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\Database\DBA;
34 use Friendica\Model\Contact;
35 use Friendica\Model\Item;
36 use Friendica\Model\Post;
37 use Friendica\Model\Tag;
38 use Friendica\Model\User;
39 use Friendica\Util\DateTimeFormat;
40 use Friendica\Util\Network;
41 use Friendica\Util\ParseUrl;
42 use Friendica\Util\Strings;
43 use Friendica\Util\XML;
46 * This class contain functions to import feeds (RSS/RDF/Atom)
51 * Read a RSS/RDF/Atom feed and create an item entry for it
53 * @param string $xml The feed data
54 * @param array $importer The user record of the importer
55 * @param array $contact The contact record of the feed
57 * @return array Returns the header and the first item in dry run mode
58 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
60 public static function import($xml, array $importer = [], array $contact = [])
62 $dryRun = empty($importer) && empty($contact);
65 Logger::info("Test Atom/RSS feed");
67 Logger::info("Import Atom/RSS feed '" . $contact["name"] . "' (Contact " . $contact["id"] . ") for user " . $importer["uid"]);
71 Logger::info('XML is empty.');
75 if (!empty($contact['poll'])) {
76 $basepath = $contact['poll'];
77 } elseif (!empty($contact['url'])) {
78 $basepath = $contact['url'];
83 $doc = new DOMDocument();
84 @$doc->loadXML(trim($xml));
85 $xpath = new DOMXPath($doc);
86 $xpath->registerNamespace('atom', ActivityNamespace::ATOM1);
87 $xpath->registerNamespace('dc', "http://purl.org/dc/elements/1.1/");
88 $xpath->registerNamespace('content', "http://purl.org/rss/1.0/modules/content/");
89 $xpath->registerNamespace('rdf', "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
90 $xpath->registerNamespace('rss', "http://purl.org/rss/1.0/");
91 $xpath->registerNamespace('media', "http://search.yahoo.com/mrss/");
92 $xpath->registerNamespace('poco', ActivityNamespace::POCO);
98 if ($xpath->query('/rdf:RDF/rss:channel')->length > 0) {
99 $author["author-link"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:link/text()');
100 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:title/text()');
102 if (empty($author["author-name"])) {
103 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rdf:RDF/rss:channel/rss:description/text()');
105 $entries = $xpath->query('/rdf:RDF/rss:item');
109 if ($xpath->query('/atom:feed')->length > 0) {
110 $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']");
111 if (is_object($alternate)) {
112 foreach ($alternate AS $attribute) {
113 if ($attribute->name == "href") {
114 $author["author-link"] = $attribute->textContent;
119 if (empty($author["author-link"])) {
120 $self = XML::getFirstAttributes($xpath, "atom:link[@rel='self']");
121 if (is_object($self)) {
122 foreach ($self AS $attribute) {
123 if ($attribute->name == "href") {
124 $author["author-link"] = $attribute->textContent;
130 if (empty($author["author-link"])) {
131 $author["author-link"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:id/text()');
133 $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:logo/text()');
135 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:title/text()');
137 if (empty($author["author-name"])) {
138 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:subtitle/text()');
141 if (empty($author["author-name"])) {
142 $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:name/text()');
145 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:displayName/text()');
147 $author["author-name"] = $value;
151 $author["author-id"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:id/text()');
153 // See https://tools.ietf.org/html/rfc4287#section-3.2.2
154 $value = XML::getFirstNodeValue($xpath, 'atom:author/atom:uri/text()');
156 $author["author-link"] = $value;
159 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:preferredUsername/text()');
161 $author["author-nick"] = $value;
164 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:address/poco:formatted/text()');
166 $author["author-location"] = $value;
169 $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:note/text()');
171 $author["author-about"] = $value;
174 $avatar = XML::getFirstAttributes($xpath, "atom:author/atom:link[@rel='avatar']");
175 if (is_object($avatar)) {
176 foreach ($avatar AS $attribute) {
177 if ($attribute->name == "href") {
178 $author["author-avatar"] = $attribute->textContent;
184 $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:updated/text()');
186 $author["app"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:generator/text()');
