3 * @copyright Copyright (C) 2010-2021, the Friendica project
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;
238 $header["post-type"] = Item::PT_ARTICLE;
240 $header["contact-id"] = $contact["id"] ?? 0;
242 if (!is_object($entries)) {
243 Logger::info("There are no entries in this feed.");
248 $creation_dates = [];
250 // Limit the number of items that are about to be fetched
251 $total_items = ($entries->length - 1);
252 $max_items = DI::config()->get('system', 'max_feed_items');
253 if (($max_items > 0) && ($total_items > $max_items)) {
254 $total_items = $max_items;
259 // Importing older entries first
260 for ($i = $total_items; $i >= 0; --$i) {
261 $entry = $entries->item($i);
263 $item = array_merge($header, $author);
265 $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']", $entry);
266 if (!is_object($alternate)) {
267 $alternate = XML::getFirstAttributes($xpath, "atom:link", $entry);
269 if (is_object($alternate)) {
270 foreach ($alternate AS $attribute) {
271 if ($attribute->name == "href") {
272 $item["plink"] = $attribute->textContent;
277 if (empty($item["plink"])) {
278 $item["plink"] = XML::getFirstNodeValue($xpath, 'link/text()', $entry);
281 if (empty($item["plink"])) {
282 $item["plink"] = XML::getFirstNodeValue($xpath, 'rss:link/text()', $entry);
285 $item["uri"] = XML::getFirstNodeValue($xpath, 'atom:id/text()', $entry);
287 if (empty($item["uri"])) {
288 $item["uri"] = XML::getFirstNodeValue($xpath, 'guid/text()', $entry);
291 if (empty($item["uri"])) {
292 $item["uri"] = $item["plink"];
295 // Add the base path if missing
296 $item["uri"] = Network::addBasePath($item["uri"], $basepath);
297 $item["plink"] = Network::addBasePath($item["plink"], $basepath);
299 $orig_plink = $item["plink"];
301 $item["plink"] = DI::httpRequest()->finalUrl($item["plink"]);
303 $item["title"] = XML::getFirstNodeValue($xpath, 'atom:title/text()', $entry);
305 if (empty($item["title"])) {
306 $item["title"] = XML::getFirstNodeValue($xpath, 'title/text()', $entry);
308 if (empty($item["title"])) {
309 $item["title"] = XML::getFirstNodeValue($xpath, 'rss:title/text()', $entry);
312 $item["title"] = html_entity_decode($item["title"], ENT_QUOTES, 'UTF-8');
314 $published = XML::getFirstNodeValue($xpath, 'atom:published/text()', $entry);
316 if (empty($published)) {
317 $published = XML::getFirstNodeValue($xpath, 'pubDate/text()', $entry);
320 if (empty($published)) {
321 $published = XML::getFirstNodeValue($xpath, 'dc:date/text()', $entry);
324 $updated = XML::getFirstNodeValue($xpath, 'atom:updated/text()', $entry);
326 if (empty($updated) && !empty($published)) {
327 $updated = $published;
330 if (empty($published) && !empty($updated)) {
331 $published = $updated;
334 if ($published != "") {
335 $item["created"] = $published;
338 if ($updated != "") {
339 $item["edited"] = $updated;
343 $condition = ["`uid` = ? AND `uri` = ? AND `network` IN (?, ?)",
344 $importer["uid"], $item["uri"], Protocol::FEED, Protocol::DFRN];
345 $previous = Post::selectFirst(['id', 'created'], $condition);
346 if (DBA::isResult($previous)) {
347 // Use the creation date when the post had been stored. It can happen this date changes in the feed.
