/**
*
* consume_feed - process atom feed and update anything/everything we might need to update
*
* $xml = the (atom) feed to consume - RSS isn't as fully supported but may work for simple feeds.
*
* $importer = the contact_record (joined to user_record) of the local user who owns this relationship.
* It is this person's stuff that is going to be updated.
* $contact = the person who is sending us stuff. If not set, we MAY be processing a "follow" activity
* from an external network and MAY create an appropriate contact record. Otherwise, we MUST
* have a contact record.
* $hub = should we find a hub declation in the feed, pass it back to our calling process, who might (or
* might not) try and subscribe to it.
* $datedir sorts in reverse order
* $pass - by default ($pass = 0) we cannot guarantee that a parent item has been
* imported prior to its children being seen in the stream unless we are certain
* of how the feed is arranged/ordered.
* With $pass = 1, we only pull parent items out of the stream.
* With $pass = 2, we only pull children (comments/likes).
*
* So running this twice, first with pass 1 and then with pass 2 will do the right
* thing regardless of feed ordering. This won't be adequate in a fully-threaded
* model where comments can have sub-threads. That would require some massive sorting
* to get all the feed items into a mostly linear ordering, and might still require
* recursion.
*/
function consume_feed($xml, $importer, &$contact, &$hub, $datedir = 0, $pass = 0)
{
if ($contact['network'] === NETWORK_OSTATUS) {
if ($pass < 2) {
// Test - remove before flight
//$tempfile = tempnam(get_temppath(), "ostatus2");
//file_put_contents($tempfile, $xml);
logger("Consume OStatus messages ", LOGGER_DEBUG);
ostatus_import($xml, $importer, $contact, $hub);
}
return;
}
if ($contact['network'] === NETWORK_FEED) {
if ($pass < 2) {
logger("Consume feeds", LOGGER_DEBUG);
feed_import($xml, $importer, $contact, $hub);
}
return;
}
require_once 'library/simplepie/simplepie.inc';
require_once 'include/contact_selectors.php';
if (!strlen($xml)) {
logger('consume_feed: empty input');
return;
}
$feed = new SimplePie();
$feed->set_raw_data($xml);
if ($datedir) {
$feed->enable_order_by_date(true);
} else {
$feed->enable_order_by_date(false);
}
$feed->init();
if ($feed->error()) {
logger('consume_feed: Error parsing XML: ' . $feed->error());
}
$permalink = $feed->get_permalink();
// Check at the feed level for updated contact name and/or photo
$name_updated = '';
$new_name = '';
$photo_timestamp = '';
$photo_url = '';
$birthday = '';
$contact_updated = '';
$hubs = $feed->get_links('hub');
logger('consume_feed: hubs: ' . print_r($hubs, true), LOGGER_DATA);
if (count($hubs)) {
$hub = implode(',', $hubs);
}
$rawtags = $feed->get_feed_tags(NAMESPACE_DFRN, 'owner');
if (!$rawtags) {
$rawtags = $feed->get_feed_tags(SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
}
if ($rawtags) {
$elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10];
if ($elems['name'][0]['attribs'][NAMESPACE_DFRN]['updated']) {
$name_updated = $elems['name'][0]['attribs'][NAMESPACE_DFRN]['updated'];
$new_name = $elems['name'][0]['data'];
// Manually checking for changed contact names
if ($new_name != $contact['name'] and $new_name != "" and $name_updated <= $contact['name-date']) {
$name_updated = date("c");
$photo_timestamp = date("c");
}
}
if (x($elems, 'link') && $elems['link'][0]['attribs']['']['rel'] === 'photo' && $elems['link'][0]['attribs'][NAMESPACE_DFRN]['updated']) {
if ($photo_timestamp == "") {
$photo_timestamp = datetime_convert('UTC', 'UTC', $elems['link'][0]['attribs'][NAMESPACE_DFRN]['updated']);
}
$photo_url = $elems['link'][0]['attribs']['']['href'];
}
if (x($rawtags[0]['child'], NAMESPACE_DFRN) && x($rawtags[0]['child'][NAMESPACE_DFRN], 'birthday')) {
$birthday = datetime_convert('UTC', 'UTC', $rawtags[0]['child'][NAMESPACE_DFRN]['birthday'][0]['data']);
}
}
if (is_array($contact) && $photo_timestamp && strlen($photo_url) && $photo_timestamp > $contact['avatar-date']) {
logger('consume_feed: Updating photo for ' . $contact['name'] . ' from ' . $photo_url . ' uid: ' . $contact['uid']);
$contact_updated = $photo_timestamp;
require_once "include/Photo.php";
$photos = import_profile_photo($photo_url, $contact['uid'], $contact['id']);
q("UPDATE `contact` SET `avatar-date` = '%s', `photo` = '%s', `thumb` = '%s', `micro` = '%s'\n\t\t\tWHERE `uid` = %d AND `id` = %d AND NOT `self`", dbesc(datetime_convert()), dbesc($photos[0]), dbesc($photos[1]), dbesc($photos[2]), intval($contact['uid']), intval($contact['id']));
}
if (is_array($contact) && $name_updated && strlen($new_name) && $name_updated > $contact['name-date']) {
if ($name_updated > $contact_updated) {
$contact_updated = $name_updated;
}
$r = q("SELECT * FROM `contact` WHERE `uid` = %d AND `id` = %d LIMIT 1", intval($contact['uid']), intval($contact['id']));
$x = q("UPDATE `contact` SET `name` = '%s', `name-date` = '%s' WHERE `uid` = %d AND `id` = %d AND `name` != '%s' AND NOT `self`", dbesc(notags(trim($new_name))), dbesc(datetime_convert()), intval($contact['uid']), intval($contact['id']), dbesc(notags(trim($new_name))));
// do our best to update the name on content items
if (count($r) and notags(trim($new_name)) != $r[0]['name']) {
q("UPDATE `item` SET `author-name` = '%s' WHERE `author-name` = '%s' AND `author-link` = '%s' AND `uid` = %d AND `author-name` != '%s'", dbesc(notags(trim($new_name))), dbesc($r[0]['name']), dbesc($r[0]['url']), intval($contact['uid']), dbesc(notags(trim($new_name))));
}
}
if ($contact_updated and $new_name and $photo_url) {
poco_check($contact['url'], $new_name, NETWORK_DFRN, $photo_url, "", "", "", "", "", $contact_updated, 2, $contact['id'], $contact['uid']);
}
if (strlen($birthday)) {
if (substr($birthday, 0, 4) != $contact['bdyear']) {
logger('consume_feed: updating birthday: ' . $birthday);
/**
*
* Add new birthday event for this person
*
* $bdtext is just a readable placeholder in case the event is shared
* with others. We will replace it during presentation to our $importer
* to contain a sparkle link and perhaps a photo.
*
*/
$bdtext = sprintf(t('%s\'s birthday'), $contact['name']);
$bdtext2 = sprintf(t('Happy Birthday %s'), ' [url=' . $contact['url'] . ']' . $contact['name'] . '[/url]');
$r = q("INSERT INTO `event` (`uid`,`cid`,`created`,`edited`,`start`,`finish`,`summary`,`desc`,`type`)\n\t\t\t\tVALUES ( %d, %d, '%s', '%s', '%s', '%s', '%s', '%s', '%s' ) ", intval($contact['uid']), intval($contact['id']), dbesc(datetime_convert()), dbesc(datetime_convert()), dbesc(datetime_convert('UTC', 'UTC', $birthday)), dbesc(datetime_convert('UTC', 'UTC', $birthday . ' + 1 day ')), dbesc($bdtext), dbesc($bdtext2), dbesc('birthday'));
// update bdyear
q("UPDATE `contact` SET `bdyear` = '%s' WHERE `uid` = %d AND `id` = %d", dbesc(substr($birthday, 0, 4)), intval($contact['uid']), intval($contact['id']));
// This function is called twice without reloading the contact
// Make sure we only create one event. This is why &$contact
// is a reference var in this function
$contact['bdyear'] = substr($birthday, 0, 4);
}
}
$community_page = 0;
$rawtags = $feed->get_feed_tags(NAMESPACE_DFRN, 'community');
if ($rawtags) {
$community_page = intval($rawtags[0]['data']);
}
if (is_array($contact) && intval($contact['forum']) != $community_page) {
q("update contact set forum = %d where id = %d", intval($community_page), intval($contact['id']));
$contact['forum'] = (string) $community_page;
}
// process any deleted entries
$del_entries = $feed->get_feed_tags(NAMESPACE_TOMB, 'deleted-entry');
if (is_array($del_entries) && count($del_entries) && $pass != 2) {
foreach ($del_entries as $dentry) {
$deleted = false;
if (isset($dentry['attribs']['']['ref'])) {
$uri = $dentry['attribs']['']['ref'];
$deleted = true;
if (isset($dentry['attribs']['']['when'])) {
$when = $dentry['attribs']['']['when'];
$when = datetime_convert('UTC', 'UTC', $when, 'Y-m-d H:i:s');
} else {
$when = datetime_convert('UTC', 'UTC', 'now', 'Y-m-d H:i:s');
}
}
if ($deleted && is_array($contact)) {
$r = q("SELECT `item`.*, `contact`.`self` FROM `item` INNER JOIN `contact` on `item`.`contact-id` = `contact`.`id`\n\t\t\t\t\tWHERE `uri` = '%s' AND `item`.`uid` = %d AND `contact-id` = %d AND NOT `item`.`file` LIKE '%%[%%' LIMIT 1", dbesc($uri), intval($importer['uid']), intval($contact['id']));
if (count($r)) {
$item = $r[0];
if (!$item['deleted']) {
logger('consume_feed: deleting item ' . $item['id'] . ' uri=' . $item['uri'], LOGGER_DEBUG);
}
if ($item['object-type'] === ACTIVITY_OBJ_EVENT) {
logger("Deleting event " . $item['event-id'], LOGGER_DEBUG);
event_delete($item['event-id']);
}
if ($item['verb'] === ACTIVITY_TAG && $item['object-type'] === ACTIVITY_OBJ_TAGTERM) {
$xo = parse_xml_string($item['object'], false);
$xt = parse_xml_string($item['target'], false);
if ($xt->type === ACTIVITY_OBJ_NOTE) {
$i = q("select * from `item` where uri = '%s' and uid = %d limit 1", dbesc($xt->id), intval($importer['importer_uid']));
if (count($i)) {
// For tags, the owner cannot remove the tag on the author's copy of the post.
