<?php
/*
*
* Basic LastFM Data Dump
* This file will download your last.fm public data
*
*/
header('Content-type: text/plain');
include 'class.feedparser.php';
$feed = new FeedParser();
$limit = 1;
$user = '******';
$stream = 'user.getrecenttracks';
$data = $feed->lastfm($user, $stream, $limit);
$fh = fopen('lastfm.log', 'w');
fwrite($fh, print_r($data, 1));
fclose($fh);
echo "Downloaded " . count($data['posts']) . " posts\n";
echo "Bandwidth Usage: {$feed->bandwidth}\n";
echo "Requests: {$feed->requests}\n";
echo "\n";
if ($_SERVER['HTTP_USER_AGENT']) {
print_r($data);
}
function execute(&$controller, &$request, &$user)
{
$member = $user->getAttribute('member', GLU_NS);
// my feeds
$source = DB_DataObject::factory('source');
$source->whereAdd('member_id = ' . $source->escape($member->id));
if ($request->hasParameter('id')) {
$source->whereAdd('id = ' . $source->escape($request->getParameter('id')));
}
$source->find();
$feeds = array();
while ($source->fetch()) {
$rss = new FeedParser($source->uri);
$rss->parse();
foreach ($rss->getItems() as $item) {
$feed['id'] = $source->id;
$feed['title'] = $item['title'];
$feed['link'] = $item['link'];
$feed['description'] = isset($item['description']) ? $item['description'] : '';
$feed['date'] = isset($item['dc:date']) ? $item['dc:date'] : (isset($item['date']) ? $item['date'] : '');
$feeds[] = $feed;
}
}
$haj = new HTML_AJAX_JSON();
$output = $haj->encode($feeds);
header('Content-Type: application/x-javascript; charset=utf-8');
echo $output;
return VIEW_NONE;
}
function testAtom03()
{
$uri = 'http://plnet.jp/test_feed/blogger-atom03.xml';
$feed = new FeedParser($uri);
$feed->parse();
$this->assertEquals('p0t blogger', $feed->getTitle());
$this->assertEquals('http://komagata.blogspot.com', $feed->getLink());
$this->assertEquals('http://komagata.blogspot.com/favicon.ico', $feed->getFavicon());
//print_r($feed->data);
}
/**
* MyBB 1.8
* Copyright 2014 MyBB Group, All Rights Reserved
*
* Website: http://www.mybb.com
* License: http://www.mybb.com/about/license
*
*/
function task_versioncheck($task)
{
global $cache, $lang, $mybb;
$current_version = rawurlencode($mybb->version_code);
$updated_cache = array('last_check' => TIME_NOW);
// Check for the latest version
require_once MYBB_ROOT . 'inc/class_xml.php';
$contents = fetch_remote_file("http://www.mybb.com/version_check.php");
if (!$contents) {
add_task_log($task, $lang->task_versioncheck_ran_errors);
return false;
}
$pos = strpos($contents, "<");
if ($pos > 1) {
$contents = substr($contents, $pos);
}
$pos = strpos(strrev($contents), ">");
if ($pos > 1) {
$contents = substr($contents, 0, -1 * ($pos - 1));
}
$parser = new XMLParser($contents);
$tree = $parser->get_tree();
$latest_code = (int) $tree['mybb']['version_code']['value'];
$latest_version = "<strong>" . htmlspecialchars_uni($tree['mybb']['latest_version']['value']) . "</strong> (" . $latest_code . ")";
if ($latest_code > $mybb->version_code) {
$latest_version = "<span style=\"color: #C00;\">" . $latest_version . "</span>";
$version_warn = 1;
$updated_cache['latest_version'] = $latest_version;
$updated_cache['latest_version_code'] = $latest_code;
} else {
$latest_version = "<span style=\"color: green;\">" . $latest_version . "</span>";
}
// Check for the latest news
require_once MYBB_ROOT . "inc/class_feedparser.php";
$feed_parser = new FeedParser();
$feed_parser->parse_feed("http://feeds.feedburner.com/MyBBDevelopmentBlog");
$updated_cache['news'] = array();
require_once MYBB_ROOT . '/inc/class_parser.php';
$post_parser = new postParser();
if ($feed_parser->error == '') {