348 $creation_dates[] = $previous['created'];
349 Logger::info("Item with uri " . $item["uri"] . " for user " . $importer["uid"] . " already existed under id " . $previous["id"]);
352 $creation_dates[] = DateTimeFormat::utc($item['created']);
355 $creator = XML::getFirstNodeValue($xpath, 'author/text()', $entry);
357 if (empty($creator)) {
358 $creator = XML::getFirstNodeValue($xpath, 'atom:author/atom:name/text()', $entry);
361 if (empty($creator)) {
362 $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
365 if ($creator != "") {
366 $item["author-name"] = $creator;
369 $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
371 if ($creator != "") {
372 $item["author-name"] = $creator;
376 // <category>Ausland</category>
377 // <media:thumbnail width="152" height="76" url="http://www.taz.de/picture/667875/192/14388767.jpg"/>
381 $enclosures = $xpath->query("enclosure|atom:link[@rel='enclosure']", $entry);
382 foreach ($enclosures AS $enclosure) {
387 foreach ($enclosure->attributes AS $attribute) {
388 if (in_array($attribute->name, ["url", "href"])) {
389 $href = $attribute->textContent;
390 } elseif ($attribute->name == "length") {
391 $length = (int)$attribute->textContent;
392 } elseif ($attribute->name == "type") {
393 $type = $attribute->textContent;
398 $attachments[] = ['type' => Post\Media::DOCUMENT, 'url' => $href, 'mimetype' => $type, 'size' => $length];
403 $categories = $xpath->query("category", $entry);
404 foreach ($categories AS $category) {
405 $taglist[] = $category->nodeValue;
408 $body = trim(XML::getFirstNodeValue($xpath, 'atom:content/text()', $entry));
411 $body = trim(XML::getFirstNodeValue($xpath, 'content:encoded/text()', $entry));
414 $summary = trim(XML::getFirstNodeValue($xpath, 'atom:summary/text()', $entry));
416 if (empty($summary)) {
417 $summary = trim(XML::getFirstNodeValue($xpath, 'description/text()', $entry));
425 if ($body == $summary) {
429 // remove the content of the title if it is identically to the body
430 // This helps with auto generated titles e.g. from tumblr
431 if (self::titleIsBody($item["title"], $body)) {
434 $item["body"] = HTML::toBBCode($body, $basepath);
436 // Remove tracking pixels
437 $item["body"] = preg_replace("/\[img=1x1\]([^\[\]]*)\[\/img\]/Usi", '', $item["body"]);
439 if (($item["body"] == '') && ($item["title"] != '')) {
440 $item["body"] = $item["title"];
447 } elseif (!Item::isValid($item)) {
448 Logger::info('Feed item is invalid', ['created' => $item['created'], 'uid' => $item['uid'], 'uri' => $item['uri']]);
450 } elseif (Item::isTooOld($item)) {
451 Logger::info('Feed is too old', ['created' => $item['created'], 'uid' => $item['uid'], 'uri' => $item['uri']]);
456 if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] < 3)) {
457 // Handle enclosures and treat them as preview picture
458 foreach ($attachments AS $attachment) {
459 if ($attachment["mimetype"] == "image/jpeg") {
460 $preview = $attachment["url"];
464 // Remove a possible link to the item itself
465 $item["body"] = str_replace($item["plink"], '', $item["body"]);
466 $item["body"] = trim(preg_replace('/\[url\=\](\w+.*?)\[\/url\]/i', '', $item["body"]));
468 // Replace the content when the title is longer than the body
469 $replace = (strlen($item["title"]) > strlen($item["body"]));
471 // Replace it, when there is an image in the body
472 if (strstr($item["body"], '[/img]')) {
476 // Replace it, when there is a link in the body
477 if (strstr($item["body"], '[/url]')) {
481 $saved_body = $item["body"];
482 $saved_title = $item["title"];
485 $item["body"] = trim($item["title"]);
488 $data = ParseUrl::getSiteinfoCached($item['plink']);
489 if (!empty($data['text']) && !empty($data['title']) && (mb_strlen($item['body']) < mb_strlen($data['text']))) {
490 // When the fetched page info text is longer than the body, we do try to enhance the body
491 if (!empty($item['body']) && (strpos($data['title'], $item['body']) === false) && (strpos($data['text'], $item['body']) === false)) {
492 // The body is not part of the fetched page info title or page info text. So we add the text to the body
493 $item['body'] .= "\n\n" . $data['text'];
495 // Else we replace the body with the page info text