188 $entries = $xpath->query('/atom:feed/atom:entry');
192 if ($xpath->query('/rss/channel')->length > 0) {
193 $author["author-link"] = XML::getFirstNodeValue($xpath, '/rss/channel/link/text()');
195 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/title/text()');
196 $author["author-avatar"] = XML::getFirstNodeValue($xpath, '/rss/channel/image/url/text()');
198 if (empty($author["author-name"])) {
199 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/copyright/text()');
202 if (empty($author["author-name"])) {
203 $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/description/text()');
206 $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/rss/channel/pubDate/text()');
208 $author["app"] = XML::getFirstNodeValue($xpath, '/rss/channel/generator/text()');
210 $entries = $xpath->query('/rss/channel/item');
214 $author["author-link"] = $contact["url"];
216 if (empty($author["author-name"])) {
217 $author["author-name"] = $contact["name"];
220 $author["author-avatar"] = $contact["thumb"];
222 $author["owner-link"] = $contact["url"];
223 $author["owner-name"] = $contact["name"];
224 $author["owner-avatar"] = $contact["thumb"];
228 $header["uid"] = $importer["uid"] ?? 0;
229 $header["network"] = Protocol::FEED;
231 $header["origin"] = 0;
232 $header["gravity"] = GRAVITY_PARENT;
233 $header["private"] = Item::PUBLIC;
234 $header["verb"] = Activity::POST;
235 $header["object-type"] = Activity\ObjectType::NOTE;
236 $header["post-type"] = Item::PT_ARTICLE;
238 $header["contact-id"] = $contact["id"] ?? 0;
240 if (!is_object($entries)) {
241 Logger::info("There are no entries in this feed.");
246 $creation_dates = [];
248 // Limit the number of items that are about to be fetched
249 $total_items = ($entries->length - 1);
250 $max_items = DI::config()->get('system', 'max_feed_items');
251 if (($max_items > 0) && ($total_items > $max_items)) {
252 $total_items = $max_items;
257 // Importing older entries first
258 for ($i = $total_items; $i >= 0; --$i) {
259 $entry = $entries->item($i);
261 $item = array_merge($header, $author);
263 $alternate = XML::getFirstAttributes($xpath, "atom:link[@rel='alternate']", $entry);
264 if (!is_object($alternate)) {
265 $alternate = XML::getFirstAttributes($xpath, "atom:link", $entry);
267 if (is_object($alternate)) {
268 foreach ($alternate AS $attribute) {
269 if ($attribute->name == "href") {
270 $item["plink"] = $attribute->textContent;
275 if (empty($item["plink"])) {
276 $item["plink"] = XML::getFirstNodeValue($xpath, 'link/text()', $entry);
279 if (empty($item["plink"])) {
280 $item["plink"] = XML::getFirstNodeValue($xpath, 'rss:link/text()', $entry);
283 $item["uri"] = XML::getFirstNodeValue($xpath, 'atom:id/text()', $entry);
285 if (empty($item["uri"])) {
286 $item["uri"] = XML::getFirstNodeValue($xpath, 'guid/text()', $entry);
289 if (empty($item["uri"])) {
290 $item["uri"] = $item["plink"];
293 // Add the base path if missing
294 $item["uri"] = Network::addBasePath($item["uri"], $basepath);
295 $item["plink"] = Network::addBasePath($item["plink"], $basepath);
297 $orig_plink = $item["plink"];
299 $item["plink"] = DI::httpRequest()->finalUrl($item["plink"]);
301 $item["title"] = XML::getFirstNodeValue($xpath, 'atom:title/text()', $entry);
303 if (empty($item["title"])) {
304 $item["title"] = XML::getFirstNodeValue($xpath, 'title/text()', $entry);
306 if (empty($item["title"])) {
307 $item["title"] = XML::getFirstNodeValue($xpath, 'rss:title/text()', $entry);
310 $item["title"] = html_entity_decode($item["title"], ENT_QUOTES, 'UTF-8');
312 $published = XML::getFirstNodeValue($xpath, 'atom:published/text()', $entry);
314 if (empty($published)) {
315 $published = XML::getFirstNodeValue($xpath, 'pubDate/text()', $entry);
318 if (empty($published)) {
319 $published = XML::getFirstNodeValue($xpath, 'dc:date/text()', $entry);
322 $updated = XML::getFirstNodeValue($xpath, 'atom:updated/text()', $entry);
324 if (empty($updated) && !empty($published)) {
325 $updated = $published;
328 if (empty($published) && !empty($updated)) {
329 $published = $updated;
332 if ($published != "") {
333 $item["created"] = $published;
336 if ($updated != "") {
337 $item["edited"] = $updated;
341 $condition = ["`uid` = ? AND `uri` = ? AND `network` IN (?, ?)",
342 $importer["uid"], $item["uri"], Protocol::FEED, Protocol::DFRN];
343 $previous = Post::selectFirst(['id', 'created'], $condition);
344 if (DBA::isResult($previous)) {
345 // Use the creation date when the post had been stored. It can happen this date changes in the feed.