foreach ($feed_parser->items as $item) {
if (isset($updated_cache['news'][2])) {
break;
}
$description = $item['description'];
$description = $post_parser->parse_message($description, array('allow_html' => true));
$description = preg_replace('#<img(.*)/>#', '', $description);
$updated_cache['news'][] = array('title' => htmlspecialchars_uni($item['title']), 'description' => $description, 'link' => htmlspecialchars_uni($item['link']), 'author' => htmlspecialchars_uni($item['author']), 'dateline' => $item['date_timestamp']);
}
}
$cache->update("update_check", $updated_cache);
add_task_log($task, $lang->task_versioncheck_ran);
}
function execute(&$controller, &$request, &$user)
{
$uri = $request->getParameter('uri');
$feed = new FeedParser($uri);
$response = '';
if ($feed->isValid() === false) {
$response = 'false';
} else {
$response = 'true';
}
header('Content-Type: text/plain; charset=utf-8');
echo $response;
return VIEW_NONE;
}
function parse()
{
$feed = DB_DataObject::factory('feed');
$feed->uri = $this->uri;
// was cached
if ($feed->count() > 0) {
$this->struct = $this->_getCache($this->uri);
// no cache
} else {
$f = new FeedParser($this->uri);
$f->parse();
$this->struct = $f->toArray();
}
}
public function executeRefresh()
{
$post = PostPeer::retrieveByPK($this->getRequestParameter('id'));
$blog = $post->getBlog();
$this->forward404Unless($blog->getId() == $this->getUser()->getId());
if ($post) {
try {
$items = FeedParser::parse($blog->getFeed());
foreach ($items as $item) {
if ($item->link == $post->getLink()) {
$post->removeTags();
foreach ($item->tags as $name) {
$tag = TagPeer::retriveByName($name, true);
$post->addTag($tag);
}
if ($post->hasTags()) {
$shortened = $post->setFullContent($item->content);
$post->setLink($item->link);
$post->setTitle($item->title);
$post->setCreatedAt($item->pubdate);
$post->setShortened($shortened);
$post->save();
}
break;
}
}
$this->setFlash('updated', 'Wpis został odświeżony.');
} catch (Exception $e) {
$this->setFlash('updated', 'Odswieżanie wpisu nie powiodło się.');
}
}
$this->redirect('ucp/index');
}
/**
* Displays a static page.
*/
function display()
{
$path = func_get_args();
$count = count($path);
if (!$count) {
$this->redirect('/');
}
$page = $subpage = $title_for_layout = null;
if (!empty($path[0])) {
$page = $path[0];
}
if (!empty($path[1])) {
$subpage = $path[1];
}
if (!empty($path[$count - 1])) {
$this->title_for_layout = Inflector::humanize($path[$count - 1]);
if ($this->title_for_layout == 'Home') {
$this->title_for_layout = '';
}
}
App::import('Vendor', 'feedparser/feedparser');
$Parser = new FeedParser();
$Parser->parse('http://blog.simpleve.com/feed/');
$items = $Parser->getItems();
$this->set('blogrss', $items);
$Parser = new FeedParser();
$Parser->parse('http://www.eveonline.com/feed/rdfdevblog.asp');
$items = $Parser->getItems();
$this->set('evedevblogrss', $items);
$motd = file_get_contents('http://www.eveonline.com/motd.asp?s=xml');
$motd = trim(str_replace('MOTD', '', $motd));
$motd = trim(str_replace('shellexec:', '', $motd));
$motd = trim(str_replace('<center>', '', $motd));
$motd = trim(str_replace('</center>', '', $motd));
$this->set('evemotd', $motd);
$this->set(compact('page', 'subpage', 'title_for_layout'));
$this->render(implode('/', $path));
}
function subscribeRoute($feedUrl, $parentFolderId)
{
// Check to see if the system already has the feed
$storage = Storage::getInstance();
$feed = $storage->getFeed($feedUrl);
if (!$feed) {
$matches = $storage->findFeedFromLinks($feedUrl);
if (count($matches) > 0) {
$feed = $matches[0];
