/';
940 $title = DI::l10n()->t('%s\'s timeline', $owner['name']);
944 $title = DI::l10n()->t('%s\'s posts', $owner['name']);
947 $title = DI::l10n()->t('%s\'s comments', $owner['name']);
952 $attributes = ["uri" => "https://friendi.ca", "version" => FRIENDICA_VERSION . "-" . DB_UPDATE_VERSION];
953 XML::addElement($doc, $root, "generator", FRIENDICA_PLATFORM, $attributes);
954 XML::addElement($doc, $root, "id", DI::baseUrl() . "/profile/" . $owner["nick"]);
955 XML::addElement($doc, $root, "title", $title);
956 XML::addElement($doc, $root, "subtitle", sprintf("Updates from %s on %s", $owner["name"], DI::config()->get('config', 'sitename')));
957 XML::addElement($doc, $root, "logo", $owner["photo"]);
958 XML::addElement($doc, $root, "updated", DateTimeFormat::utcNow(DateTimeFormat::ATOM));
960 $author = self::addAuthor($doc, $owner);
961 $root->appendChild($author);
963 $attributes = ["href" => $owner["url"], "rel" => "alternate", "type" => "text/html"];
964 XML::addElement($doc, $root, "link", "", $attributes);
966 OStatus::hublinks($doc, $root, $owner["nick"]);
968 $attributes = ["href" => DI::baseUrl() . $selfUri, "rel" => "self", "type" => "application/atom+xml"];
969 XML::addElement($doc, $root, "link", "", $attributes);
975 * Adds the author element to the XML document
977 * @param DOMDocument $doc XML document
978 * @param array $owner Contact data of the poster
980 * @return \DOMElement author element
981 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
983 private static function addAuthor(DOMDocument $doc, array $owner)
985 $author = $doc->createElement("author");
986 XML::addElement($doc, $author, "uri", $owner["url"]);
987 XML::addElement($doc, $author, "name", $owner["nick"]);
988 XML::addElement($doc, $author, "email", $owner["addr"]);
994 * Adds an entry element to the XML document
996 * @param DOMDocument $doc XML document
997 * @param array $item Data of the item that is to be posted
998 * @param array $owner Contact data of the poster
999 * @param bool $toplevel optional default false
1001 * @return \DOMElement Entry element
1002 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1003 * @throws \ImagickException
1005 private static function entry(DOMDocument $doc, array $item, array $owner)
1009 $repeated_guid = OStatus::getResharedGuid($item);
1010 if ($repeated_guid != "") {
1011 $xml = self::reshareEntry($doc, $item, $owner, $repeated_guid);
1018 return self::noteEntry($doc, $item, $owner);
1022 * Adds an entry element with reshared content
1024 * @param DOMDocument $doc XML document
1025 * @param array $item Data of the item that is to be posted
1026 * @param array $owner Contact data of the poster
1027 * @param string $repeated_guid guid
1028 * @param bool $toplevel Is it for en entry element (false) or a feed entry (true)?
1030 * @return bool Entry element
1031 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1032 * @throws \ImagickException
1034 private static function reshareEntry(DOMDocument $doc, array $item, array $owner, $repeated_guid)
1036 if (($item['gravity'] != GRAVITY_PARENT) && (Strings::normaliseLink($item["author-link"]) != Strings::normaliseLink($owner["url"]))) {
1037 Logger::info('Feed entry author does not match feed owner', ['owner' => $owner["url"], 'author' => $item["author-link"]]);
1040 $entry = OStatus::entryHeader($doc, $owner, $item, false);
1042 $condition = ['uid' => $owner["uid"], 'guid' => $repeated_guid, 'private' => [Item::PUBLIC, Item::UNLISTED],
1043 'network' => Protocol::FEDERATED];
1044 $repeated_item = Post::selectFirst([], $condition);
1045 if (!DBA::isResult($repeated_item)) {
1049 self::entryContent($doc, $entry, $item, self::getTitle($repeated_item), Activity::SHARE, false);
1051 self::entryFooter($doc, $entry, $item, $owner);
1057 * Adds a regular entry element
1059 * @param DOMDocument $doc XML document
1060 * @param array $item Data of the item that is to be posted
1061 * @param array $owner Contact data of the poster
1062 * @param bool $toplevel Is it for en entry element (false) or a feed entry (true)?