496 $item['body'] = $data['text'];
500 $data = PageInfo::queryUrl($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_denylist"] ?? '');
503 // Take the data that was provided by the feed if the query is empty
504 if (($data['type'] == 'link') && empty($data['title']) && empty($data['text'])) {
505 $data['title'] = $saved_title;
506 $item["body"] = $saved_body;
509 $data_text = strip_tags(trim($data['text'] ?? ''));
510 $item_body = strip_tags(trim($item['body'] ?? ''));
512 if (!empty($data_text) && (($data_text == $item_body) || strstr($item_body, $data_text))) {
516 // We always strip the title since it will be added in the page information
518 $item["body"] = $item["body"] . "\n" . PageInfo::getFooterFromData($data, false);
519 $taglist = $contact["fetch_further_information"] == 2 ? PageInfo::getTagsFromUrl($item["plink"], $preview, $contact["ffi_keyword_denylist"] ?? '') : [];
520 $item["object-type"] = Activity\ObjectType::BOOKMARK;
524 if (!empty($summary)) {
525 $item["body"] = '[abstract]' . HTML::toBBCode($summary, $basepath) . "[/abstract]\n" . $item["body"];
528 if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] == 3)) {
529 if (empty($taglist)) {
530 $taglist = PageInfo::getTagsFromUrl($item["plink"], $preview, $contact["ffi_keyword_denylist"] ?? '');
532 $item["body"] .= "\n" . self::tagToString($taglist);
537 // Add the link to the original feed entry if not present in feed
538 if (($item['plink'] != '') && !strstr($item["body"], $item['plink'])) {
539 $item["body"] .= "[hr][url]" . $item['plink'] . "[/url]";
543 if (empty($item['title'])) {
544 $item['post-type'] = Item::PT_NOTE;
547 Logger::info('Stored feed', ['item' => $item]);
549 $notify = Item::isRemoteSelf($contact, $item);
551 // Distributed items should have a well formatted URI.
552 // Additionally we have to avoid conflicts with identical URI between imported feeds and these items.
554 $item['guid'] = Item::guidFromUri($orig_plink, DI::baseUrl()->getHostname());
555 $item['uri'] = Item::newURI($item['uid'], $item['guid']);
556 unset($item['thr-parent']);
557 unset($item['parent-uri']);
559 // Set the delivery priority for "remote self" to "medium"
560 $notify = PRIORITY_MEDIUM;
563 $condition = ['uid' => $item['uid'], 'uri' => $item['uri']];
564 if (!Post::exists($condition) && !Post\Delayed::exists($item["uri"], $item['uid'])) {
566 Post\Delayed::publish($item, $notify, $taglist, $attachments);
568 $postings[] = ['item' => $item, 'notify' => $notify,
569 'taglist' => $taglist, 'attachments' => $attachments];
572 Logger::info('Post already created or exists in the delayed posts queue', ['uid' => $item['uid'], 'uri' => $item["uri"]]);
576 if (!empty($postings)) {
577 $min_posting = DI::config()->get('system', 'minimum_posting_interval', 0);
578 $total = count($postings);
580 // Posts shouldn't be delayed more than a day
581 $interval = min(1440, self::getPollInterval($contact));
582 $delay = max(round(($interval * 60) / $total), 60 * $min_posting);
583 Logger::info('Got posting delay', ['delay' => $delay, 'interval' => $interval, 'items' => $total, 'cid' => $contact['id'], 'url' => $contact['url']]);
590 foreach ($postings as $posting) {
592 $publish_time = time() + $post_delay;
593 $post_delay += $delay;
595 $publish_time = time();
598 $last_publish = DI::pConfig()->get($posting['item']['uid'], 'system', 'last_publish', 0, true);
599 $next_publish = max($last_publish + (60 * $min_posting), time());
600 if ($publish_time < $next_publish) {
601 $publish_time = $next_publish;
603 $publish_at = date(DateTimeFormat::MYSQL, $publish_time);
605 Post\Delayed::add($posting['item']['uri'], $posting['item'], $posting['notify'], false, $publish_at, $posting['taglist'], $posting['attachments']);
609 if (!$dryRun && DI::config()->get('system', 'adjust_poll_frequency')) {
610 self::adjustPollFrequency($contact, $creation_dates);
613 return ["header" => $author, "items" => $items];
617 * Automatically adjust the poll frequency according to the post frequency
619 * @param array $contact
620 * @param array $creation_dates
623 private static function adjustPollFrequency(array $contact, array $creation_dates)
625 if ($contact['network'] != Protocol::FEED) {