}
// TODO: allow selection if > 1
}
if (!$feed) {
// Not in the system. Fetch it from www
// Check URL for validity
if (!filter_var($feedUrl, FILTER_VALIDATE_URL)) {
throw new JsonError(l("Incorrect or unrecognized URL"));
}
// Fetch and parse the feed
try {
$parser = FeedParser::create($feedUrl, true);
} catch (Exception $e) {
throw new JsonError(l("Could not read contents of feed"));
}
if (!$parser) {
throw new JsonError(l("Could not determine type of feed"));
}
try {
$feed = $parser->parse();
} catch (Exception $e) {
throw new JsonError(l("Could not parse the contents of the feed"));
}
// Import the feed contents
$feed->id = $storage->importFeed($this->user->id, $feed);
if ($feed->id === false) {
throw new JsonError(l("An error occurred while adding feed"));
}
// Import any links
$links = $parser->getLinks();
if (!in_array($feed->link, $links)) {
$links[] = $feed->link;
}
$storage->addLinks($feed->id, $links);
}
// Subscribe to feed
if (!$storage->subscribeToFeed($this->user->id, $feed->id, $parentFolderId)) {
throw new JsonError(l("Could not subscribe to feed"));
}
return array("feed" => array("title" => $feed->title), "allItems" => $storage->getUserFeeds($this->user));
}
private function parseFeed($feedUrl)
{
$now = microtime(true);
$parser = FeedParser::create($feedUrl);
$this->secondsSpentDownloading += microtime(true) - $now;
if (!$parser) {
throw new Exception("Could not determine type of feed from content");
}
$now = microtime(true);
echo " (-) parsing (" . get_class($parser) . ") ... ";
$feed = $parser->parse();
$secondsSpentParsingFeed = microtime(true) - $now;
$this->secondsSpentParsing += $secondsSpentParsingFeed;
echo sprintf("done! (%.04fs)\n", $secondsSpentParsingFeed);
return $feed;
}
/**
* Loads images from a MediaRSS or ATOM feed
*/
function _loadRSS($url)
{
require_once DOKU_INC . 'inc/FeedParser.php';
$feed = new FeedParser();
$feed->set_feed_url($url);
$feed->init();
$files = array();
// base url to use for broken feeds with non-absolute links
$main = parse_url($url);
$host = $main['scheme'] . '://' . $main['host'] . ($main['port'] ? ':' . $main['port'] : '');
$path = dirname($main['path']) . '/';
foreach ($feed->get_items() as $item) {
if ($enclosure = $item->get_enclosure()) {
// skip non-image enclosures
if ($enclosure->get_type()) {
if (substr($enclosure->get_type(), 0, 5) != 'image') {
continue;
}
} else {
if (!preg_match('/\\.(jpe?g|png|gif)(\\?|$)/i', $enclosure->get_link())) {
continue;
}
}
// non absolute links
$ilink = $enclosure->get_link();
if (!preg_match('/^https?:\\/\\//i', $ilink)) {
if ($ilink[0] == '/') {
$ilink = $host . $ilink;
} else {
$ilink = $host . $path . $ilink;
}
}
$link = $item->link;
if (!preg_match('/^https?:\\/\\//i', $link)) {
if ($link[0] == '/') {
$link = $host . $link;
} else {
$link = $host . $path . $link;
}
}
$files[] = array('id' => $ilink, 'isimg' => true, 'file' => basename($ilink), 'title' => SimplePie_Misc::htmlspecialchars_decode($enclosure->get_title(), ENT_COMPAT), 'desc' => strip_tags(SimplePie_Misc::htmlspecialchars_decode($enclosure->get_description(), ENT_COMPAT)), 'width' => $enclosure->get_width(), 'height' => $enclosure->get_height(), 'mtime' => $item->get_date('U'), 'ctime' => $item->get_date('U'), 'detail' => $link);
}
}
return $files;
}
<?php require_once dirname(dirname(__FILE__)) . '/webapp/config.php'; require_once 'FeedParser.php'; $uri = $_SERVER['argv'][1]; $feed = new FeedParser($uri); $res = $feed->parse(); echo "favicon: " . $feed->getFavicon() . "\n";
/**
* Render the output of an RSS feed
*
* @param string $url URL of the feed