$owner_remove = $item['contact-id'] == $i[0]['contact-id'] ? true : false;
$author_remove = $item['origin'] && $item['self'] ? true : false;
$author_copy = $item['origin'] ? true : false;
if ($owner_remove && $author_copy) {
continue;
}
if ($author_remove || $owner_remove) {
$tags = explode(',', $i[0]['tag']);
$newtags = array();
if (count($tags)) {
foreach ($tags as $tag) {
if (trim($tag) !== trim($xo->body)) {
$newtags[] = trim($tag);
}
}
}
q("update item set tag = '%s' where id = %d", dbesc(implode(',', $newtags)), intval($i[0]['id']));
create_tags_from_item($i[0]['id']);
}
}
}
}
if ($item['uri'] == $item['parent-uri']) {
$r = q("UPDATE `item` SET `deleted` = 1, `edited` = '%s', `changed` = '%s',\n\t\t\t\t\t\t\t`body` = '', `title` = ''\n\t\t\t\t\t\t\tWHERE `parent-uri` = '%s' AND `uid` = %d", dbesc($when), dbesc(datetime_convert()), dbesc($item['uri']), intval($importer['uid']));
create_tags_from_itemuri($item['uri'], $importer['uid']);
create_files_from_itemuri($item['uri'], $importer['uid']);
update_thread_uri($item['uri'], $importer['uid']);
} else {
$r = q("UPDATE `item` SET `deleted` = 1, `edited` = '%s', `changed` = '%s',\n\t\t\t\t\t\t\t`body` = '', `title` = ''\n\t\t\t\t\t\t\tWHERE `uri` = '%s' AND `uid` = %d", dbesc($when), dbesc(datetime_convert()), dbesc($uri), intval($importer['uid']));
create_tags_from_itemuri($uri, $importer['uid']);
create_files_from_itemuri($uri, $importer['uid']);
if ($item['last-child']) {
// ensure that last-child is set in case the comment that had it just got wiped.
q("UPDATE `item` SET `last-child` = 0, `changed` = '%s' WHERE `parent-uri` = '%s' AND `uid` = %d ", dbesc(datetime_convert()), dbesc($item['parent-uri']), intval($item['uid']));
// who is the last child now?
$r = q("SELECT `id` FROM `item` WHERE `parent-uri` = '%s' AND `type` != 'activity' AND `deleted` = 0 AND `moderated` = 0 AND `uid` = %d\n\t\t\t\t\t\t\t\tORDER BY `created` DESC LIMIT 1", dbesc($item['parent-uri']), intval($importer['uid']));
if (count($r)) {
q("UPDATE `item` SET `last-child` = 1 WHERE `id` = %d", intval($r[0]['id']));
}
}
}
}
}
}
}
// Now process the feed
if ($feed->get_item_quantity()) {
logger('consume_feed: feed item count = ' . $feed->get_item_quantity());
// in inverse date order
if ($datedir) {
$items = array_reverse($feed->get_items());
} else {
$items = $feed->get_items();
}
foreach ($items as $item) {
$is_reply = false;
$item_id = $item->get_id();
$rawthread = $item->get_item_tags(NAMESPACE_THREAD, 'in-reply-to');
if (isset($rawthread[0]['attribs']['']['ref'])) {
$is_reply = true;
$parent_uri = $rawthread[0]['attribs']['']['ref'];
}
if ($is_reply && is_array($contact)) {
if ($pass == 1) {
continue;
}
// not allowed to post
if ($contact['rel'] == CONTACT_IS_FOLLOWER) {
continue;
}
// Have we seen it? If not, import it.
$item_id = $item->get_id();
$datarray = get_atom_elements($feed, $item, $contact);
if (!x($datarray, 'author-name') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-name'] = $contact['name'];
}
if (!x($datarray, 'author-link') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-link'] = $contact['url'];
}
if (!x($datarray, 'author-avatar') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-avatar'] = $contact['thumb'];
}
if (!x($datarray, 'author-name') || !x($datarray, 'author-link')) {
logger('consume_feed: no author information! ' . print_r($datarray, true));
continue;
}
$force_parent = false;
if ($contact['network'] === NETWORK_OSTATUS || stristr($contact['url'], 'twitter.com')) {
if ($contact['network'] === NETWORK_OSTATUS) {
$force_parent = true;
}
if (strlen($datarray['title'])) {
unset($datarray['title']);
}
$r = q("UPDATE `item` SET `last-child` = 0, `changed` = '%s' WHERE `parent-uri` = '%s' AND `uid` = %d", dbesc(datetime_convert()), dbesc($parent_uri), intval($importer['uid']));
$datarray['last-child'] = 1;
update_thread_uri($parent_uri, $importer['uid']);
}
$r = q("SELECT `uid`, `last-child`, `edited`, `body` FROM `item` WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", dbesc($item_id), intval($importer['uid']));
// Update content if 'updated' changes
if (count($r)) {
if (edited_timestamp_is_newer($r[0], $datarray)) {
// do not accept (ignore) an earlier edit than one we currently have.
if (datetime_convert('UTC', 'UTC', $datarray['edited']) < $r[0]['edited']) {
continue;
}
$r = q("UPDATE `item` SET `title` = '%s', `body` = '%s', `tag` = '%s', `edited` = '%s', `changed` = '%s' WHERE `uri` = '%s' AND `uid` = %d", dbesc($datarray['title']), dbesc($datarray['body']), dbesc($datarray['tag']), dbesc(datetime_convert('UTC', 'UTC', $datarray['edited'])), dbesc(datetime_convert()), dbesc($item_id), intval($importer['uid']));
create_tags_from_itemuri($item_id, $importer['uid']);
update_thread_uri($item_id, $importer['uid']);
}
// update last-child if it changes
$allow = $item->get_item_tags(NAMESPACE_DFRN, 'comment-allow');
if ($allow && $allow[0]['data'] != $r[0]['last-child']) {
$r = q("UPDATE `item` SET `last-child` = 0, `changed` = '%s' WHERE `parent-uri` = '%s' AND `uid` = %d", dbesc(datetime_convert()), dbesc($parent_uri), intval($importer['uid']));
$r = q("UPDATE `item` SET `last-child` = %d , `changed` = '%s' WHERE `uri` = '%s' AND `uid` = %d", intval($allow[0]['data']), dbesc(datetime_convert()), dbesc($item_id), intval($importer['uid']));
update_thread_uri($item_id, $importer['uid']);
}
continue;
}
if ($contact['network'] === NETWORK_FEED || !strlen($contact['notify'])) {
// one way feed - no remote comment ability
$datarray['last-child'] = 0;
}
$datarray['parent-uri'] = $parent_uri;
$datarray['uid'] = $importer['uid'];
$datarray['contact-id'] = $contact['id'];
if ($datarray['verb'] === ACTIVITY_LIKE || $datarray['verb'] === ACTIVITY_DISLIKE || $datarray['verb'] === ACTIVITY_ATTEND || $datarray['verb'] === ACTIVITY_ATTENDNO || $datarray['verb'] === ACTIVITY_ATTENDMAYBE) {
$datarray['type'] = 'activity';
$datarray['gravity'] = GRAVITY_LIKE;
// only one like or dislike per person
// splitted into two queries for performance issues
$r = q("select id from item where uid = %d and `contact-id` = %d and verb ='%s' and deleted = 0 and (`parent-uri` = '%s') limit 1", intval($datarray['uid']), intval($datarray['contact-id']), dbesc($datarray['verb']), dbesc($parent_uri));
if ($r && count($r)) {
continue;
}
$r = q("select id from item where uid = %d and `contact-id` = %d and verb ='%s' and deleted = 0 and (`thr-parent` = '%s') limit 1", intval($datarray['uid']), intval($datarray['contact-id']), dbesc($datarray['verb']), dbesc($parent_uri));
if ($r && count($r)) {
continue;
}
}
if ($datarray['verb'] === ACTIVITY_TAG && $datarray['object-type'] === ACTIVITY_OBJ_TAGTERM) {
$xo = parse_xml_string($datarray['object'], false);
$xt = parse_xml_string($datarray['target'], false);
if ($xt->type == ACTIVITY_OBJ_NOTE) {
$r = q("select * from item where `uri` = '%s' AND `uid` = %d limit 1", dbesc($xt->id), intval($importer['importer_uid']));
if (!count($r)) {
continue;
}
// extract tag, if not duplicate, add to parent item
if ($xo->id && $xo->content) {
$newtag = '#[url=' . $xo->id . ']' . $xo->content . '[/url]';
if (!stristr($r[0]['tag'], $newtag)) {
q("UPDATE item SET tag = '%s' WHERE id = %d", dbesc($r[0]['tag'] . (strlen($r[0]['tag']) ? ',' : '') . $newtag), intval($r[0]['id']));
create_tags_from_item($r[0]['id']);
}
}
}
}
$r = item_store($datarray, $force_parent);
continue;
} else {
// Head post of a conversation. Have we seen it? If not, import it.