1064 * @return \DOMElement Entry element
1065 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1066 * @throws \ImagickException
1068 private static function noteEntry(DOMDocument $doc, array $item, array $owner)
1070 if (($item['gravity'] != GRAVITY_PARENT) && (Strings::normaliseLink($item["author-link"]) != Strings::normaliseLink($owner["url"]))) {
1071 Logger::info('Feed entry author does not match feed owner', ['owner' => $owner["url"], 'author' => $item["author-link"]]);
1074 $entry = OStatus::entryHeader($doc, $owner, $item, false);
1076 self::entryContent($doc, $entry, $item, self::getTitle($item), '', true);
1078 self::entryFooter($doc, $entry, $item, $owner);
1084 * Adds elements to the XML document
1086 * @param DOMDocument $doc XML document
1087 * @param \DOMElement $entry Entry element where the content is added
1088 * @param array $item Data of the item that is to be posted
1089 * @param array $owner Contact data of the poster
1090 * @param string $title Title for the post
1091 * @param string $verb The activity verb
1092 * @param bool $complete Add the "status_net" element?
1093 * @param bool $feed_mode Behave like a regular feed for users if true
1095 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1097 private static function entryContent(DOMDocument $doc, \DOMElement $entry, array $item, $title, $verb = "", $complete = true)
1100 $verb = OStatus::constructVerb($item);
1103 XML::addElement($doc, $entry, "id", $item["uri"]);
1104 XML::addElement($doc, $entry, "title", html_entity_decode($title, ENT_QUOTES, 'UTF-8'));
1106 $body = OStatus::formatPicturePost($item['body']);
1108 $body = BBCode::convert($body, false, BBCode::OSTATUS);
1110 XML::addElement($doc, $entry, "content", $body, ["type" => "html"]);
1112 XML::addElement($doc, $entry, "link", "", ["rel" => "alternate", "type" => "text/html",
1113 "href" => DI::baseUrl()."/display/".$item["guid"]]
1116 XML::addElement($doc, $entry, "published", DateTimeFormat::utc($item["created"]."+00:00", DateTimeFormat::ATOM));
1117 XML::addElement($doc, $entry, "updated", DateTimeFormat::utc($item["edited"]."+00:00", DateTimeFormat::ATOM));
1121 * Adds the elements at the foot of an entry to the XML document
1123 * @param DOMDocument $doc XML document
1124 * @param object $entry The entry element where the elements are added
1125 * @param array $item Data of the item that is to be posted
1126 * @param array $owner Contact data of the poster
1127 * @param bool $complete default true
1129 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1131 private static function entryFooter(DOMDocument $doc, $entry, array $item, array $owner)
1135 if ($item['gravity'] != GRAVITY_PARENT) {
1136 $parent = Post::selectFirst(['guid', 'author-link', 'owner-link'], ['id' => $item['parent']]);
1138 $thrparent = Post::selectFirst(['guid', 'author-link', 'owner-link', 'plink'], ['uid' => $owner["uid"], 'uri' => $item['thr-parent']]);
1140 if (DBA::isResult($thrparent)) {
1141 $mentioned[$thrparent["author-link"]] = $thrparent["author-link"];
1142 $mentioned[$thrparent["owner-link"]] = $thrparent["owner-link"];
1143 $parent_plink = $thrparent["plink"];
1145 $mentioned[$parent["author-link"]] = $parent["author-link"];
1146 $mentioned[$parent["owner-link"]] = $parent["owner-link"];
1147 $parent_plink = DI::baseUrl()."/display/".$parent["guid"];
1151 "ref" => $item['thr-parent'],
1152 "href" => $parent_plink];
1153 XML::addElement($doc, $entry, "thr:in-reply-to", "", $attributes);
1157 "href" => $parent_plink];
1158 XML::addElement($doc, $entry, "link", "", $attributes);
1161 // uri-id isn't present for follow entry pseudo-items
1162 $tags = Tag::getByURIId($item['uri-id'] ?? 0);
1163 foreach ($tags as $tag) {
1164 $mentioned[$tag['url']] = $tag['url'];
1167 foreach ($tags as $tag) {
1168 if ($tag['type'] == Tag::HASHTAG) {
1169 XML::addElement($doc, $entry, "category", "", ["term" => $tag['name']]);
1173 OStatus::getAttachment($doc, $entry, $item);
1177 * Fetch or create title for feed entry
1179 * @param array $item
1180 * @return string title
1182 private static function getTitle(array $item)
1184 if ($item['title'] != '') {
1185 return BBCode::convert($item['title'], false, BBCode::OSTATUS);
1188 // Fetch information about the post
1189 $siteinfo = BBCode::getAttachedData($item["body"]);
1190 if (isset($siteinfo["title"])) {
1191 return $siteinfo["title"];
1194 // If no bookmark is found then take the first line
1195 // Remove the share element before fetching the first line
1196 $title = trim(preg_replace("/\[share.*?\](.*?)\[\/share\]/ism","\n$1\n",$item['body']));
1198 $title = HTML::toPlaintext(BBCode::convert($title, false), 0, true)."\n";
1199 $pos = strpos($title, "\n");
1201 if (($pos == 0) || ($pos > 100)) {
1206 return substr($title, 0, $pos) . $trailer;