* @param array $params Finetuning of the output
*/
function rss($url, $params)
{
global $lang;
require_once DOKU_INC . 'inc/FeedParser.php';
$feed = new FeedParser();
$feed->feed_url($url);
//disable warning while fetching
$elvl = null;
if (!defined('DOKU_E_LEVEL')) {
$elvl = error_reporting(E_ERROR);
}
$rc = $feed->init();
if (!defined('DOKU_E_LEVEL')) {
error_reporting($elvl);
}
//decide on start and end
if ($params['reverse']) {
$mod = -1;
$start = $feed->get_item_quantity() - 1;
$end = $start - $params['max'];
$end = $end < -1 ? -1 : $end;
} else {
$mod = 1;
$start = 0;
$end = $feed->get_item_quantity();
$end = $end > $params['max'] ? $params['max'] : $end;
}
$this->listu_open();
if ($rc) {
for ($x = $start; $x != $end; $x += $mod) {
$item = $feed->get_item($x);
$this->listitem_open(0);
$this->listcontent_open();
$this->externallink($item->get_permalink(), $item->get_title());
if ($params['author']) {
$author = $item->get_author(0);
if ($author) {
$name = $author->get_name();
if (!$name) {
$name = $author->get_email();
}
if ($name) {
$this->cdata(' ' . $lang['by'] . ' ' . $name);
}
}
}
if ($params['date']) {
$this->cdata(' (' . $item->get_date($this->config->getParam('dformat')) . ')');
}
if ($params['details']) {
$this->cdata(strip_tags($item->get_description()));
}
$this->listcontent_close();
$this->listitem_close();
}
} else {
$this->listitem_open(0);
$this->listcontent_open();
$this->emphasis_open();
$this->cdata($lang['rssfailed']);
$this->emphasis_close();
$this->externallink($url);
$this->listcontent_close();
$this->listitem_close();
}
$this->listu_close();
}
/**
*
*
* @param Doku_Event $event Not used
* @param mixed $param Not used
* @return void
*/
public function handle(Doku_Event &$event, $param)
{
global $conf;
if ($event->data != 'feedaggregator') {
return;
}
$event->preventDefault();
// See if we need a token and whether it matches.
$requiredToken = $this->getConf('token');
$suppliedToken = isset($_GET['token']) ? $_GET['token'] : false;
if (!empty($requiredToken) && $suppliedToken !== $requiredToken) {
msg("Token doesn't match for feedaggregator");
return true;
}
// Get the feed list.
$feeds = file(fullpath($conf['tmpdir'] . '/feedaggregator.csv'));
// Set up SimplePie and merge all the feeds together.
$simplepie = new FeedParser();
$ua = 'Mozilla/4.0 (compatible; DokuWiki feedaggregator plugin ' . wl('', '', true) . ')';
$simplepie->set_useragent($ua);
$simplepie->force_feed($this->getConf('force_feed'));
$simplepie->force_fsockopen($this->getConf('force_fsockopen'));
$simplepie->set_feed_url($feeds);
// Set up caching.
$cacheDir = fullpath($conf['cachedir'] . '/feedaggregator');
io_mkdir_p($cacheDir);
$simplepie->enable_cache();
$simplepie->set_cache_location($cacheDir);
// Run the actual feed aggregation.
$simplepie->init();
// Check for errors.
if ($simplepie->error()) {
header("Content-type:text/plain");
echo join("\n", $simplepie->error());
}
// Create the output HTML and cache it for use by the syntax component.
$html = '';
foreach ($simplepie->get_items() as $item) {
$html .= "<div class='feedaggregator_item'>\n" . "<h2>" . $item->get_title() . "</h2>\n" . $item->get_content() . "\n" . "<p>" . " <a href='" . $item->get_permalink() . "'>Published " . $item->get_date('j M Y') . "</a> " . " in <a href='" . $item->get_feed()->get_permalink() . "'>" . $item->get_feed()->get_title() . "</a>" . "</p>\n" . "</div>\n\n";
}
io_saveFile($cacheDir . '/output.html', $html);
// Output nothing, as this should be run from cron and we don't want to
// flood the logs with success.