$item_id = $item->get_id();
$datarray = get_atom_elements($feed, $item, $contact);
if (is_array($contact)) {
if (!x($datarray, 'author-name') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-name'] = $contact['name'];
}
if (!x($datarray, 'author-link') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-link'] = $contact['url'];
}
if (!x($datarray, 'author-avatar') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-avatar'] = $contact['thumb'];
}
}
if (!x($datarray, 'author-name') || !x($datarray, 'author-link')) {
logger('consume_feed: no author information! ' . print_r($datarray, true));
continue;
}
// special handling for events
if (x($datarray, 'object-type') && $datarray['object-type'] === ACTIVITY_OBJ_EVENT) {
$ev = bbtoevent($datarray['body']);
if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) {
$ev['uid'] = $importer['uid'];
$ev['uri'] = $item_id;
$ev['edited'] = $datarray['edited'];
$ev['private'] = $datarray['private'];
$ev['guid'] = $datarray['guid'];
if (is_array($contact)) {
$ev['cid'] = $contact['id'];
}
$r = q("SELECT * FROM `event` WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", dbesc($item_id), intval($importer['uid']));
if (count($r)) {
$ev['id'] = $r[0]['id'];
}
$xyz = event_store($ev);
continue;
}
}
if ($contact['network'] === NETWORK_OSTATUS || stristr($contact['url'], 'twitter.com')) {
if (strlen($datarray['title'])) {
unset($datarray['title']);
}
$datarray['last-child'] = 1;
}
$r = q("SELECT `uid`, `last-child`, `edited`, `body` FROM `item` WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", dbesc($item_id), intval($importer['uid']));
// Update content if 'updated' changes
if (count($r)) {
if (edited_timestamp_is_newer($r[0], $datarray)) {
// do not accept (ignore) an earlier edit than one we currently have.
if (datetime_convert('UTC', 'UTC', $datarray['edited']) < $r[0]['edited']) {
continue;
}
$r = q("UPDATE `item` SET `title` = '%s', `body` = '%s', `tag` = '%s', `edited` = '%s', `changed` = '%s' WHERE `uri` = '%s' AND `uid` = %d", dbesc($datarray['title']), dbesc($datarray['body']), dbesc($datarray['tag']), dbesc(datetime_convert('UTC', 'UTC', $datarray['edited'])), dbesc(datetime_convert()), dbesc($item_id), intval($importer['uid']));
create_tags_from_itemuri($item_id, $importer['uid']);
update_thread_uri($item_id, $importer['uid']);
}
// update last-child if it changes
$allow = $item->get_item_tags(NAMESPACE_DFRN, 'comment-allow');
if ($allow && $allow[0]['data'] != $r[0]['last-child']) {
$r = q("UPDATE `item` SET `last-child` = %d , `changed` = '%s' WHERE `uri` = '%s' AND `uid` = %d", intval($allow[0]['data']), dbesc(datetime_convert()), dbesc($item_id), intval($importer['uid']));
update_thread_uri($item_id, $importer['uid']);
}
continue;
}
if (activity_match($datarray['verb'], ACTIVITY_FOLLOW)) {
logger('consume-feed: New follower');
new_follower($importer, $contact, $datarray, $item);
return;
}
if (activity_match($datarray['verb'], ACTIVITY_UNFOLLOW)) {
lose_follower($importer, $contact, $datarray, $item);
return;
}
if (activity_match($datarray['verb'], ACTIVITY_REQ_FRIEND)) {
logger('consume-feed: New friend request');
new_follower($importer, $contact, $datarray, $item, true);
return;
}
if (activity_match($datarray['verb'], ACTIVITY_UNFRIEND)) {
lose_sharer($importer, $contact, $datarray, $item);
return;
}
if (!is_array($contact)) {
return;
}
if ($contact['network'] === NETWORK_FEED || !strlen($contact['notify'])) {
// one way feed - no remote comment ability
$datarray['last-child'] = 0;
}
if ($contact['network'] === NETWORK_FEED) {
$datarray['private'] = 2;
}
$datarray['parent-uri'] = $item_id;
$datarray['uid'] = $importer['uid'];
$datarray['contact-id'] = $contact['id'];
if (!link_compare($datarray['owner-link'], $contact['url'])) {
// The item owner info is not our contact. It's OK and is to be expected if this is a tgroup delivery,
// but otherwise there's a possible data mixup on the sender's system.
// the tgroup delivery code called from item_store will correct it if it's a forum,
// but we're going to unconditionally correct it here so that the post will always be owned by our contact.
logger('consume_feed: Correcting item owner.', LOGGER_DEBUG);
$datarray['owner-name'] = $contact['name'];
$datarray['owner-link'] = $contact['url'];
$datarray['owner-avatar'] = $contact['thumb'];
}
// We've allowed "followers" to reach this point so we can decide if they are
// posting an @-tag delivery, which followers are allowed to do for certain
// page types. Now that we've parsed the post, let's check if it is legit. Otherwise ignore it.
if ($contact['rel'] == CONTACT_IS_FOLLOWER && !tgroup_check($importer['uid'], $datarray)) {
continue;
}
// This is my contact on another system, but it's really me.
// Turn this into a wall post.
$notify = item_is_remote_self($contact, $datarray);
$r = item_store($datarray, false, $notify);
logger('Stored - Contact ' . $contact['url'] . ' Notify ' . $notify . ' return ' . $r . ' Item ' . print_r($datarray, true), LOGGER_DEBUG);
continue;
}
}
}
}
/**
*
* consume_feed - process atom feed and update anything/everything we might need to update
*
* $xml = the (atom) feed to consume - RSS isn't as fully supported but may work for simple feeds.
*
* $importer = the contact_record (joined to user_record) of the local user who owns this relationship.
* It is this person's stuff that is going to be updated.
* $contact = the person who is sending us stuff. If not set, we MAY be processing a "follow" activity
* from an external network and MAY create an appropriate contact record. Otherwise, we MUST
* have a contact record.
* $hub = should we find a hub declation in the feed, pass it back to our calling process, who might (or
* might not) try and subscribe to it.
* $datedir sorts in reverse order
* $pass - by default ($pass = 0) we cannot guarantee that a parent item has been
* imported prior to its children being seen in the stream unless we are certain
* of how the feed is arranged/ordered.
* With $pass = 1, we only pull parent items out of the stream.
* With $pass = 2, we only pull children (comments/likes).
*
* So running this twice, first with pass 1 and then with pass 2 will do the right
* thing regardless of feed ordering. This won't be adequate in a fully-threaded
* model where comments can have sub-threads. That would require some massive sorting
* to get all the feed items into a mostly linear ordering, and might still require
* recursion.
*/
function consume_feed($xml, $importer, &$contact, &$hub, $datedir = 0, $pass = 0)
{
require_once 'library/simplepie/simplepie.inc';
if (!strlen($xml)) {
logger('consume_feed: empty input');
return;
}
$feed = new SimplePie();
$feed->set_raw_data($xml);
if ($datedir) {
$feed->enable_order_by_date(true);
} else {
$feed->enable_order_by_date(false);
}
$feed->init();
if ($feed->error()) {
logger('consume_feed: Error parsing XML: ' . $feed->error());
}
$permalink = $feed->get_permalink();
// Check at the feed level for updated contact name and/or photo
$name_updated = '';
$new_name = '';
$photo_timestamp = '';
$photo_url = '';
$birthday = '';
$hubs = $feed->get_links('hub');
if (count($hubs)) {
$hub = implode(',', $hubs);
}
$rawtags = $feed->get_feed_tags(NAMESPACE_DFRN, 'owner');
if (!$rawtags) {
$rawtags = $feed->get_feed_tags(SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
}
if ($rawtags) {
$elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10];
if ($elems['name'][0]['attribs'][NAMESPACE_DFRN]['updated']) {
$name_updated = $elems['name'][0]['attribs'][NAMESPACE_DFRN]['updated'];
$new_name = $elems['name'][0]['data'];
}
if (x($elems, 'link') && $elems['link'][0]['attribs']['']['rel'] === 'photo' && $elems['link'][0]['attribs'][NAMESPACE_DFRN]['updated']) {
$photo_timestamp = datetime_convert('UTC', 'UTC', $elems['link'][0]['attribs'][NAMESPACE_DFRN]['updated']);
$photo_url = $elems['link'][0]['attribs']['']['href'];
}
if (x($rawtags[0]['child'], NAMESPACE_DFRN) && x($rawtags[0]['child'][NAMESPACE_DFRN], 'birthday')) {
$birthday = datetime_convert('UTC', 'UTC', $rawtags[0]['child'][NAMESPACE_DFRN]['birthday'][0]['data']);
}
}
if (is_array($contact) && $photo_timestamp && strlen($photo_url) && $photo_timestamp > $contact['avatar-date']) {
logger('consume_feed: Updating photo for ' . $contact['name']);
require_once "Photo.php";
$photo_failure = false;
$have_photo = false;
$r = q("SELECT `resource-id` FROM `photo` WHERE `contact-id` = %d AND `uid` = %d LIMIT 1", intval($contact['id']), intval($contact['uid']));
if (count($r)) {
$resource_id = $r[0]['resource-id'];
$have_photo = true;
} else {
$resource_id = photo_new_resource();
}
$img_str = fetch_url($photo_url, true);
$img = new Photo($img_str);
if ($img->is_valid()) {
if ($have_photo) {
q("DELETE FROM `photo` WHERE `resource-id` = '%s' AND `contact-id` = %d AND `uid` = %d", dbesc($resource_id), intval($contact['id']), intval($contact['uid']));
}
$img->scaleImageSquare(175);
$hash = $resource_id;
$r = $img->store($contact['uid'], $contact['id'], $hash, basename($photo_url), 'Contact Photos', 4);
$img->scaleImage(80);
$r = $img->store($contact['uid'], $contact['id'], $hash, basename($photo_url), 'Contact Photos', 5);
$img->scaleImage(48);
$r = $img->store($contact['uid'], $contact['id'], $hash, basename($photo_url), 'Contact Photos', 6);
$a = get_app();
q("UPDATE `contact` SET `avatar-date` = '%s', `photo` = '%s', `thumb` = '%s', `micro` = '%s' \n\t\t\t\tWHERE `uid` = %d AND `id` = %d LIMIT 1", dbesc(datetime_convert()), dbesc($a->get_baseurl() . '/photo/' . $hash . '-4.jpg'), dbesc($a->get_baseurl() . '/photo/' . $hash . '-5.jpg'), dbesc($a->get_baseurl() . '/photo/' . $hash . '-6.jpg'), intval($contact['uid']), intval($contact['id']));
}
}
if (is_array($contact) && $name_updated && strlen($new_name) && $name_updated > $contact['name-date']) {
$r = q("select * from contact where uid = %d and id = %d limit 1", intval($contact['uid']), intval($contact['id']));
$x = q("UPDATE `contact` SET `name` = '%s', `name-date` = '%s' WHERE `uid` = %d AND `id` = %d LIMIT 1", dbesc(notags(trim($new_name))), dbesc(datetime_convert()), intval($contact['uid']), intval($contact['id']));
// do our best to update the name on content items
if (count($r)) {
q("update item set `author-name` = '%s' where `author-name` = '%s' and `author-link` = '%s' and uid = %d", dbesc(notags(trim($new_name))), dbesc($r[0]['name']), dbesc($r[0]['url']), intval($contact['uid']));
}
}
if (strlen($birthday)) {
if (substr($birthday, 0, 4) != $contact['bdyear']) {
logger('consume_feed: updating birthday: ' . $birthday);
/**
*
* Add new birthday event for this person
*
* $bdtext is just a readable placeholder in case the event is shared
* with others. We will replace it during presentation to our $importer
* to contain a sparkle link and perhaps a photo.