346 $creation_dates[] = $previous['created'];
347 Logger::info("Item with uri " . $item["uri"] . " for user " . $importer["uid"] . " already existed under id " . $previous["id"]);
350 $creation_dates[] = DateTimeFormat::utc($item['created']);
353 $creator = XML::getFirstNodeValue($xpath, 'author/text()', $entry);
355 if (empty($creator)) {
356 $creator = XML::getFirstNodeValue($xpath, 'atom:author/atom:name/text()', $entry);
359 if (empty($creator)) {
360 $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
363 if ($creator != "") {
364 $item["author-name"] = $creator;
367 $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry);
369 if ($creator != "") {
370 $item["author-name"] = $creator;
374 // <category>Ausland</category>
375 // <media:thumbnail width="152" height="76" url="http://www.taz.de/picture/667875/192/14388767.jpg"/>
379 $enclosures = $xpath->query("enclosure|atom:link[@rel='enclosure']", $entry);
380 foreach ($enclosures AS $enclosure) {
385 foreach ($enclosure->attributes AS $attribute) {
386 if (in_array($attribute->name, ["url", "href"])) {
387 $href = $attribute->textContent;
388 } elseif ($attribute->name == "length") {
389 $length = (int)$attribute->textContent;
390 } elseif ($attribute->name == "type") {
391 $type = $attribute->textContent;
396 $attachments[] = ['type' => Post\Media::DOCUMENT, 'url' => $href, 'mimetype' => $type, 'size' => $length];
401 $categories = $xpath->query("category", $entry);
402 foreach ($categories AS $category) {
403 $taglist[] = $category->nodeValue;
406 $body = trim(XML::getFirstNodeValue($xpath, 'atom:content/text()', $entry));
409 $body = trim(XML::getFirstNodeValue($xpath, 'content:encoded/text()', $entry));
412 $summary = trim(XML::getFirstNodeValue($xpath, 'atom:summary/text()', $entry));
414 if (empty($summary)) {
415 $summary = trim(XML::getFirstNodeValue($xpath, 'description/text()', $entry));
423 if ($body == $summary) {
427 // remove the content of the title if it is identically to the body
428 // This helps with auto generated titles e.g. from tumblr
429 if (self::titleIsBody($item["title"], $body)) {
432 $item["body"] = HTML::toBBCode($body, $basepath);
434 // Remove tracking pixels
435 $item["body"] = preg_replace("/\[img=1x1\]([^\[\]]*)\[\/img\]/Usi", '', $item["body"]);
437 if (($item["body"] == '') && ($item["title"] != '')) {
438 $item["body"] = $item["title"];
445 } elseif (!Item::isValid($item)) {
446 Logger::info('Feed item is invalid', ['created' => $item['created'], 'uid' => $item['uid'], 'uri' => $item['uri']]);
448 } elseif (Item::isTooOld($item)) {
449 Logger::info('Feed is too old', ['created' => $item['created'], 'uid' => $item['uid'], 'uri' => $item['uri']]);
454 if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] < 3)) {
455 // Handle enclosures and treat them as preview picture
456 foreach ($attachments AS $attachment) {
457 if ($attachment["mimetype"] == "image/jpeg") {
458 $preview = $attachment["url"];
462 // Remove a possible link to the item itself
463 $item["body"] = str_replace($item["plink"], '', $item["body"]);
464 $item["body"] = trim(preg_replace('/\[url\=\](\w+.*?)\[\/url\]/i', '', $item["body"]));
466 // Replace the content when the title is longer than the body
467 $replace = (strlen($item["title"]) > strlen($item["body"]));
469 // Replace it, when there is an image in the body
470 if (strstr($item["body"], '[/img]')) {
474 // Replace it, when there is a link in the body
475 if (strstr($item["body"], '[/url]')) {
479 $saved_body = $item["body"];
480 $saved_title = $item["title"];
483 $item["body"] = trim($item["title"]);
486 $data = ParseUrl::getSiteinfoCached($item['plink']);
487 if (!empty($data['text']) && !empty($data['title']) && (mb_strlen($item['body']) < mb_strlen($data['text']))) {
488 // When the fetched page info text is longer than the body, we do try to enhance the body
489 if (!empty($item['body']) && (strpos($data['title'], $item['body']) === false) && (strpos($data['text'], $item['body']) === false)) {
490 // The body is not part of the fetched page info title or page info text. So we add the text to the body
491 $item['body'] .= "\n\n" . $data['text'];
493 // Else we replace the body with the page info text