exit(0);
}
/**
* Renders an RSS feed
*
* @author Andreas Gohr <*****@*****.**>
*/
function rss($url, $params)
{
global $lang;
global $conf;
require_once DOKU_INC . 'inc/FeedParser.php';
$feed = new FeedParser();
$feed->feed_url($url);
//disable warning while fetching
if (!defined('DOKU_E_LEVEL')) {
$elvl = error_reporting(E_ERROR);
}
$rc = $feed->init();
if (!defined('DOKU_E_LEVEL')) {
error_reporting($elvl);
}
//decide on start and end
if ($params['reverse']) {
$mod = -1;
$start = $feed->get_item_quantity() - 1;
$end = $start - $params['max'];
$end = $end < -1 ? -1 : $end;
} else {
$mod = 1;
$start = 0;
$end = $feed->get_item_quantity();
$end = $end > $params['max'] ? $params['max'] : $end;
}
//$this->doc .= '<ul class="rss">';
$this->listu_open();
if ($rc) {
for ($x = $start; $x != $end; $x += $mod) {
$item = $feed->get_item($x);
//$this->doc .= '<li><div class="li">';
$this->listitem_open(1);
$this->externallink($item->get_permalink(), $item->get_title());
if ($params['author']) {
$author = $item->get_author(0);
if ($author) {
$name = $author->get_name();
if (!$name) {
$name = $author->get_email();
}
if ($name) {
$this->cdata(' ' . $lang['by'] . ' ' . $name);
}
//$this->doc .= ' '.$lang['by'].' '.$name;
}
}
if ($params['date']) {
$this->cdata(' (' . $item->get_date($conf['dformat']) . ')');
//$this->doc .= ' ('.$item->get_date($conf['dformat']).')';
}
if ($params['details']) {
//$this->doc .= '<div class="detail">';
if ($htmlok) {
$this->cdata($item->get_description());
} else {
$this->cdata(strip_tags($item->get_description()));
}
//$this->doc .= '</div>';
}
//$this->doc .= '</div></li>';
$this->listitem_close();
}
} else {
//$this->doc .= '<li><div class="li">';
$this->listitem_open(1);
$this->emphasis_open();
//$this->doc .= '<em>'.$lang['rssfailed'].'</em>';
$this->cdata($lang['rssfailed']);
$this->emphasis_close();
$this->externallink($url);
// if($conf['allowdebug']){
// $this->doc .= '<!--'.hsc($feed->error).'-->';
// }
// $this->doc .= '</div></li>';
$this->listitem_close();
}
// $this->doc .= '</ul>';
$this->listu_close();
}
private static function extractLinks($sourceUrl, $html)
{
$links = array();
if (preg_match_all("/<link(?:\\s+\\w+=[\"'][^\"']*[\"'])+\\s*\\/?>/", $html, $matches)) {
foreach ($matches[0] as $match) {
if (preg_match_all("/(\\w+)=[\"']([^\"']*)[\"']/", $match, $submatches, PREG_SET_ORDER)) {
$attrs = array();
foreach ($submatches as $submatch) {
list(, $key, $value) = $submatch;
$attrs[strtolower($key)] = $value;
}
if (strcasecmp($attrs["rel"], "alternate") === 0 && (strcasecmp($attrs["type"], "application/rss+xml") === 0 || strcasecmp($attrs["type"], "application/atom+xml") === 0)) {
$link = new stdClass();
$link->title = $attrs["title"];
$link->url = FeedParser::resolveUrl($sourceUrl, $attrs["href"]);
$links[] = $link;
}
}
}
}
return $links;
}
<?php
// The only thing we need is include base class file
require_once 'FeedParser.php';
// Get XML serialization of feed
$xml = file_get_contents($_POST['filename']);
// This is great. To work with feed we invoke only base class. All other work is
// transparent.
$feed = new FeedParser($xml);
//Because we have interface for feeds, we invoke interface methods
echo '<b>Type:</b>' . $feed->getFeedType() . "<br/>";
echo '<b>Title:</b>' . $feed->getTitle() . "<br/>";
echo '<b>Description:</b>' . $feed->getDescription() . "<br/>";
echo '<b>Feed link:</b>' . $feed->getFeedLink() . "<br/>";
echo '<b>Link:</b>' . $feed->getLink() . "<br/>";
$items = $feed->getItems();
// Stuff in your items can be empty, so you should somehow handle it.
// I've prepared is_empty function for you - enjoy.