*
*/
$bdtext = t('Birthday:') . ' [url=' . $contact['url'] . ']' . $contact['name'] . '[/url]';
$r = q("INSERT INTO `event` (`uid`,`cid`,`created`,`edited`,`start`,`finish`,`desc`,`type`)\n\t\t\t\tVALUES ( %d, %d, '%s', '%s', '%s', '%s', '%s', '%s' ) ", intval($contact['uid']), intval($contact['id']), dbesc(datetime_convert()), dbesc(datetime_convert()), dbesc(datetime_convert('UTC', 'UTC', $birthday)), dbesc(datetime_convert('UTC', 'UTC', $birthday . ' + 1 day ')), dbesc($bdtext), dbesc('birthday'));
// update bdyear
q("UPDATE `contact` SET `bdyear` = '%s' WHERE `uid` = %d AND `id` = %d LIMIT 1", dbesc(substr($birthday, 0, 4)), intval($contact['uid']), intval($contact['id']));
// This function is called twice without reloading the contact
// Make sure we only create one event. This is why &$contact
// is a reference var in this function
$contact['bdyear'] = substr($birthday, 0, 4);
}
}
$community_page = 0;
$rawtags = $feed->get_feed_tags(NAMESPACE_DFRN, 'community');
if ($rawtags) {
$community_page = intval($rawtags[0]['data']);
}
if (is_array($contact) && intval($contact['forum']) != $community_page) {
q("update contact set forum = %d where id = %d limit 1", intval($community_page), intval($contact['id']));
$contact['forum'] = (string) $community_page;
}
// process any deleted entries
$del_entries = $feed->get_feed_tags(NAMESPACE_TOMB, 'deleted-entry');
if (is_array($del_entries) && count($del_entries) && $pass != 2) {
foreach ($del_entries as $dentry) {
$deleted = false;
if (isset($dentry['attribs']['']['ref'])) {
$uri = $dentry['attribs']['']['ref'];
$deleted = true;
if (isset($dentry['attribs']['']['when'])) {
$when = $dentry['attribs']['']['when'];
$when = datetime_convert('UTC', 'UTC', $when, 'Y-m-d H:i:s');
} else {
$when = datetime_convert('UTC', 'UTC', 'now', 'Y-m-d H:i:s');
}
}
if ($deleted && is_array($contact)) {
$r = q("SELECT `item`.*, `contact`.`self` FROM `item` left join `contact` on `item`.`contact-id` = `contact`.`id` \n\t\t\t\t\tWHERE `uri` = '%s' AND `item`.`uid` = %d AND `contact-id` = %d AND NOT `item`.`file` LIKE '%%[%%' LIMIT 1", dbesc($uri), intval($importer['uid']), intval($contact['id']));
if (count($r)) {
$item = $r[0];
if (!$item['deleted']) {
logger('consume_feed: deleting item ' . $item['id'] . ' uri=' . $item['uri'], LOGGER_DEBUG);
}
if ($item['verb'] === ACTIVITY_TAG && $item['object-type'] === ACTIVITY_OBJ_TAGTERM) {
$xo = parse_xml_string($item['object'], false);
$xt = parse_xml_string($item['target'], false);
if ($xt->type === ACTIVITY_OBJ_NOTE) {
$i = q("select * from `item` where uri = '%s' and uid = %d limit 1", dbesc($xt->id), intval($importer['importer_uid']));
if (count($i)) {
// For tags, the owner cannot remove the tag on the author's copy of the post.
$owner_remove = $item['contact-id'] == $i[0]['contact-id'] ? true : false;
$author_remove = $item['origin'] && $item['self'] ? true : false;
$author_copy = $item['origin'] ? true : false;
if ($owner_remove && $author_copy) {
continue;
}
if ($author_remove || $owner_remove) {
$tags = explode(',', $i[0]['tag']);
$newtags = array();
if (count($tags)) {
foreach ($tags as $tag) {
if (trim($tag) !== trim($xo->body)) {
$newtags[] = trim($tag);
}
}
}
q("update item set tag = '%s' where id = %d limit 1", dbesc(implode(',', $newtags)), intval($i[0]['id']));
}
}
}
}
if ($item['uri'] == $item['parent-uri']) {
$r = q("UPDATE `item` SET `deleted` = 1, `edited` = '%s', `changed` = '%s',\n\t\t\t\t\t\t\t`body` = '', `title` = ''\n\t\t\t\t\t\t\tWHERE `parent-uri` = '%s' AND `uid` = %d", dbesc($when), dbesc(datetime_convert()), dbesc($item['uri']), intval($importer['uid']));
} else {
$r = q("UPDATE `item` SET `deleted` = 1, `edited` = '%s', `changed` = '%s',\n\t\t\t\t\t\t\t`body` = '', `title` = '' \n\t\t\t\t\t\t\tWHERE `uri` = '%s' AND `uid` = %d LIMIT 1", dbesc($when), dbesc(datetime_convert()), dbesc($uri), intval($importer['uid']));
if ($item['last-child']) {
// ensure that last-child is set in case the comment that had it just got wiped.
q("UPDATE `item` SET `last-child` = 0, `changed` = '%s' WHERE `parent-uri` = '%s' AND `uid` = %d ", dbesc(datetime_convert()), dbesc($item['parent-uri']), intval($item['uid']));
// who is the last child now?
$r = q("SELECT `id` FROM `item` WHERE `parent-uri` = '%s' AND `type` != 'activity' AND `deleted` = 0 AND `moderated` = 0 AND `uid` = %d \n\t\t\t\t\t\t\t\tORDER BY `created` DESC LIMIT 1", dbesc($item['parent-uri']), intval($importer['uid']));
if (count($r)) {
q("UPDATE `item` SET `last-child` = 1 WHERE `id` = %d LIMIT 1", intval($r[0]['id']));
}
}
}
}
}
}
}
// Now process the feed
if ($feed->get_item_quantity()) {
logger('consume_feed: feed item count = ' . $feed->get_item_quantity());
// in inverse date order
if ($datedir) {
$items = array_reverse($feed->get_items());
} else {
$items = $feed->get_items();
}
foreach ($items as $item) {
$is_reply = false;
$item_id = $item->get_id();
$rawthread = $item->get_item_tags(NAMESPACE_THREAD, 'in-reply-to');
if (isset($rawthread[0]['attribs']['']['ref'])) {
$is_reply = true;
$parent_uri = $rawthread[0]['attribs']['']['ref'];
}
if ($is_reply && is_array($contact)) {
if ($pass == 1) {
continue;
}
// Have we seen it? If not, import it.
$item_id = $item->get_id();
$datarray = get_atom_elements($feed, $item);
if (!x($datarray, 'author-name') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-name'] = $contact['name'];
}
if (!x($datarray, 'author-link') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-link'] = $contact['url'];
}
if (!x($datarray, 'author-avatar') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-avatar'] = $contact['thumb'];
}
if (!x($datarray, 'author-name') || !x($datarray, 'author-link')) {
logger('consume_feed: no author information! ' . print_r($datarray, true));
continue;
}
$r = q("SELECT `uid`, `last-child`, `edited`, `body` FROM `item` WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", dbesc($item_id), intval($importer['uid']));
// Update content if 'updated' changes
if (count($r)) {
if (x($datarray, 'edited') !== false && datetime_convert('UTC', 'UTC', $datarray['edited']) !== $r[0]['edited']) {
$r = q("UPDATE `item` SET `title` = '%s', `body` = '%s', `tag` = '%s', `edited` = '%s' WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", dbesc($datarray['title']), dbesc($datarray['body']), dbesc($datarray['tag']), dbesc(datetime_convert('UTC', 'UTC', $datarray['edited'])), dbesc($item_id), intval($importer['uid']));
}
// update last-child if it changes
$allow = $item->get_item_tags(NAMESPACE_DFRN, 'comment-allow');
if ($allow && $allow[0]['data'] != $r[0]['last-child']) {
$r = q("UPDATE `item` SET `last-child` = 0, `changed` = '%s' WHERE `parent-uri` = '%s' AND `uid` = %d", dbesc(datetime_convert()), dbesc($parent_uri), intval($importer['uid']));
$r = q("UPDATE `item` SET `last-child` = %d , `changed` = '%s' WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", intval($allow[0]['data']), dbesc(datetime_convert()), dbesc($item_id), intval($importer['uid']));
}
continue;
}
$force_parent = false;
if ($contact['network'] === NETWORK_OSTATUS || stristr($contact['url'], 'twitter.com')) {
if ($contact['network'] === NETWORK_OSTATUS) {
$force_parent = true;
}
if (strlen($datarray['title'])) {
unset($datarray['title']);
}
$r = q("UPDATE `item` SET `last-child` = 0, `changed` = '%s' WHERE `parent-uri` = '%s' AND `uid` = %d", dbesc(datetime_convert()), dbesc($parent_uri), intval($importer['uid']));
$datarray['last-child'] = 1;
}
if ($contact['network'] === NETWORK_FEED || !strlen($contact['notify'])) {
// one way feed - no remote comment ability
$datarray['last-child'] = 0;
}
$datarray['parent-uri'] = $parent_uri;
$datarray['uid'] = $importer['uid'];
$datarray['contact-id'] = $contact['id'];
if (activity_match($datarray['verb'], ACTIVITY_LIKE) || activity_match($datarray['verb'], ACTIVITY_DISLIKE)) {
$datarray['type'] = 'activity';
$datarray['gravity'] = GRAVITY_LIKE;
// only one like or dislike per person
$r = q("select id from item where uid = %d and `contact-id` = %d and verb ='%s' and deleted = 0 limit 1", intval($datarray['uid']), intval($datarray['contact-id']), dbesc($datarray['verb']));
if ($r && count($r)) {
continue;
}
}
if ($datarray['verb'] === ACTIVITY_TAG && $datarray['object-type'] === ACTIVITY_OBJ_TAGTERM) {
$xo = parse_xml_string($datarray['object'], false);
$xt = parse_xml_string($datarray['target'], false);
if ($xt->type == ACTIVITY_OBJ_NOTE) {
$r = q("select * from item where `uri` = '%s' AND `uid` = %d limit 1", dbesc($xt->id), intval($importer['importer_uid']));
if (!count($r)) {
continue;
}
// extract tag, if not duplicate, add to parent item
if ($xo->id && $xo->content) {
$newtag = '#[url=' . $xo->id . ']' . $xo->content . '[/url]';
if (!stristr($r[0]['tag'], $newtag)) {
q("UPDATE item SET tag = '%s' WHERE id = %d LIMIT 1", dbesc($r[0]['tag'] . (strlen($r[0]['tag']) ? ',' : '') . $newtag), intval($r[0]['id']));
}
}
}
}
$r = item_store($datarray, $force_parent);
continue;
} else {
// Head post of a conversation. Have we seen it? If not, import it.