626 Logger::info('Contact is no feed, skip.', ['id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url'], 'network' => $contact['network']]);
630 if (!empty($creation_dates)) {
631 // Count the post frequency and the earliest and latest post date
635 $oldest_date = $newest_date = '';
637 foreach ($creation_dates as $date) {
638 $timestamp = strtotime($date);
639 $day = intdiv($timestamp, 86400);
640 $hour = $timestamp % 86400;
642 // Only have a look at values from the last seven days
643 if (((time() / 86400) - $day) < 7) {
644 if (empty($frequency[$day])) {
645 $frequency[$day] = ['count' => 1, 'low' => $hour, 'high' => $hour];
647 ++$frequency[$day]['count'];
648 if ($frequency[$day]['low'] > $hour) {
649 $frequency[$day]['low'] = $hour;
651 if ($frequency[$day]['high'] < $hour) {
652 $frequency[$day]['high'] = $hour;
656 if ($oldest > $day) {
658 $oldest_date = $date;
661 if ($newest < $day) {
663 $newest_date = $date;
667 if (count($creation_dates) == 1) {
668 Logger::info('Feed had posted a single time, switching to daily polling', ['newest' => $newest_date, 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
669 $priority = 8; // Poll once a day
672 if (empty($priority) && (((time() / 86400) - $newest) > 730)) {
673 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']]);
674 $priority = 10; // Poll every month
677 if (empty($priority) && (((time() / 86400) - $newest) > 365)) {
678 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']]);
679 $priority = 9; // Poll every week
682 if (empty($priority) && empty($frequency)) {
683 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']]);
684 $priority = 8; // Poll once a day
687 if (empty($priority)) {
688 // Calculate the highest "posts per day" value
690 foreach ($frequency as $entry) {
691 if (($entry['count'] == 1) || ($entry['high'] == $entry['low'])) {
695 // We take the earliest and latest post day and interpolate the number of post per day
696 // that would had been created with this post frequency
698 // Assume at least four hours between oldest and newest post per day - should be okay for news outlets
699 $duration = max($entry['high'] - $entry['low'], 14400);
700 $ppd = (86400 / $duration) * $entry['count'];
706 $priority = 1; // Poll every quarter hour
707 } elseif ($max > 24) {
708 $priority = 2; // Poll half an hour
709 } elseif ($max > 12) {
710 $priority = 3; // Poll hourly
711 } elseif ($max > 8) {
712 $priority = 4; // Poll every two hours
713 } elseif ($max > 4) {
714 $priority = 5; // Poll every three hours
715 } elseif ($max > 2) {
716 $priority = 6; // Poll every six hours
718 $priority = 7; // Poll twice a day
720 Logger::info('Calculated priority by the posts per day', ['priority' => $priority, 'max' => round($max, 2), 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
723 Logger::info('No posts, switching to daily polling', ['id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
724 $priority = 8; // Poll once a day
727 if ($contact['rating'] != $priority) {
728 Logger::notice('Adjusting priority', ['old' => $contact['rating'], 'new' => $priority, 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
729 DBA::update('contact', ['rating' => $priority], ['id' => $contact['id']]);
734 * Get the poll interval for the given contact array
736 * @param array $contact
737 * @return int Poll interval in minutes
739 public static function getPollInterval(array $contact)
741 if (in_array($contact['network'], [Protocol::MAIL, Protocol::FEED])) {
742 $ratings = [0, 3, 7, 8, 9, 10];
743 if (DI::config()->get('system', 'adjust_poll_frequency') && ($contact['network'] == Protocol::FEED)) {
744 $rating = $contact['rating'];
745 } elseif (array_key_exists($contact['priority'], $ratings)) {
746 $rating = $ratings[$contact['priority']];
751 // Check once a week per default for all other networks
755 // Friendica and OStatus are checked once a day
756 if (in_array($contact['network'], [Protocol::DFRN, Protocol::OSTATUS])) {
760 // Check archived contacts or contacts with unsupported protocols once a month
761 if ($contact['archive'] || in_array($contact['network'], [Protocol::ZOT, Protocol::PHANTOM])) {
769 * Based on $contact['priority'], should we poll this site now? Or later?