494 $item['body'] = $data['text'];
498 $data = PageInfo::queryUrl($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_denylist"] ?? '');
501 // Take the data that was provided by the feed if the query is empty
502 if (($data['type'] == 'link') && empty($data['title']) && empty($data['text'])) {
503 $data['title'] = $saved_title;
504 $item["body"] = $saved_body;
507 $data_text = strip_tags(trim($data['text'] ?? ''));
508 $item_body = strip_tags(trim($item['body'] ?? ''));
510 if (!empty($data_text) && (($data_text == $item_body) || strstr($item_body, $data_text))) {
514 // We always strip the title since it will be added in the page information
516 $item["body"] = $item["body"] . "\n" . PageInfo::getFooterFromData($data, false);
517 $taglist = $contact["fetch_further_information"] == 2 ? PageInfo::getTagsFromUrl($item["plink"], $preview, $contact["ffi_keyword_denylist"] ?? '') : [];
518 $item["object-type"] = Activity\ObjectType::BOOKMARK;
522 if (!empty($summary)) {
523 $item["body"] = '[abstract]' . HTML::toBBCode($summary, $basepath) . "[/abstract]\n" . $item["body"];
526 if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] == 3)) {
527 if (empty($taglist)) {
528 $taglist = PageInfo::getTagsFromUrl($item["plink"], $preview, $contact["ffi_keyword_denylist"] ?? '');
530 $item["body"] .= "\n" . self::tagToString($taglist);
535 // Add the link to the original feed entry if not present in feed
536 if (($item['plink'] != '') && !strstr($item["body"], $item['plink'])) {
537 $item["body"] .= "[hr][url]" . $item['plink'] . "[/url]";
541 if (empty($item['title'])) {
542 $item['post-type'] = Item::PT_NOTE;
545 Logger::info('Stored feed', ['item' => $item]);
547 $notify = Item::isRemoteSelf($contact, $item);
549 // Distributed items should have a well formatted URI.
550 // Additionally we have to avoid conflicts with identical URI between imported feeds and these items.
552 $item['guid'] = Item::guidFromUri($orig_plink, DI::baseUrl()->getHostname());
553 $item['uri'] = Item::newURI($item['uid'], $item['guid']);
554 unset($item['thr-parent']);
555 unset($item['parent-uri']);
557 // Set the delivery priority for "remote self" to "medium"
558 $notify = PRIORITY_MEDIUM;
561 $condition = ['uid' => $item['uid'], 'uri' => $item['uri']];
562 if (!Post::exists($condition) && !Post\Delayed::exists($item["uri"], $item['uid'])) {
564 Post\Delayed::publish($item, $notify, $taglist, $attachments);
566 $postings[] = ['item' => $item, 'notify' => $notify,
567 'taglist' => $taglist, 'attachments' => $attachments];
570 Logger::info('Post already created or exists in the delayed posts queue', ['uid' => $item['uid'], 'uri' => $item["uri"]]);
574 if (!empty($postings)) {
575 $min_posting = DI::config()->get('system', 'minimum_posting_interval', 0);
576 $total = count($postings);
578 // Posts shouldn't be delayed more than a day
579 $interval = min(1440, self::getPollInterval($contact));
580 $delay = max(round(($interval * 60) / $total), 60 * $min_posting);
581 Logger::info('Got posting delay', ['delay' => $delay, 'interval' => $interval, 'items' => $total, 'cid' => $contact['id'], 'url' => $contact['url']]);
588 foreach ($postings as $posting) {
590 $publish_time = time() + $post_delay;
591 $post_delay += $delay;
593 $publish_time = time();
596 $last_publish = DI::pConfig()->get($posting['item']['uid'], 'system', 'last_publish', 0, true);
597 $next_publish = max($last_publish + (60 * $min_posting), time());
598 if ($publish_time < $next_publish) {
599 $publish_time = $next_publish;
601 $publish_at = date(DateTimeFormat::MYSQL, $publish_time);
603 Post\Delayed::add($posting['item']['uri'], $posting['item'], $posting['notify'], false, $publish_at, $posting['taglist'], $posting['attachments']);
607 if (!$dryRun && DI::config()->get('system', 'adjust_poll_frequency')) {
608 self::adjustPollFrequency($contact, $creation_dates);
611 return ["header" => $author, "items" => $items];
615 * Automatically adjust the poll frequency according to the post frequency
617 * @param array $contact
618 * @param array $creation_dates
621 private static function adjustPollFrequency(array $contact, array $creation_dates)
623 if ($contact['network'] != Protocol::FEED) {