$i = 1;
foreach ($items as $item) {
//Because we have interface for items, we invoke interface methods
echo "<h1>";
if (is_empty($item->getLink())) {
echo '<a href="#">';
} else {
echo '<a href="' . $item->getLink() . '">';
}
if (is_empty($item->getTitle())) {
echo "No title";
} else {
echo "{$i}. " . $item->getTitle();
<?php
/*
*
* Simple Example of how to use Feed Parser Class
* by Marc Qualie
*
*/
include 'class.feedparser.php';
$feed = new FeedParser();
// Two different ways of getting blogger feeds, both end up in at the same url
$blog1 = $feed->blogger('blog.marcqualie.com', 1);
$blog2 = $feed->blogger('marcqualie', 1);
// Twitter via Username
$tweets = $feed->twitter('marcqualie', 1);
// Output
header('Content-type: text/plain');
echo "Bandwidth: {$feed->bandwidth}\n";
echo "Requests: {$feed->requests}\n";
echo "\n";
// Dump
$output = 'blog1 = ' . print_r($blog1, 1) . "\n";
$output .= 'blog2 = ' . print_r($blog2, 1) . "\n";
$output .= 'tweets = ' . print_r($tweets, 1) . "\n";
$fh = fopen('example.log', 'w');
fwrite($fh, $output);
fclose($fh);
set_time_limit(0);
define('SF_ROOT_DIR', realpath(dirname(__FILE__) . '/..'));
define('SF_APP', 'backend');
define('SF_ENVIRONMENT', 'prod');
define('SF_DEBUG', false);
require_once SF_ROOT_DIR . DIRECTORY_SEPARATOR . 'apps' . DIRECTORY_SEPARATOR . SF_APP . DIRECTORY_SEPARATOR . 'config' . DIRECTORY_SEPARATOR . 'config.php';
$databaseManager = new sfDatabaseManager();
$databaseManager->initialize();
$blogs = BlogPeer::getApproved();
logmsg('Skrypt odswiezajacy');
logmsg('Feedow do sprawdzenia: %d', count($blogs));
logmsg(str_repeat('-', 80));
foreach ($blogs as $blog) {
logmsg('Parsowanie feedu %s', $blog->getFeed());
try {
$items = FeedParser::parse($blog->getFeed());
$ts = PostPeer::getNewestTimestamp($blog);
logmsg('Najnowszy wpis (timestamp): %d', $ts);
foreach ($items as $item) {
if (!parseItem($blog, $item, $ts)) {
break;
}
}
} catch (Exception $e) {
logmsg('Blad: %s', $e->getMessage());
}
logmsg(str_repeat('-', 80) . "\n");
}
logmsg('Odswiezanie zakonczone.');
function parseItem($blog, $item, $ts)
{
/**
* Renders an RSS feed
*
* @author Andreas Gohr <*****@*****.**>
*/
function rss($url, $params)
{
global $lang;
global $conf;
require_once DOKU_INC . 'inc/FeedParser.php';
$feed = new FeedParser();
$feed->set_feed_url($url);
//disable warning while fetching
if (!defined('DOKU_E_LEVEL')) {
$elvl = error_reporting(E_ERROR);
}
$rc = $feed->init();
if (isset($elvl)) {
error_reporting($elvl);
}
if ($params['nosort']) {
$feed->enable_order_by_date(false);
}
//decide on start and end
if ($params['reverse']) {
$mod = -1;
$start = $feed->get_item_quantity() - 1;
$end = $start - $params['max'];
$end = $end < -1 ? -1 : $end;
} else {
$mod = 1;
$start = 0;
$end = $feed->get_item_quantity();
$end = $end > $params['max'] ? $params['max'] : $end;
}
$this->doc .= '<ul class="rss">';
if ($rc) {
for ($x = $start; $x != $end; $x += $mod) {
$item = $feed->get_item($x);
$this->doc .= '<li><div class="li">';
// support feeds without links
$lnkurl = $item->get_permalink();
if ($lnkurl) {
// title is escaped by SimplePie, we unescape here because it
// is escaped again in externallink() FS#1705
$this->externallink($item->get_permalink(), html_entity_decode($item->get_title(), ENT_QUOTES, 'UTF-8'));
} else {
$this->doc .= ' ' . $item->get_title();
}
if ($params['author']) {
$author = $item->get_author(0);
if ($author) {
$name = $author->get_name();
if (!$name) {
$name = $author->get_email();
}
if ($name) {
$this->doc .= ' ' . $lang['by'] . ' ' . $name;