772 $min_poll_interval = max(1, DI::config()->get('system', 'min_poll_interval'));
774 $poll_intervals = [$min_poll_interval, 15, 30, 60, 120, 180, 360, 720 ,1440, 10080, 43200];
776 //$poll_intervals = [$min_poll_interval . ' minute', '15 minute', '30 minute',
777 // '1 hour', '2 hour', '3 hour', '6 hour', '12 hour' ,'1 day', '1 week', '1 month'];
779 return $poll_intervals[$rating];
783 * Convert a tag array to a tag string
786 * @return string tag string
788 private static function tagToString(array $tags)
792 foreach ($tags as $tag) {
797 $tagstr .= "#[url=" . DI::baseUrl() . "/search?tag=" . urlencode($tag) . "]" . $tag . "[/url]";
803 private static function titleIsBody($title, $body)
805 $title = strip_tags($title);
806 $title = trim($title);
807 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
808 $title = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $title);
810 $body = strip_tags($body);
812 $body = html_entity_decode($body, ENT_QUOTES, 'UTF-8');
813 $body = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $body);
815 if (strlen($title) < strlen($body)) {
816 $body = substr($body, 0, strlen($title));
819 if (($title != $body) && (substr($title, -3) == "...")) {
820 $pos = strrpos($title, "...");
822 $title = substr($title, 0, $pos);
823 $body = substr($body, 0, $pos);
826 return ($title == $body);
830 * Creates the Atom feed for a given nickname
833 * - activity (default): all the public posts
834 * - posts: all the public top-level posts
835 * - comments: all the public replies
837 * Updates the provided last_update parameter if the result comes from the
838 * cache or it is empty
840 * @param string $owner_nick Nickname of the feed owner
841 * @param string $last_update Date of the last update
842 * @param integer $max_items Number of maximum items to fetch
843 * @param string $filter Feed items filter (activity, posts or comments)
844 * @param boolean $nocache Wether to bypass caching
846 * @return string Atom feed
847 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
848 * @throws \ImagickException
850 public static function atom($owner_nick, $last_update, $max_items = 300, $filter = 'activity', $nocache = false)
852 $stamp = microtime(true);
854 $owner = User::getOwnerDataByNick($owner_nick);
859 $cachekey = "feed:feed:" . $owner_nick . ":" . $filter . ":" . $last_update;
861 $previous_created = $last_update;
863 // Don't cache when the last item was posted less then 15 minutes ago (Cache duration)
864 if ((time() - strtotime($owner['last-item'])) < 15*60) {
865 $result = DI::cache()->get($cachekey);
866 if (!$nocache && !is_null($result)) {
867 Logger::info('Cached feed duration', ['seconds' => number_format(microtime(true) - $stamp, 3), 'nick' => $owner_nick, 'filter' => $filter, 'created' => $previous_created]);
868 return $result['feed'];
872 $check_date = empty($last_update) ? '' : DateTimeFormat::utc($last_update);
873 $authorid = Contact::getIdForURL($owner["url"]);
875 $condition = ["`uid` = ? AND `received` > ? AND NOT `deleted` AND `gravity` IN (?, ?)