624 Logger::info('Contact is no feed, skip.', ['id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url'], 'network' => $contact['network']]);
628 if (!empty($creation_dates)) {
629 // Count the post frequency and the earliest and latest post date
633 $oldest_date = $newest_date = '';
635 foreach ($creation_dates as $date) {
636 $timestamp = strtotime($date);
637 $day = intdiv($timestamp, 86400);
638 $hour = $timestamp % 86400;
640 // Only have a look at values from the last seven days
641 if (((time() / 86400) - $day) < 7) {
642 if (empty($frequency[$day])) {
643 $frequency[$day] = ['count' => 1, 'low' => $hour, 'high' => $hour];
645 ++$frequency[$day]['count'];
646 if ($frequency[$day]['low'] > $hour) {
647 $frequency[$day]['low'] = $hour;
649 if ($frequency[$day]['high'] < $hour) {
650 $frequency[$day]['high'] = $hour;
654 if ($oldest > $day) {
656 $oldest_date = $date;
659 if ($newest < $day) {
661 $newest_date = $date;
665 if (count($creation_dates) == 1) {
666 Logger::info('Feed had posted a single time, switching to daily polling', ['newest' => $newest_date, 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
667 $priority = 8; // Poll once a day
670 if (empty($priority) && (((time() / 86400) - $newest) > 730)) {
671 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']]);
672 $priority = 10; // Poll every month
675 if (empty($priority) && (((time() / 86400) - $newest) > 365)) {
676 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']]);
677 $priority = 9; // Poll every week
680 if (empty($priority) && empty($frequency)) {
681 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']]);
682 $priority = 8; // Poll once a day
685 if (empty($priority)) {
686 // Calculate the highest "posts per day" value
688 foreach ($frequency as $entry) {
689 if (($entry['count'] == 1) || ($entry['high'] == $entry['low'])) {
693 // We take the earliest and latest post day and interpolate the number of post per day
694 // that would had been created with this post frequency
696 // Assume at least four hours between oldest and newest post per day - should be okay for news outlets
697 $duration = max($entry['high'] - $entry['low'], 14400);
698 $ppd = (86400 / $duration) * $entry['count'];
704 $priority = 1; // Poll every quarter hour
705 } elseif ($max > 24) {
706 $priority = 2; // Poll half an hour
707 } elseif ($max > 12) {
708 $priority = 3; // Poll hourly
709 } elseif ($max > 8) {
710 $priority = 4; // Poll every two hours
711 } elseif ($max > 4) {
712 $priority = 5; // Poll every three hours
713 } elseif ($max > 2) {
714 $priority = 6; // Poll every six hours
716 $priority = 7; // Poll twice a day
718 Logger::info('Calculated priority by the posts per day', ['priority' => $priority, 'max' => round($max, 2), 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
721 Logger::info('No posts, switching to daily polling', ['id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
722 $priority = 8; // Poll once a day
725 if ($contact['rating'] != $priority) {
726 Logger::notice('Adjusting priority', ['old' => $contact['rating'], 'new' => $priority, 'id' => $contact['id'], 'uid' => $contact['uid'], 'url' => $contact['url']]);
727 DBA::update('contact', ['rating' => $priority], ['id' => $contact['id']]);
732 * Get the poll interval for the given contact array
734 * @param array $contact
735 * @return int Poll interval in minutes
737 public static function getPollInterval(array $contact)
739 if (in_array($contact['network'], [Protocol::MAIL, Protocol::FEED])) {
740 $ratings = [0, 3, 7, 8, 9, 10];
741 if (DI::config()->get('system', 'adjust_poll_frequency') && ($contact['network'] == Protocol::FEED)) {
742 $rating = $contact['rating'];
743 } elseif (array_key_exists($contact['priority'], $ratings)) {
744 $rating = $ratings[$contact['priority']];
749 // Check once a week per default for all other networks
753 // Friendica and OStatus are checked once a day
754 if (in_array($contact['network'], [Protocol::DFRN, Protocol::OSTATUS])) {
758 // Check archived contacts or contacts with unsupported protocols once a month
759 if ($contact['archive'] || in_array($contact['network'], [Protocol::ZOT, Protocol::PHANTOM])) {
767 * Based on $contact['priority'], should we poll this site now? Or later?