$item_id = $item->get_id();
$datarray = get_atom_elements($feed, $item);
if (is_array($contact)) {
if (!x($datarray, 'author-name') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-name'] = $contact['name'];
}
if (!x($datarray, 'author-link') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-link'] = $contact['url'];
}
if (!x($datarray, 'author-avatar') && $contact['network'] != NETWORK_DFRN) {
$datarray['author-avatar'] = $contact['thumb'];
}
}
if (!x($datarray, 'author-name') || !x($datarray, 'author-link')) {
logger('consume_feed: no author information! ' . print_r($datarray, true));
continue;
}
// special handling for events
if (x($datarray, 'object-type') && $datarray['object-type'] === ACTIVITY_OBJ_EVENT) {
$ev = bbtoevent($datarray['body']);
if (x($ev, 'desc') && x($ev, 'start')) {
$ev['uid'] = $importer['uid'];
$ev['uri'] = $item_id;
$ev['edited'] = $datarray['edited'];
$ev['private'] = $datarray['private'];
if (is_array($contact)) {
$ev['cid'] = $contact['id'];
}
$r = q("SELECT * FROM `event` WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", dbesc($item_id), intval($importer['uid']));
if (count($r)) {
$ev['id'] = $r[0]['id'];
}
$xyz = event_store($ev);
continue;
}
}
$r = q("SELECT `uid`, `last-child`, `edited`, `body` FROM `item` WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", dbesc($item_id), intval($importer['uid']));
// Update content if 'updated' changes
if (count($r)) {
if (x($datarray, 'edited') !== false && datetime_convert('UTC', 'UTC', $datarray['edited']) !== $r[0]['edited']) {
$r = q("UPDATE `item` SET `title` = '%s', `body` = '%s', `tag` = '%s', `edited` = '%s' WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", dbesc($datarray['title']), dbesc($datarray['body']), dbesc($datarray['tag']), dbesc(datetime_convert('UTC', 'UTC', $datarray['edited'])), dbesc($item_id), intval($importer['uid']));
}
// update last-child if it changes
$allow = $item->get_item_tags(NAMESPACE_DFRN, 'comment-allow');
if ($allow && $allow[0]['data'] != $r[0]['last-child']) {
$r = q("UPDATE `item` SET `last-child` = %d , `changed` = '%s' WHERE `uri` = '%s' AND `uid` = %d LIMIT 1", intval($allow[0]['data']), dbesc(datetime_convert()), dbesc($item_id), intval($importer['uid']));
}
continue;
}
if (activity_match($datarray['verb'], ACTIVITY_FOLLOW)) {
logger('consume-feed: New follower');
new_follower($importer, $contact, $datarray, $item);
return;
}
if (activity_match($datarray['verb'], ACTIVITY_UNFOLLOW)) {
lose_follower($importer, $contact, $datarray, $item);
return;
}
if (activity_match($datarray['verb'], ACTIVITY_REQ_FRIEND)) {
logger('consume-feed: New friend request');
new_follower($importer, $contact, $datarray, $item, true);
return;
}
if (activity_match($datarray['verb'], ACTIVITY_UNFRIEND)) {
lose_sharer($importer, $contact, $datarray, $item);
return;
}
if (!is_array($contact)) {
return;
}
if ($contact['network'] === NETWORK_OSTATUS || stristr($contact['url'], 'twitter.com')) {
if (strlen($datarray['title'])) {
unset($datarray['title']);
}
$datarray['last-child'] = 1;
}
if ($contact['network'] === NETWORK_FEED || !strlen($contact['notify'])) {
// one way feed - no remote comment ability
$datarray['last-child'] = 0;
}
// This is my contact on another system, but it's really me.
// Turn this into a wall post.
if ($contact['remote_self']) {
$datarray['wall'] = 1;
}
$datarray['parent-uri'] = $item_id;
$datarray['uid'] = $importer['uid'];
$datarray['contact-id'] = $contact['id'];
if (!link_compare($datarray['owner-link'], $contact['url'])) {
// The item owner info is not our contact. It's OK and is to be expected if this is a tgroup delivery,
// but otherwise there's a possible data mixup on the sender's system.
// the tgroup delivery code called from item_store will correct it if it's a forum,
// but we're going to unconditionally correct it here so that the post will always be owned by our contact.
logger('consume_feed: Correcting item owner.', LOGGER_DEBUG);
$datarray['owner-name'] = $contact['name'];
$datarray['owner-link'] = $contact['url'];
$datarray['owner-avatar'] = $contact['thumb'];
}
$r = item_store($datarray);
continue;
}
}
}
}
function feed_meta($xml)
{
require_once 'library/simplepie/simplepie.inc';
$ret = array();
if (!strlen($xml)) {
logger('empty input');
return $ret;
}
$feed = new SimplePie();
$feed->set_raw_data($xml);
$feed->init();
if ($feed->error()) {
logger('Error parsing XML: ' . $feed->error());
return $ret;
}
$ret['hubs'] = $feed->get_links('hub');
// logger('consume_feed: hubs: ' . print_r($hubs,true), LOGGER_DATA);
$author = array();
$found_author = $feed->get_author();
if ($found_author) {
$author['author_name'] = unxmlify($found_author->get_name());
$author['author_link'] = unxmlify($found_author->get_link());
$rawauthor = $feed->get_feed_tags(SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
logger('rawauthor: ' . print_r($rawauthor, true));
if ($rawauthor) {
if ($rawauthor[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10]['link']) {
$base = $rawauthor[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10]['link'];
foreach ($base as $link) {
if (!x($author, 'author_photo') || !$author['author_photo']) {
if ($link['attribs']['']['rel'] === 'photo' || $link['attribs']['']['rel'] === 'avatar') {
$author['author_photo'] = unxmlify($link['attribs']['']['href']);
break;
}
}
}
}
if ($rawauthor[0]['child'][NAMESPACE_POCO]['displayName'][0]['data']) {
$author['full_name'] = unxmlify($rawauthor[0]['child'][NAMESPACE_POCO]['displayName'][0]['data']);
}
if ($rawauthor[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10]['uri'][0]['data']) {
$author['author_uri'] = unxmlify($rawauthor[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10]['uri'][0]['data']);
}
}
}
if (substr($author['author_link'], -1, 1) == '/') {
$author['author_link'] = substr($author['author_link'], 0, -1);
}
$ret['author'] = $author;
return $ret;
}
function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false, $override_url = false)
{
require_once "lib/simplepie/simplepie.inc";
require_once "lib/magpierss/rss_fetch.inc";
require_once 'lib/magpierss/rss_utils.inc';
$debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
if (!$_REQUEST["daemon"] && !$ignore_daemon) {
return false;
}
if ($debug_enabled) {
_debug("update_rss_feed: start");
}
if (!$ignore_daemon) {
if (DB_TYPE == "pgsql") {
$updstart_thresh_qpart = "(ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '120 seconds')";
} else {
$updstart_thresh_qpart = "(ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 120 SECOND))";
}
$result = db_query($link, "SELECT id,update_interval,auth_login,\n\t\t\t\tauth_pass,cache_images,update_method,last_updated\n\t\t\t\tFROM ttrss_feeds WHERE id = '{$feed}' AND {$updstart_thresh_qpart}");
} else {
$result = db_query($link, "SELECT id,update_interval,auth_login,\n\t\t\t\tfeed_url,auth_pass,cache_images,update_method,last_updated,\n\t\t\t\tmark_unread_on_update, owner_uid, update_on_checksum_change,\n\t\t\t\tpubsub_state\n\t\t\t\tFROM ttrss_feeds WHERE id = '{$feed}'");
}
if (db_num_rows($result) == 0) {
if ($debug_enabled) {
_debug("update_rss_feed: feed {$feed} NOT FOUND/SKIPPED");
}
return false;
}
$update_method = db_fetch_result($result, 0, "update_method");
$last_updated = db_fetch_result($result, 0, "last_updated");
$owner_uid = db_fetch_result($result, 0, "owner_uid");
$mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, 0, "mark_unread_on_update"));
$update_on_checksum_change = sql_bool_to_bool(db_fetch_result($result, 0, "update_on_checksum_change"));
$pubsub_state = db_fetch_result($result, 0, "pubsub_state");
db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW()\n\t\t\tWHERE id = '{$feed}'");
$auth_login = db_fetch_result($result, 0, "auth_login");
$auth_pass = db_fetch_result($result, 0, "auth_pass");
if ($update_method == 0) {
$update_method = DEFAULT_UPDATE_METHOD + 1;
}
// 1 - Magpie
// 2 - SimplePie
// 3 - Twitter OAuth
if ($update_method == 2) {
$use_simplepie = true;
} else {
$use_simplepie = false;
}
if ($debug_enabled) {
_debug("update method: {$update_method} (feed setting: {$update_method}) (use simplepie: {$use_simplepie})\n");
}
if ($update_method == 1) {
$auth_login = urlencode($auth_login);
$auth_pass = urlencode($auth_pass);
}
$cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
$fetch_url = db_fetch_result($result, 0, "feed_url");
$feed = db_escape_string($feed);
if ($auth_login && $auth_pass) {
$url_parts = array();
preg_match("/(^[^:]*):\\/\\/(.*)/", $fetch_url, $url_parts);
if ($url_parts[1] && $url_parts[2]) {
$fetch_url = $url_parts[1] . "://{$auth_login}:{$auth_pass}@" . $url_parts[2];
}
}
if ($override_url) {
$fetch_url = $override_url;
}
if ($debug_enabled) {
_debug("update_rss_feed: fetching [{$fetch_url}]...");
}
// Ignore cache if new feed or manual update.