876 AND `private` != ? AND `visible` AND `wall` AND `parent-network` IN (?, ?, ?, ?)",
877 $owner["uid"], $check_date, GRAVITY_PARENT, GRAVITY_COMMENT,
878 Item::PRIVATE, Protocol::ACTIVITYPUB,
879 Protocol::OSTATUS, Protocol::DFRN, Protocol::DIASPORA];
881 if ($filter === 'comments') {
882 $condition[0] .= " AND `object-type` = ? ";
883 $condition[] = Activity\ObjectType::COMMENT;
886 if ($owner['account-type'] != User::ACCOUNT_TYPE_COMMUNITY) {
887 $condition[0] .= " AND `contact-id` = ? AND `author-id` = ?";
888 $condition[] = $owner["id"];
889 $condition[] = $authorid;
892 $params = ['order' => ['received' => true], 'limit' => $max_items];
894 if ($filter === 'posts') {
895 $ret = Post::selectThread(Item::DELIVER_FIELDLIST, $condition, $params);
897 $ret = Post::select(Item::DELIVER_FIELDLIST, $condition, $params);
900 $items = Post::toArray($ret);
902 $doc = new DOMDocument('1.0', 'utf-8');
903 $doc->formatOutput = true;
905 $root = self::addHeader($doc, $owner, $filter);
907 foreach ($items as $item) {
908 $entry = self::entry($doc, $item, $owner);
909 $root->appendChild($entry);
911 if ($last_update < $item['created']) {
912 $last_update = $item['created'];
916 $feeddata = trim($doc->saveXML());
918 $msg = ['feed' => $feeddata, 'last_update' => $last_update];
919 DI::cache()->set($cachekey, $msg, Duration::QUARTER_HOUR);
921 Logger::info('Feed duration', ['seconds' => number_format(microtime(true) - $stamp, 3), 'nick' => $owner_nick, 'filter' => $filter, 'created' => $previous_created]);
927 * Adds the header elements to the XML document
929 * @param DOMDocument $doc XML document
930 * @param array $owner Contact data of the poster
931 * @param string $filter The related feed filter (activity, posts or comments)
933 * @return object header root element
934 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
936 private static function addHeader(DOMDocument $doc, array $owner, $filter)
938 $root = $doc->createElementNS(ActivityNamespace::ATOM1, 'feed');
939 $doc->appendChild($root);
942 $selfUri = '/feed/' . $owner["nick"] . '/';
945 $title = DI::l10n()->t('%s\'s timeline', $owner['name']);
949 $title = DI::l10n()->t('%s\'s posts', $owner['name']);
952 $title = DI::l10n()->t('%s\'s comments', $owner['name']);
957 $attributes = ["uri" => "https://friendi.ca", "version" => FRIENDICA_VERSION . "-" . DB_UPDATE_VERSION];
958 XML::addElement($doc, $root, "generator", FRIENDICA_PLATFORM, $attributes);
959 XML::addElement($doc, $root, "id", DI::baseUrl() . "/profile/" . $owner["nick"]);
960 XML::addElement($doc, $root, "title", $title);
961 XML::addElement($doc, $root, "subtitle", sprintf("Updates from %s on %s", $owner["name"], DI::config()->get('config', 'sitename')));
962 XML::addElement($doc, $root, "logo", $owner["photo"]);
963 XML::addElement($doc, $root, "updated", DateTimeFormat::utcNow(DateTimeFormat::ATOM));
965 $author = self::addAuthor($doc, $owner);
966 $root->appendChild($author);
968 $attributes = ["href" => $owner["url"], "rel" => "alternate", "type" => "text/html"];
969 XML::addElement($doc, $root, "link", "", $attributes);
971 OStatus::hublinks($doc, $root, $owner["nick"]);
973 $attributes = ["href" => DI::baseUrl() . $selfUri, "rel" => "self", "type" => "application/atom+xml"];
974 XML::addElement($doc, $root, "link", "", $attributes);
980 * Adds the author element to the XML document
982 * @param DOMDocument $doc XML document
983 * @param array $owner Contact data of the poster
985 * @return \DOMElement author element
986 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
988 private static function addAuthor(DOMDocument $doc, array $owner)