770 $min_poll_interval = max(1, DI::config()->get('system', 'min_poll_interval'));
772 $poll_intervals = [$min_poll_interval, 15, 30, 60, 120, 180, 360, 720 ,1440, 10080, 43200];
774 //$poll_intervals = [$min_poll_interval . ' minute', '15 minute', '30 minute',
775 // '1 hour', '2 hour', '3 hour', '6 hour', '12 hour' ,'1 day', '1 week', '1 month'];
777 return $poll_intervals[$rating];
781 * Convert a tag array to a tag string
784 * @return string tag string
786 private static function tagToString(array $tags)
790 foreach ($tags as $tag) {
795 $tagstr .= "#[url=" . DI::baseUrl() . "/search?tag=" . urlencode($tag) . "]" . $tag . "[/url]";
801 private static function titleIsBody($title, $body)
803 $title = strip_tags($title);
804 $title = trim($title);
805 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
806 $title = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $title);
808 $body = strip_tags($body);
810 $body = html_entity_decode($body, ENT_QUOTES, 'UTF-8');
811 $body = str_replace(["\n", "\r", "\t", " "], ["", "", "", ""], $body);
813 if (strlen($title) < strlen($body)) {
814 $body = substr($body, 0, strlen($title));
817 if (($title != $body) && (substr($title, -3) == "...")) {
818 $pos = strrpos($title, "...");
820 $title = substr($title, 0, $pos);
821 $body = substr($body, 0, $pos);
824 return ($title == $body);
828 * Creates the Atom feed for a given nickname
831 * - activity (default): all the public posts
832 * - posts: all the public top-level posts
833 * - comments: all the public replies
835 * Updates the provided last_update parameter if the result comes from the
836 * cache or it is empty
838 * @param string $owner_nick Nickname of the feed owner
839 * @param string $last_update Date of the last update
840 * @param integer $max_items Number of maximum items to fetch
841 * @param string $filter Feed items filter (activity, posts or comments)
842 * @param boolean $nocache Wether to bypass caching
844 * @return string Atom feed
845 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
846 * @throws \ImagickException
848 public static function atom($owner_nick, $last_update, $max_items = 300, $filter = 'activity', $nocache = false)
850 $stamp = microtime(true);
852 $owner = User::getOwnerDataByNick($owner_nick);
857 $cachekey = "feed:feed:" . $owner_nick . ":" . $filter . ":" . $last_update;
859 $previous_created = $last_update;
861 // Don't cache when the last item was posted less then 15 minutes ago (Cache duration)
862 if ((time() - strtotime($owner['last-item'])) < 15*60) {
863 $result = DI::cache()->get($cachekey);
864 if (!$nocache && !is_null($result)) {
865 Logger::info('Cached feed duration', ['seconds' => number_format(microtime(true) - $stamp, 3), 'nick' => $owner_nick, 'filter' => $filter, 'created' => $previous_created]);
866 return $result['feed'];
870 $check_date = empty($last_update) ? '' : DateTimeFormat::utc($last_update);
871 $authorid = Contact::getIdForURL($owner["url"]);
873 $condition = ["`uid` = ? AND `received` > ? AND NOT `deleted` AND `gravity` IN (?, ?)
874 AND `private` != ? AND `visible` AND `wall` AND `parent-network` IN (?, ?, ?, ?)",
875 $owner["uid"], $check_date, GRAVITY_PARENT, GRAVITY_COMMENT,
876 Item::PRIVATE, Protocol::ACTIVITYPUB,
877 Protocol::OSTATUS, Protocol::DFRN, Protocol::DIASPORA];
879 if ($filter === 'comments') {
880 $condition[0] .= " AND `object-type` = ? ";
881 $condition[] = Activity\ObjectType::COMMENT;
884 if ($owner['account-type'] != User::ACCOUNT_TYPE_COMMUNITY) {
885 $condition[0] .= " AND `contact-id` = ? AND `author-id` = ?";
886 $condition[] = $owner["id"];
887 $condition[] = $authorid;
890 $params = ['order' => ['received' => true], 'limit' => $max_items];
892 if ($filter === 'posts') {
893 $ret = Post::selectThread(Item::DELIVER_FIELDLIST, $condition, $params);
895 $ret = Post::select(Item::DELIVER_FIELDLIST, $condition, $params);
898 $items = Post::toArray($ret);
900 $doc = new DOMDocument('1.0', 'utf-8');
901 $doc->formatOutput = true;
903 $root = self::addHeader($doc, $owner, $filter);
905 foreach ($items as $item) {
906 $entry = self::entry($doc, $item, $owner);
907 $root->appendChild($entry);
909 if ($last_update < $item['created']) {
910 $last_update = $item['created'];
914 $feeddata = trim($doc->saveXML());
916 $msg = ['feed' => $feeddata, 'last_update' => $last_update];
917 DI::cache()->set($cachekey, $msg, Duration::QUARTER_HOUR);
919 Logger::info('Feed duration', ['seconds' => number_format(microtime(true) - $stamp, 3), 'nick' => $owner_nick, 'filter' => $filter, 'created' => $previous_created]);
925 * Adds the header elements to the XML document
927 * @param DOMDocument $doc XML document
928 * @param array $owner Contact data of the poster
929 * @param string $filter The related feed filter (activity, posts or comments)
931 * @return object header root element
932 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
934 private static function addHeader(DOMDocument $doc, array $owner, $filter)
936 $root = $doc->createElementNS(ActivityNamespace::ATOM1, 'feed');
937 $doc->appendChild($root);
940 $selfUri = '/feed/' . $owner["nick"] . '/';