$cache_age = is_null($last_updated) || $last_updated == '1970-01-01 00:00:00' ? -1 : get_feed_update_interval($link, $feed) * 60;
if ($update_method == 3) {
$rss = fetch_twitter_rss($link, $fetch_url, $owner_uid);
} else {
if ($update_method == 1) {
define('MAGPIE_CACHE_AGE', $cache_age);
define('MAGPIE_CACHE_ON', !$no_cache);
define('MAGPIE_FETCH_TIME_OUT', 60);
define('MAGPIE_CACHE_DIR', CACHE_DIR . "/magpie");
$rss = @fetch_rss($fetch_url);
} else {
$simplepie_cache_dir = CACHE_DIR . "/simplepie";
if (!is_dir($simplepie_cache_dir)) {
mkdir($simplepie_cache_dir);
}
$rss = new SimplePie();
$rss->set_useragent(SELF_USER_AGENT);
# $rss->set_timeout(10);
$rss->set_feed_url($fetch_url);
$rss->set_output_encoding('UTF-8');
//$rss->force_feed(true);
if ($debug_enabled) {
_debug("feed update interval (sec): " . get_feed_update_interval($link, $feed) * 60);
}
$rss->enable_cache(!$no_cache);
if (!$no_cache) {
$rss->set_cache_location($simplepie_cache_dir);
$rss->set_cache_duration($cache_age);
}
$rss->init();
}
}
// print_r($rss);
if ($debug_enabled) {
_debug("update_rss_feed: fetch done, parsing...");
}
$feed = db_escape_string($feed);
if ($update_method == 2) {
$fetch_ok = !$rss->error();
} else {
$fetch_ok = !!$rss;
}
if ($fetch_ok) {
if ($debug_enabled) {
_debug("update_rss_feed: processing feed data...");
}
// db_query($link, "BEGIN");
if (DB_TYPE == "pgsql") {
$favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
} else {
$favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
}
$result = db_query($link, "SELECT title,icon_url,site_url,owner_uid,\n\t\t\t\t(favicon_last_checked IS NULL OR {$favicon_interval_qpart}) AS\n\t\t\t\t\t\tfavicon_needs_check\n\t\t\t\tFROM ttrss_feeds WHERE id = '{$feed}'");
$registered_title = db_fetch_result($result, 0, "title");
$orig_icon_url = db_fetch_result($result, 0, "icon_url");
$orig_site_url = db_fetch_result($result, 0, "site_url");
$favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, "favicon_needs_check"));
$owner_uid = db_fetch_result($result, 0, "owner_uid");
if ($use_simplepie) {
$site_url = db_escape_string(trim($rss->get_link()));
} else {
$site_url = db_escape_string(trim($rss->channel["link"]));
}
// weird, weird Magpie
if (!$use_simplepie) {
if (!$site_url) {
$site_url = db_escape_string($rss->channel["link_"]);
}
}
$site_url = rewrite_relative_url($fetch_url, $site_url);
$site_url = substr($site_url, 0, 250);
if ($debug_enabled) {
_debug("update_rss_feed: checking favicon...");
}
if ($favicon_needs_check) {
check_feed_favicon($site_url, $feed, $link);
db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW()\n\t\t\t\t\tWHERE id = '{$feed}'");
}
if (!$registered_title || $registered_title == "[Unknown]") {
if ($use_simplepie) {
$feed_title = db_escape_string($rss->get_title());
} else {
$feed_title = db_escape_string($rss->channel["title"]);
}
if ($debug_enabled) {
_debug("update_rss_feed: registering title: {$feed_title}");
}
db_query($link, "UPDATE ttrss_feeds SET\n\t\t\t\t\ttitle = '{$feed_title}' WHERE id = '{$feed}'");
}
if ($site_url && $orig_site_url != $site_url) {
db_query($link, "UPDATE ttrss_feeds SET\n\t\t\t\t\tsite_url = '{$site_url}' WHERE id = '{$feed}'");
}
// print "I: " . $rss->channel["image"]["url"];
if (!$use_simplepie) {
$icon_url = db_escape_string(trim($rss->image["url"]));
} else {
$icon_url = db_escape_string(trim($rss->get_image_url()));
}
$icon_url = rewrite_relative_url($fetch_url, $icon_url);
$icon_url = substr($icon_url, 0, 250);
if ($icon_url && $orig_icon_url != $icon_url) {
db_query($link, "UPDATE ttrss_feeds SET icon_url = '{$icon_url}' WHERE id = '{$feed}'");
}
if ($debug_enabled) {
_debug("update_rss_feed: loading filters...");
}
$filters = load_filters($link, $feed, $owner_uid);
// if ($debug_enabled) {
// print_r($filters);
// }
if ($use_simplepie) {
$iterator = $rss->get_items();
} else {
$iterator = $rss->items;
if (!$iterator || !is_array($iterator)) {
$iterator = $rss->entries;
}
if (!$iterator || !is_array($iterator)) {
$iterator = $rss;
}
}
if (!is_array($iterator)) {
/* db_query($link, "UPDATE ttrss_feeds
SET last_error = 'Parse error: can\'t find any articles.'
WHERE id = '$feed'"); */
// clear any errors and mark feed as updated if fetched okay
// even if it's blank
if ($debug_enabled) {
_debug("update_rss_feed: entry iterator is not an array, no articles?");
}
db_query($link, "UPDATE ttrss_feeds\n\t\t\t\t\tSET last_updated = NOW(), last_error = '' WHERE id = '{$feed}'");
return;
// no articles
}
if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) {
if ($debug_enabled) {
_debug("update_rss_feed: checking for PUSH hub...");
}
$feed_hub_url = false;
if ($use_simplepie) {
$links = $rss->get_links('hub');
if ($links && is_array($links)) {
foreach ($links as $l) {
$feed_hub_url = $l;
break;
}
}
} else {
$atom = $rss->channel['atom'];
if ($atom) {
if ($atom['link@rel'] == 'hub') {
$feed_hub_url = $atom['link@href'];
}
if (!$feed_hub_url && $atom['link#'] > 1) {
for ($i = 2; $i <= $atom['link#']; $i++) {
if ($atom["link#{$i}@rel"] == 'hub') {
$feed_hub_url = $atom["link#{$i}@href"];
break;
}
}
}
} else {
$feed_hub_url = $rss->channel['link_hub'];
}
}
if ($debug_enabled) {
_debug("update_rss_feed: feed hub url: {$feed_hub_url}");
}
if ($feed_hub_url && function_exists('curl_init') && !ini_get("open_basedir")) {
require_once 'lib/pubsubhubbub/subscriber.php';
$callback_url = get_self_url_prefix() . "/public.php?op=pubsub&id={$feed}";
$s = new Subscriber($feed_hub_url, $callback_url);
$rc = $s->subscribe($fetch_url);
if ($debug_enabled) {
_debug("update_rss_feed: feed hub url found, subscribe request sent.");
}
db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1\n\t\t\t\t\t\tWHERE id = '{$feed}'");
}
}
if ($debug_enabled) {
_debug("update_rss_feed: processing articles...");
}
foreach ($iterator as $item) {
if ($_REQUEST['xdebug'] == 2) {
print_r($item);
}
if ($use_simplepie) {
$entry_guid = $item->get_id();
if (!$entry_guid) {
$entry_guid = $item->get_link();
}
if (!$entry_guid) {
$entry_guid = make_guid_from_title($item->get_title());
}
} else {
$entry_guid = $item["id"];
if (!$entry_guid) {
$entry_guid = $item["guid"];
}
if (!$entry_guid) {
$entry_guid = $item["about"];
}
if (!$entry_guid) {
$entry_guid = $item["link"];
}
if (!$entry_guid) {
$entry_guid = make_guid_from_title($item["title"]);
}
}
if ($debug_enabled) {
_debug("update_rss_feed: guid {$entry_guid}");
}
if (!$entry_guid) {
continue;
}
$entry_timestamp = "";
if ($use_simplepie) {
$entry_timestamp = strtotime($item->get_date());
} else {
$rss_2_date = $item['pubdate'];
$rss_1_date = $item['dc']['date'];
$atom_date = $item['issued'];
if (!$atom_date) {
$atom_date = $item['updated'];
}
if ($atom_date != "") {
$entry_timestamp = parse_w3cdtf($atom_date);
}
if ($rss_1_date != "") {
$entry_timestamp = parse_w3cdtf($rss_1_date);
}
if ($rss_2_date != "") {
$entry_timestamp = strtotime($rss_2_date);
}
}
if ($entry_timestamp == "" || $entry_timestamp == -1 || !$entry_timestamp) {
$entry_timestamp = time();
$no_orig_date = 'true';
} else {
$no_orig_date = 'false';
}
$entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
if ($debug_enabled) {
_debug("update_rss_feed: date {$entry_timestamp} [{$entry_timestamp_fmt}]");
}
if ($use_simplepie) {
$entry_title = $item->get_title();
} else {
$entry_title = trim(strip_tags($item["title"]));
}
if ($use_simplepie) {
$entry_link = $item->get_link();
} else {
// strange Magpie workaround
$entry_link = $item["link_"];
if (!$entry_link) {
$entry_link = $item["link"];