990 $author = $doc->createElement("author");
991 XML::addElement($doc, $author, "uri", $owner["url"]);
992 XML::addElement($doc, $author, "name", $owner["nick"]);
993 XML::addElement($doc, $author, "email", $owner["addr"]);
999 * Adds an entry element to the XML document
1001 * @param DOMDocument $doc XML document
1002 * @param array $item Data of the item that is to be posted
1003 * @param array $owner Contact data of the poster
1004 * @param bool $toplevel optional default false
1006 * @return \DOMElement Entry element
1007 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1008 * @throws \ImagickException
1010 private static function entry(DOMDocument $doc, array $item, array $owner)
1014 $repeated_guid = OStatus::getResharedGuid($item);
1015 if ($repeated_guid != "") {
1016 $xml = self::reshareEntry($doc, $item, $owner, $repeated_guid);
1023 return self::noteEntry($doc, $item, $owner);
1027 * Adds an entry element with reshared content
1029 * @param DOMDocument $doc XML document
1030 * @param array $item Data of the item that is to be posted
1031 * @param array $owner Contact data of the poster
1032 * @param string $repeated_guid guid
1033 * @param bool $toplevel Is it for en entry element (false) or a feed entry (true)?
1035 * @return bool Entry element
1036 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1037 * @throws \ImagickException
1039 private static function reshareEntry(DOMDocument $doc, array $item, array $owner, $repeated_guid)
1041 if (($item['gravity'] != GRAVITY_PARENT) && (Strings::normaliseLink($item["author-link"]) != Strings::normaliseLink($owner["url"]))) {
1042 Logger::info('Feed entry author does not match feed owner', ['owner' => $owner["url"], 'author' => $item["author-link"]]);
1045 $entry = OStatus::entryHeader($doc, $owner, $item, false);
1047 $condition = ['uid' => $owner["uid"], 'guid' => $repeated_guid, 'private' => [Item::PUBLIC, Item::UNLISTED],
1048 'network' => Protocol::FEDERATED];
1049 $repeated_item = Post::selectFirst(Item::DELIVER_FIELDLIST, $condition);
1050 if (!DBA::isResult($repeated_item)) {
1054 self::entryContent($doc, $entry, $item, self::getTitle($repeated_item), Activity::SHARE, false);
1056 self::entryFooter($doc, $entry, $item, $owner);
1062 * Adds a regular entry element
1064 * @param DOMDocument $doc XML document
1065 * @param array $item Data of the item that is to be posted
1066 * @param array $owner Contact data of the poster
1067 * @param bool $toplevel Is it for en entry element (false) or a feed entry (true)?
1069 * @return \DOMElement Entry element
1070 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1071 * @throws \ImagickException
1073 private static function noteEntry(DOMDocument $doc, array $item, array $owner)
1075 if (($item['gravity'] != GRAVITY_PARENT) && (Strings::normaliseLink($item["author-link"]) != Strings::normaliseLink($owner["url"]))) {
1076 Logger::info('Feed entry author does not match feed owner', ['owner' => $owner["url"], 'author' => $item["author-link"]]);
1079 $entry = OStatus::entryHeader($doc, $owner, $item, false);
1081 self::entryContent($doc, $entry, $item, self::getTitle($item), '', true);
1083 self::entryFooter($doc, $entry, $item, $owner);
1089 * Adds elements to the XML document
1091 * @param DOMDocument $doc XML document
1092 * @param \DOMElement $entry Entry element where the content is added
1093 * @param array $item Data of the item that is to be posted
1094 * @param array $owner Contact data of the poster
1095 * @param string $title Title for the post
1096 * @param string $verb The activity verb
1097 * @param bool $complete Add the "status_net" element?