943 $title = DI::l10n()->t('%s\'s timeline', $owner['name']);
947 $title = DI::l10n()->t('%s\'s posts', $owner['name']);
950 $title = DI::l10n()->t('%s\'s comments', $owner['name']);
955 $attributes = ["uri" => "https://friendi.ca", "version" => FRIENDICA_VERSION . "-" . DB_UPDATE_VERSION];
956 XML::addElement($doc, $root, "generator", FRIENDICA_PLATFORM, $attributes);
957 XML::addElement($doc, $root, "id", DI::baseUrl() . "/profile/" . $owner["nick"]);
958 XML::addElement($doc, $root, "title", $title);
959 XML::addElement($doc, $root, "subtitle", sprintf("Updates from %s on %s", $owner["name"], DI::config()->get('config', 'sitename')));
960 XML::addElement($doc, $root, "logo", $owner["photo"]);
961 XML::addElement($doc, $root, "updated", DateTimeFormat::utcNow(DateTimeFormat::ATOM));
963 $author = self::addAuthor($doc, $owner);
964 $root->appendChild($author);
966 $attributes = ["href" => $owner["url"], "rel" => "alternate", "type" => "text/html"];
967 XML::addElement($doc, $root, "link", "", $attributes);
969 OStatus::hublinks($doc, $root, $owner["nick"]);
971 $attributes = ["href" => DI::baseUrl() . $selfUri, "rel" => "self", "type" => "application/atom+xml"];
972 XML::addElement($doc, $root, "link", "", $attributes);
978 * Adds the author element to the XML document
980 * @param DOMDocument $doc XML document
981 * @param array $owner Contact data of the poster
983 * @return \DOMElement author element
984 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
986 private static function addAuthor(DOMDocument $doc, array $owner)
988 $author = $doc->createElement("author");
989 XML::addElement($doc, $author, "uri", $owner["url"]);
990 XML::addElement($doc, $author, "name", $owner["nick"]);
991 XML::addElement($doc, $author, "email", $owner["addr"]);
997 * Adds an entry element to the XML document
999 * @param DOMDocument $doc XML document
1000 * @param array $item Data of the item that is to be posted
1001 * @param array $owner Contact data of the poster
1002 * @param bool $toplevel optional default false
1004 * @return \DOMElement Entry element
1005 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1006 * @throws \ImagickException
1008 private static function entry(DOMDocument $doc, array $item, array $owner)
1012 $repeated_guid = OStatus::getResharedGuid($item);
1013 if ($repeated_guid != "") {
1014 $xml = self::reshareEntry($doc, $item, $owner, $repeated_guid);
1021 return self::noteEntry($doc, $item, $owner);
1025 * Adds an entry element with reshared content
1027 * @param DOMDocument $doc XML document
1028 * @param array $item Data of the item that is to be posted
1029 * @param array $owner Contact data of the poster
1030 * @param string $repeated_guid guid
1031 * @param bool $toplevel Is it for en entry element (false) or a feed entry (true)?
1033 * @return bool Entry element
1034 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1035 * @throws \ImagickException
1037 private static function reshareEntry(DOMDocument $doc, array $item, array $owner, $repeated_guid)
1039 if (($item['gravity'] != GRAVITY_PARENT) && (Strings::normaliseLink($item["author-link"]) != Strings::normaliseLink($owner["url"]))) {
1040 Logger::info('Feed entry author does not match feed owner', ['owner' => $owner["url"], 'author' => $item["author-link"]]);
1043 $entry = OStatus::entryHeader($doc, $owner, $item, false);
1045 $condition = ['uid' => $owner["uid"], 'guid' => $repeated_guid, 'private' => [Item::PUBLIC, Item::UNLISTED],
1046 'network' => Protocol::FEDERATED];
1047 $repeated_item = Post::selectFirst(Item::DELIVER_FIELDLIST, $condition);
1048 if (!DBA::isResult($repeated_item)) {
1052 self::entryContent($doc, $entry, $item, self::getTitle($repeated_item), Activity::SHARE, false);
1054 self::entryFooter($doc, $entry, $item, $owner);
1060 * Adds a regular entry element
1062 * @param DOMDocument $doc XML document
1063 * @param array $item Data of the item that is to be posted
1064 * @param array $owner Contact data of the poster
1065 * @param bool $toplevel Is it for en entry element (false) or a feed entry (true)?
1067 * @return \DOMElement Entry element
1068 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1069 * @throws \ImagickException
1071 private static function noteEntry(DOMDocument $doc, array $item, array $owner)
1073 if (($item['gravity'] != GRAVITY_PARENT) && (Strings::normaliseLink($item["author-link"]) != Strings::normaliseLink($owner["url"]))) {
1074 Logger::info('Feed entry author does not match feed owner', ['owner' => $owner["url"], 'author' => $item["author-link"]]);
1077 $entry = OStatus::entryHeader($doc, $owner, $item, false);
1079 self::entryContent($doc, $entry, $item, self::getTitle($item), '', true);
1081 self::entryFooter($doc, $entry, $item, $owner);
1087 * Adds elements to the XML document
1089 * @param DOMDocument $doc XML document
1090 * @param \DOMElement $entry Entry element where the content is added
1091 * @param array $item Data of the item that is to be posted
1092 * @param array $owner Contact data of the poster
1093 * @param string $title Title for the post
1094 * @param string $verb The activity verb
1095 * @param bool $complete Add the "status_net" element?