}
}
$entry_link = rewrite_relative_url($site_url, $entry_link);
if ($debug_enabled) {
_debug("update_rss_feed: title {$entry_title}");
_debug("update_rss_feed: link {$entry_link}");
}
if (!$entry_title) {
$entry_title = date("Y-m-d H:i:s", $entry_timestamp);
}
$entry_link = strip_tags($entry_link);
if ($use_simplepie) {
$entry_content = $item->get_content();
if (!$entry_content) {
$entry_content = $item->get_description();
}
} else {
$entry_content = $item["content:escaped"];
if (!$entry_content) {
$entry_content = $item["content:encoded"];
}
if (!$entry_content && is_array($entry_content)) {
$entry_content = $item["content"]["encoded"];
}
if (!$entry_content) {
$entry_content = $item["content"];
}
if (is_array($entry_content)) {
$entry_content = $entry_content[0];
}
// Magpie bugs are getting ridiculous
if (trim($entry_content) == "Array") {
$entry_content = false;
}
if (!$entry_content) {
$entry_content = $item["atom_content"];
}
if (!$entry_content) {
$entry_content = $item["summary"];
}
if (!$entry_content || strlen($entry_content) < strlen($item["description"])) {
$entry_content = $item["description"];
}
// WTF
if (is_array($entry_content)) {
$entry_content = $entry_content["encoded"];
if (!$entry_content) {
$entry_content = $entry_content["escaped"];
}
}
}
if ($cache_images && is_writable(CACHE_DIR . '/images')) {
$entry_content = cache_images($entry_content, $site_url, $debug_enabled);
}
if ($_REQUEST["xdebug"] == 2) {
print "update_rss_feed: content: ";
print $entry_content;
print "\n";
}
$entry_content_unescaped = $entry_content;
if ($use_simplepie) {
$entry_comments = strip_tags($item->data["comments"]);
if ($item->get_author()) {
$entry_author_item = $item->get_author();
$entry_author = $entry_author_item->get_name();
if (!$entry_author) {
$entry_author = $entry_author_item->get_email();
}
$entry_author = db_escape_string($entry_author);
}
} else {
$entry_comments = strip_tags($item["comments"]);
$entry_author = db_escape_string(strip_tags($item['dc']['creator']));
if ($item['author']) {
if (is_array($item['author'])) {
if (!$entry_author) {
$entry_author = db_escape_string(strip_tags($item['author']['name']));
}
if (!$entry_author) {
$entry_author = db_escape_string(strip_tags($item['author']['email']));
}
}
if (!$entry_author) {
$entry_author = db_escape_string(strip_tags($item['author']));
}
}
}
if (preg_match('/^[\\t\\n\\r ]*$/', $entry_author)) {
$entry_author = '';
}
$entry_guid = db_escape_string(strip_tags($entry_guid));
$entry_guid = mb_substr($entry_guid, 0, 250);
$result = db_query($link, "SELECT id FROM\tttrss_entries\n\t\t\t\t\tWHERE guid = '{$entry_guid}'");
$entry_content = db_escape_string($entry_content, false);
$content_hash = "SHA1:" . sha1(strip_tags($entry_content));
$entry_title = db_escape_string($entry_title);
$entry_link = db_escape_string($entry_link);
$entry_comments = mb_substr(db_escape_string($entry_comments), 0, 250);
$entry_author = mb_substr($entry_author, 0, 250);
if ($use_simplepie) {
$num_comments = 0;
#FIXME#
} else {
$num_comments = db_escape_string($item["slash"]["comments"]);
}
if (!$num_comments) {
$num_comments = 0;
}
if ($debug_enabled) {
_debug("update_rss_feed: looking for tags [1]...");
}
// parse <category> entries into tags
$additional_tags = array();
if ($use_simplepie) {
$additional_tags_src = $item->get_categories();
if (is_array($additional_tags_src)) {
foreach ($additional_tags_src as $tobj) {
array_push($additional_tags, $tobj->get_term());
}
}
if ($debug_enabled) {
_debug("update_rss_feed: category tags:");
print_r($additional_tags);
}
} else {
$t_ctr = $item['category#'];
if ($t_ctr == 0) {
$additional_tags = array();
} else {
if ($t_ctr > 0) {
$additional_tags = array($item['category']);
if ($item['category@term']) {
array_push($additional_tags, $item['category@term']);
}
for ($i = 0; $i <= $t_ctr; $i++) {
if ($item["category#{$i}"]) {
array_push($additional_tags, $item["category#{$i}"]);
}
if ($item["category#{$i}@term"]) {
array_push($additional_tags, $item["category#{$i}@term"]);
}
}
}
}
// parse <dc:subject> elements
$t_ctr = $item['dc']['subject#'];
if ($t_ctr > 0) {
array_push($additional_tags, $item['dc']['subject']);
for ($i = 0; $i <= $t_ctr; $i++) {
if ($item['dc']["subject#{$i}"]) {
array_push($additional_tags, $item['dc']["subject#{$i}"]);
}
}
}
}
if ($debug_enabled) {
_debug("update_rss_feed: looking for tags [2]...");
}
/* taaaags */
// <a href="..." rel="tag">Xorg</a>, //
$entry_tags = null;
preg_match_all("/<a.*?rel=['\"]tag['\"].*?\\>([^<]+)<\\/a>/i", $entry_content_unescaped, $entry_tags);
$entry_tags = $entry_tags[1];
$entry_tags = array_merge($entry_tags, $additional_tags);
$entry_tags = array_unique($entry_tags);
for ($i = 0; $i < count($entry_tags); $i++) {
$entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
}
if ($debug_enabled) {
//_debug("update_rss_feed: unfiltered tags found:");
//print_r($entry_tags);
}
# sanitize content
$entry_content = sanitize_article_content($entry_content);
$entry_title = sanitize_article_content($entry_title);
if ($debug_enabled) {
_debug("update_rss_feed: done collecting data [TITLE:{$entry_title}]");
}
db_query($link, "BEGIN");
if (db_num_rows($result) == 0) {
if ($debug_enabled) {
_debug("update_rss_feed: base guid not found");
}
// base post entry does not exist, create it
$result = db_query($link, "INSERT INTO ttrss_entries\n\t\t\t\t\t\t\t(title,\n\t\t\t\t\t\t\tguid,\n\t\t\t\t\t\t\tlink,\n\t\t\t\t\t\t\tupdated,\n\t\t\t\t\t\t\tcontent,\n\t\t\t\t\t\t\tcontent_hash,\n\t\t\t\t\t\t\tno_orig_date,\n\t\t\t\t\t\t\tdate_updated,\n\t\t\t\t\t\t\tdate_entered,\n\t\t\t\t\t\t\tcomments,\n\t\t\t\t\t\t\tnum_comments,\n\t\t\t\t\t\t\tauthor)\n\t\t\t\t\t\tVALUES\n\t\t\t\t\t\t\t('{$entry_title}',\n\t\t\t\t\t\t\t'{$entry_guid}',\n\t\t\t\t\t\t\t'{$entry_link}',\n\t\t\t\t\t\t\t'{$entry_timestamp_fmt}',\n\t\t\t\t\t\t\t'{$entry_content}',\n\t\t\t\t\t\t\t'{$content_hash}',\n\t\t\t\t\t\t\t{$no_orig_date},\n\t\t\t\t\t\t\tNOW(),\n\t\t\t\t\t\t\tNOW(),\n\t\t\t\t\t\t\t'{$entry_comments}',\n\t\t\t\t\t\t\t'{$num_comments}',\n\t\t\t\t\t\t\t'{$entry_author}')");
} else {
// we keep encountering the entry in feeds, so we need to
// update date_updated column so that we don't get horrible
// dupes when the entry gets purged and reinserted again e.g.
// in the case of SLOW SLOW OMG SLOW updating feeds
$base_entry_id = db_fetch_result($result, 0, "id");
db_query($link, "UPDATE ttrss_entries SET date_updated = NOW()\n\t\t\t\t\t\tWHERE id = '{$base_entry_id}'");
}
// now it should exist, if not - bad luck then
$result = db_query($link, "SELECT\n\t\t\t\t\t\tid,content_hash,no_orig_date,title,\n\t\t\t\t\t\t" . SUBSTRING_FOR_DATE . "(date_updated,1,19) as date_updated,\n\t\t\t\t\t\t" . SUBSTRING_FOR_DATE . "(updated,1,19) as updated,\n\t\t\t\t\t\tnum_comments\n\t\t\t\t\tFROM\n\t\t\t\t\t\tttrss_entries\n\t\t\t\t\tWHERE guid = '{$entry_guid}'");
$entry_ref_id = 0;
$entry_int_id = 0;
if (db_num_rows($result) == 1) {
if ($debug_enabled) {
_debug("update_rss_feed: base guid found, checking for user record");
}
// this will be used below in update handler
$orig_content_hash = db_fetch_result($result, 0, "content_hash");
$orig_title = db_fetch_result($result, 0, "title");
$orig_num_comments = db_fetch_result($result, 0, "num_comments");
$orig_date_updated = strtotime(db_fetch_result($result, 0, "date_updated"));
$ref_id = db_fetch_result($result, 0, "id");
$entry_ref_id = $ref_id;
// check for user post link to main table
// do we allow duplicate posts with same GUID in different feeds?