1098 * @param bool $feed_mode Behave like a regular feed for users if true
1100 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1102 private static function entryContent(DOMDocument $doc, \DOMElement $entry, array $item, $title, $verb = "", $complete = true)
1105 $verb = OStatus::constructVerb($item);
1108 XML::addElement($doc, $entry, "id", $item["uri"]);
1109 XML::addElement($doc, $entry, "title", html_entity_decode($title, ENT_QUOTES, 'UTF-8'));
1111 $body = OStatus::formatPicturePost($item['body']);
1113 $body = BBCode::convert($body, false, BBCode::OSTATUS);
1115 XML::addElement($doc, $entry, "content", $body, ["type" => "html"]);
1117 XML::addElement($doc, $entry, "link", "", ["rel" => "alternate", "type" => "text/html",
1118 "href" => DI::baseUrl()."/display/".$item["guid"]]
1121 XML::addElement($doc, $entry, "published", DateTimeFormat::utc($item["created"]."+00:00", DateTimeFormat::ATOM));
1122 XML::addElement($doc, $entry, "updated", DateTimeFormat::utc($item["edited"]."+00:00", DateTimeFormat::ATOM));
1126 * Adds the elements at the foot of an entry to the XML document
1128 * @param DOMDocument $doc XML document
1129 * @param object $entry The entry element where the elements are added
1130 * @param array $item Data of the item that is to be posted
1131 * @param array $owner Contact data of the poster
1132 * @param bool $complete default true
1134 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1136 private static function entryFooter(DOMDocument $doc, $entry, array $item, array $owner)
1140 if ($item['gravity'] != GRAVITY_PARENT) {
1141 $parent = Post::selectFirst(['guid', 'author-link', 'owner-link'], ['id' => $item['parent']]);
1143 $thrparent = Post::selectFirst(['guid', 'author-link', 'owner-link', 'plink'], ['uid' => $owner["uid"], 'uri' => $item['thr-parent']]);
1145 if (DBA::isResult($thrparent)) {
1146 $mentioned[$thrparent["author-link"]] = $thrparent["author-link"];
1147 $mentioned[$thrparent["owner-link"]] = $thrparent["owner-link"];
1148 $parent_plink = $thrparent["plink"];
1150 $mentioned[$parent["author-link"]] = $parent["author-link"];
1151 $mentioned[$parent["owner-link"]] = $parent["owner-link"];
1152 $parent_plink = DI::baseUrl()."/display/".$parent["guid"];
1156 "ref" => $item['thr-parent'],
1157 "href" => $parent_plink];
1158 XML::addElement($doc, $entry, "thr:in-reply-to", "", $attributes);
1162 "href" => $parent_plink];
1163 XML::addElement($doc, $entry, "link", "", $attributes);
1166 // uri-id isn't present for follow entry pseudo-items
1167 $tags = Tag::getByURIId($item['uri-id'] ?? 0);
1168 foreach ($tags as $tag) {
1169 $mentioned[$tag['url']] = $tag['url'];
1172 foreach ($tags as $tag) {
1173 if ($tag['type'] == Tag::HASHTAG) {
1174 XML::addElement($doc, $entry, "category", "", ["term" => $tag['name']]);
1178 OStatus::getAttachment($doc, $entry, $item);
1182 * Fetch or create title for feed entry
1184 * @param array $item
1185 * @return string title
1187 private static function getTitle(array $item)
1189 if ($item['title'] != '') {
1190 return BBCode::convert($item['title'], false, BBCode::OSTATUS);
1193 // Fetch information about the post
1194 $siteinfo = BBCode::getAttachedData($item["body"]);
1195 if (isset($siteinfo["title"])) {
1196 return $siteinfo["title"];
1199 // If no bookmark is found then take the first line
1200 // Remove the share element before fetching the first line
1201 $title = trim(preg_replace("/\[share.*?\](.*?)\[\/share\]/ism","\n$1\n",$item['body']));
1203 $title = HTML::toPlaintext(BBCode::convert($title, false), 0, true)."\n";
1204 $pos = strpos($title, "\n");
1206 if (($pos == 0) || ($pos > 100)) {
1211 return substr($title, 0, $pos) . $trailer;