1096 * @param bool $feed_mode Behave like a regular feed for users if true
1098 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1100 private static function entryContent(DOMDocument $doc, \DOMElement $entry, array $item, $title, $verb = "", $complete = true)
1103 $verb = OStatus::constructVerb($item);
1106 XML::addElement($doc, $entry, "id", $item["uri"]);
1107 XML::addElement($doc, $entry, "title", html_entity_decode($title, ENT_QUOTES, 'UTF-8'));
1109 $body = OStatus::formatPicturePost($item['body']);
1111 $body = BBCode::convert($body, false, BBCode::OSTATUS);
1113 XML::addElement($doc, $entry, "content", $body, ["type" => "html"]);
1115 XML::addElement($doc, $entry, "link", "", ["rel" => "alternate", "type" => "text/html",
1116 "href" => DI::baseUrl()."/display/".$item["guid"]]
1119 XML::addElement($doc, $entry, "published", DateTimeFormat::utc($item["created"]."+00:00", DateTimeFormat::ATOM));
1120 XML::addElement($doc, $entry, "updated", DateTimeFormat::utc($item["edited"]."+00:00", DateTimeFormat::ATOM));
1124 * Adds the elements at the foot of an entry to the XML document
1126 * @param DOMDocument $doc XML document
1127 * @param object $entry The entry element where the elements are added
1128 * @param array $item Data of the item that is to be posted
1129 * @param array $owner Contact data of the poster
1130 * @param bool $complete default true
1132 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
1134 private static function entryFooter(DOMDocument $doc, $entry, array $item, array $owner)
1138 if ($item['gravity'] != GRAVITY_PARENT) {
1139 $parent = Post::selectFirst(['guid', 'author-link', 'owner-link'], ['id' => $item['parent']]);
1141 $thrparent = Post::selectFirst(['guid', 'author-link', 'owner-link', 'plink'], ['uid' => $owner["uid"], 'uri' => $item['thr-parent']]);
1143 if (DBA::isResult($thrparent)) {
1144 $mentioned[$thrparent["author-link"]] = $thrparent["author-link"];
1145 $mentioned[$thrparent["owner-link"]] = $thrparent["owner-link"];
1146 $parent_plink = $thrparent["plink"];
1148 $mentioned[$parent["author-link"]] = $parent["author-link"];
1149 $mentioned[$parent["owner-link"]] = $parent["owner-link"];
1150 $parent_plink = DI::baseUrl()."/display/".$parent["guid"];
1154 "ref" => $item['thr-parent'],
1155 "href" => $parent_plink];
1156 XML::addElement($doc, $entry, "thr:in-reply-to", "", $attributes);
1160 "href" => $parent_plink];
1161 XML::addElement($doc, $entry, "link", "", $attributes);
1164 // uri-id isn't present for follow entry pseudo-items
1165 $tags = Tag::getByURIId($item['uri-id'] ?? 0);
1166 foreach ($tags as $tag) {
1167 $mentioned[$tag['url']] = $tag['url'];
1170 foreach ($tags as $tag) {
1171 if ($tag['type'] == Tag::HASHTAG) {
1172 XML::addElement($doc, $entry, "category", "", ["term" => $tag['name']]);
1176 OStatus::getAttachment($doc, $entry, $item);
1180 * Fetch or create title for feed entry
1182 * @param array $item
1183 * @return string title
1185 private static function getTitle(array $item)
1187 if ($item['title'] != '') {
1188 return BBCode::convert($item['title'], false, BBCode::OSTATUS);
1191 // Fetch information about the post
1192 $siteinfo = BBCode::getAttachedData($item["body"]);
1193 if (isset($siteinfo["title"])) {
1194 return $siteinfo["title"];
1197 // If no bookmark is found then take the first line
1198 // Remove the share element before fetching the first line
1199 $title = trim(preg_replace("/\[share.*?\](.*?)\[\/share\]/ism","\n$1\n",$item['body']));
1201 $title = HTML::toPlaintext(BBCode::convert($title, false), 0, true)."\n";
1202 $pos = strpos($title, "\n");
1204 if (($pos == 0) || ($pos > 100)) {
1209 return substr($title, 0, $pos) . $trailer;