if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
$dupcheck_qpart = "AND (feed_id = '{$feed}' OR feed_id IS NULL)";
} else {
$dupcheck_qpart = "";
}
/* Collect article tags here so we could filter by them: */
$article_filters = get_article_filters($filters, $entry_title, $entry_content, $entry_link, $entry_timestamp, $entry_author, $entry_tags);
if ($debug_enabled) {
_debug("update_rss_feed: article filters: ");
if (count($article_filters) != 0) {
print_r($article_filters);
}
}
if (find_article_filter($article_filters, "filter")) {
db_query($link, "COMMIT");
// close transaction in progress
continue;
}
$score = calculate_article_score($article_filters);
if ($debug_enabled) {
_debug("update_rss_feed: initial score: {$score}");
}
$query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE\n\t\t\t\t\t\t\tref_id = '{$ref_id}' AND owner_uid = '{$owner_uid}'\n\t\t\t\t\t\t\t{$dupcheck_qpart}";
// if ($_REQUEST["xdebug"]) print "$query\n";
$result = db_query($link, $query);
// okay it doesn't exist - create user entry
if (db_num_rows($result) == 0) {
if ($debug_enabled) {
_debug("update_rss_feed: user record not found, creating...");
}
if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
$unread = 'true';
$last_read_qpart = 'NULL';
} else {
$unread = 'false';
$last_read_qpart = 'NOW()';
}
if (find_article_filter($article_filters, 'mark') || $score > 1000) {
$marked = 'true';
} else {
$marked = 'false';
}
if (find_article_filter($article_filters, 'publish')) {
$published = 'true';
} else {
$published = 'false';
}
// N-grams
if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
$result = db_query($link, "SELECT COUNT(*) AS similar FROM\n\t\t\t\t\t\t\t\t\tttrss_entries,ttrss_user_entries\n\t\t\t\t\t\t\t\tWHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'\n\t\t\t\t\t\t\t\t\tAND similarity(title, '{$entry_title}') >= " . _NGRAM_TITLE_DUPLICATE_THRESHOLD . "\n\t\t\t\t\t\t\t\t\tAND owner_uid = {$owner_uid}");
$ngram_similar = db_fetch_result($result, 0, "similar");
if ($debug_enabled) {
_debug("update_rss_feed: N-gram similar results: {$ngram_similar}");
}
if ($ngram_similar > 0) {
$unread = 'false';
}
}
$result = db_query($link, "INSERT INTO ttrss_user_entries\n\t\t\t\t\t\t\t\t(ref_id, owner_uid, feed_id, unread, last_read, marked,\n\t\t\t\t\t\t\t\t\tpublished, score, tag_cache, label_cache, uuid)\n\t\t\t\t\t\t\tVALUES ('{$ref_id}', '{$owner_uid}', '{$feed}', {$unread},\n\t\t\t\t\t\t\t\t{$last_read_qpart}, {$marked}, {$published}, '{$score}', '', '', '')");
if (PUBSUBHUBBUB_HUB && $published == 'true') {
$rss_link = get_self_url_prefix() . "/public.php?op=rss&id=-2&key=" . get_feed_access_key($link, -2, false, $owner_uid);
$p = new Publisher(PUBSUBHUBBUB_HUB);
$pubsub_result = $p->publish_update($rss_link);
}
$result = db_query($link, "SELECT int_id FROM ttrss_user_entries WHERE\n\t\t\t\t\t\t\t\tref_id = '{$ref_id}' AND owner_uid = '{$owner_uid}' AND\n\t\t\t\t\t\t\t\tfeed_id = '{$feed}' LIMIT 1");
if (db_num_rows($result) == 1) {
$entry_int_id = db_fetch_result($result, 0, "int_id");
}
} else {
if ($debug_enabled) {
_debug("update_rss_feed: user record FOUND");
}
$entry_ref_id = db_fetch_result($result, 0, "ref_id");
$entry_int_id = db_fetch_result($result, 0, "int_id");
}
if ($debug_enabled) {
_debug("update_rss_feed: RID: {$entry_ref_id}, IID: {$entry_int_id}");
}
$post_needs_update = false;
$update_insignificant = false;
if ($orig_num_comments != $num_comments) {
$post_needs_update = true;
$update_insignificant = true;
}
if ($content_hash != $orig_content_hash) {
$post_needs_update = true;
$update_insignificant = false;
}
if (db_escape_string($orig_title) != $entry_title) {
$post_needs_update = true;
$update_insignificant = false;
}
// if post needs update, update it and mark all user entries
// linking to this post as updated
if ($post_needs_update) {
if (defined('DAEMON_EXTENDED_DEBUG')) {
_debug("update_rss_feed: post {$entry_guid} needs update...");
}
// print "<!-- post $orig_title needs update : $post_needs_update -->";
db_query($link, "UPDATE ttrss_entries\n\t\t\t\t\t\t\tSET title = '{$entry_title}', content = '{$entry_content}',\n\t\t\t\t\t\t\t\tcontent_hash = '{$content_hash}',\n\t\t\t\t\t\t\t\tupdated = '{$entry_timestamp_fmt}',\n\t\t\t\t\t\t\t\tnum_comments = '{$num_comments}'\n\t\t\t\t\t\t\tWHERE id = '{$ref_id}'");
if (!$update_insignificant) {
if ($mark_unread_on_update) {
db_query($link, "UPDATE ttrss_user_entries\n\t\t\t\t\t\t\t\t\tSET last_read = null, unread = true WHERE ref_id = '{$ref_id}'");
} else {
if ($update_on_checksum_change) {
db_query($link, "UPDATE ttrss_user_entries\n\t\t\t\t\t\t\t\t\tSET last_read = null WHERE ref_id = '{$ref_id}'\n\t\t\t\t\t\t\t\t\t\tAND unread = false");
}
}
}
}
}
db_query($link, "COMMIT");
if ($debug_enabled) {
_debug("update_rss_feed: assigning labels...");
}
assign_article_to_labels($link, $entry_ref_id, $article_filters, $owner_uid);
if ($debug_enabled) {
_debug("update_rss_feed: looking for enclosures...");
}
// enclosures
$enclosures = array();
if ($use_simplepie) {
$encs = $item->get_enclosures();
if (is_array($encs)) {
foreach ($encs as $e) {
$e_item = array($e->link, $e->type, $e->length);
array_push($enclosures, $e_item);
}
}
} else {
// <enclosure>
$e_ctr = $item['enclosure#'];
if ($e_ctr > 0) {
$e_item = array($item['enclosure@url'], $item['enclosure@type'], $item['enclosure@length']);
array_push($enclosures, $e_item);
for ($i = 0; $i <= $e_ctr; $i++) {
if ($item["enclosure#{$i}@url"]) {
$e_item = array($item["enclosure#{$i}@url"], $item["enclosure#{$i}@type"], $item["enclosure#{$i}@length"]);
array_push($enclosures, $e_item);
}
}
}
// <media:content>
// can there be many of those? yes -fox
$m_ctr = $item['media']['content#'];
if ($m_ctr > 0) {
$e_item = array($item['media']['content@url'], $item['media']['content@medium'], $item['media']['content@length']);
array_push($enclosures, $e_item);
for ($i = 0; $i <= $m_ctr; $i++) {
if ($item["media"]["content#{$i}@url"]) {
$e_item = array($item["media"]["content#{$i}@url"], $item["media"]["content#{$i}@medium"], $item["media"]["content#{$i}@length"]);
array_push($enclosures, $e_item);
}
}
}
}
if ($debug_enabled) {
_debug("update_rss_feed: article enclosures:");
print_r($enclosures);
}
db_query($link, "BEGIN");
foreach ($enclosures as $enc) {
$enc_url = db_escape_string($enc[0]);
$enc_type = db_escape_string($enc[1]);
$enc_dur = db_escape_string($enc[2]);
$result = db_query($link, "SELECT id FROM ttrss_enclosures\n\t\t\t\t\t\tWHERE content_url = '{$enc_url}' AND post_id = '{$entry_ref_id}'");
if (db_num_rows($result) == 0) {
db_query($link, "INSERT INTO ttrss_enclosures\n\t\t\t\t\t\t\t(content_url, content_type, title, duration, post_id) VALUES\n\t\t\t\t\t\t\t('{$enc_url}', '{$enc_type}', '', '{$enc_dur}', '{$entry_ref_id}')");
}
}
db_query($link, "COMMIT");
// check for manual tags (we have to do it here since they're loaded from filters)
foreach ($article_filters as $f) {
if ($f["type"] == "tag") {
$manual_tags = trim_array(explode(",", $f["param"]));
foreach ($manual_tags as $tag) {
if (tag_is_valid($tag)) {
array_push($entry_tags, $tag);
}
}
}
}
// Skip boring tags
$boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link, 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
$filtered_tags = array();
$tags_to_cache = array();
if ($entry_tags && is_array($entry_tags)) {
foreach ($entry_tags as $tag) {
if (array_search($tag, $boring_tags) === false) {
array_push($filtered_tags, $tag);
}
}
}
$filtered_tags = array_unique($filtered_tags);
if ($debug_enabled) {
_debug("update_rss_feed: filtered article tags:");
print_r($filtered_tags);
}
// Save article tags in the database
if (count($filtered_tags) > 0) {
db_query($link, "BEGIN");
foreach ($filtered_tags as $tag) {
$tag = sanitize_tag($tag);
$tag = db_escape_string($tag);
if (!tag_is_valid($tag)) {
continue;
}
$result = db_query($link, "SELECT id FROM ttrss_tags\n\t\t\t\t\t\t\tWHERE tag_name = '{$tag}' AND post_int_id = '{$entry_int_id}' AND\n\t\t\t\t\t\t\towner_uid = '{$owner_uid}' LIMIT 1");
if ($result && db_num_rows($result) == 0) {
db_query($link, "INSERT INTO ttrss_tags\n\t\t\t\t\t\t\t\t\t(owner_uid,tag_name,post_int_id)\n\t\t\t\t\t\t\t\t\tVALUES ('{$owner_uid}','{$tag}', '{$entry_int_id}')");
}
array_push($tags_to_cache, $tag);
}
/* update the cache */
$tags_to_cache = array_unique($tags_to_cache);
$tags_str = db_escape_string(join(",", $tags_to_cache));
db_query($link, "UPDATE ttrss_user_entries\n\t\t\t\t\t\tSET tag_cache = '{$tags_str}' WHERE ref_id = '{$entry_ref_id}'\n\t\t\t\t\t\tAND owner_uid = {$owner_uid}");
db_query($link, "COMMIT");
}
if ($debug_enabled) {
_debug("update_rss_feed: article processed");
}
}
if (!$last_updated) {
if ($debug_enabled) {
_debug("update_rss_feed: new feed, catching it up...");
}
catchup_feed($link, $feed, false, $owner_uid);
}
if ($debug_enabled) {
_debug("purging feed...");
}
purge_feed($link, $feed, 0, $debug_enabled);
db_query($link, "UPDATE ttrss_feeds\n\t\t\t\tSET last_updated = NOW(), last_error = '' WHERE id = '{$feed}'");
// db_query($link, "COMMIT");
} else {
if ($use_simplepie) {
$error_msg = mb_substr($rss->error(), 0, 250);
} else {
$error_msg = mb_substr(magpie_error(), 0, 250);
}
if ($debug_enabled) {
_debug("update_rss_feed: error fetching feed: {$error_msg}");
}
$error_msg = db_escape_string($error_msg);
db_query($link, "UPDATE ttrss_feeds SET last_error = '{$error_msg}',\n\t\t\t\t\tlast_updated = NOW() WHERE id = '{$feed}'");
}
if ($use_simplepie) {
unset($rss);
}
if ($debug_enabled) {
_debug("update_rss_feed: done");
}
}