}
}
}
if ($params['date']) {
$this->doc .= ' (' . $item->get_local_date($conf['dformat']) . ')';
}
if ($params['details']) {
$this->doc .= '<div class="detail">';
if ($conf['htmlok']) {
$this->doc .= $item->get_description();
} else {
$this->doc .= strip_tags($item->get_description());
}
$this->doc .= '</div>';
}
$this->doc .= '</div></li>';
}
} else {
$this->doc .= '<li><div class="li">';
$this->doc .= '<em>' . $lang['rssfailed'] . '</em>';
$this->externallink($url);
if ($conf['allowdebug']) {
$this->doc .= '<!--' . hsc($feed->error) . '-->';
}
$this->doc .= '</div></li>';
}
$this->doc .= '</ul>';
}
function execute(&$controller, &$request, &$user)
{
$uri = $request->getParameter('uri');
$member = $user->getAttribute('member', GLU_NS);
$redirect = $request->hasParameter('redirect') ? $request->getParameter('redirect') : null;
$cc_id = $this->getContentCategoryIdByUri($uri) ? $this->getContentCategoryIdByUri($uri) : 8;
// get feed
$f = new FeedParser($uri);
$parse_result = $f->parse();
if ($parse_result === false) {
trigger_error(__CLASS__ . '::' . __FUNCTION__ . '(): ' . "Failed to parse feed uri: {$uri}", E_USER_NOTICE);
header('Content-Type: text/plain; charset=utf-8');
echo 'false';
return VIEW_NONE;
}
$name = $f->getTitle();
$favicon = $f->getFavicon() ? $f->getFavicon() : SCRIPT_PATH . 'images/favicon.ico';
$db =& DBUtils::connect(false);
$db->query('BEGIN');
// get feedId
$feedId = $this->getFeedIdByUri($uri);
if ($feedId === false) {
// add feed
$fields = array('uri' => $uri, 'link' => $f->getLink(), 'title' => $f->getTitle(), 'description' => $f->getDescription(), 'favicon' => $favicon, 'lastupdatedtime' => date("Y-m-d H:i:s"));
$res = $db->autoExecute('feed', $fields, DB_AUTOQUERY_INSERT);
if (DB::isError($res)) {
$db->rollback();
trigger_error(__CLASS__ . '::' . __FUNCTION__ . '(): ' . "Failed to insert. " . $res->toString(), E_USER_ERROR);
header('Content-Type: text/plain; charset=utf-8');
echo 'false';
return VIEW_NONE;
}
// specific mysql
$feedId = $db->getOne('SELECT LAST_INSERT_ID()');
}
// add member_to_feed
$m2f_fields = array('member_id' => MemberUtils::get_id_by_account($member->account), 'feed_id' => $feedId);
if (DBUtils::get('member_to_feed', $m2f_fields)) {
echo "'already exists'";
// already exists
return VIEW_NONE;
}
$res = $db->autoExecute('member_to_feed', $m2f_fields, DB_AUTOQUERY_INSERT);
if (DB::isError($res)) {
$db->rollback();
trigger_error(__CLASS__ . '::' . __FUNCTION__ . '(): ' . "Failed to insert m2f. " . $res->toString(), E_USER_WARNING);
header('Content-Type: text/plain; charset=utf-8');
echo 'false';
return VIEW_NONE;
}
// add member_to_content_category_to_feed
$m2cc2f_fields = array('member_id' => $member->id, 'content_category_id' => $cc_id, 'feed_id' => $feedId);
$sql = "SELECT count(*)\n FROM member_to_content_category_to_feed\n WHERE member_id = ?\n AND content_category_id = ?\n AND feed_id = ? ";
if ($db->getOne($sql, array_values($m2cc2f_fields)) == 0) {
$res = $db->autoExecute('member_to_content_category_to_feed', $m2cc2f_fields, DB_AUTOQUERY_INSERT);
if (DB::isError($res)) {
$db->rollback();
trigger_error(__CLASS__ . '::' . __FUNCTION__ . '(): ' . 'Failed to insert m2cc2f. ' . $res->toString(), E_USER_WARNING);
header('Content-Type: text/plain; charset=utf-8');
echo 'false';
return VIEW_NONE;
}
}
// try to crawl
$crawler =& new Crawler();
$crawl_result = $crawler->crawl($uri);
if ($crawl_result === false) {
$db->rollback();
trigger_error("AddFeedAction::execute(): Failed to crawl: {$uri}", E_USER_NOTICE);
header('Content-Type: text/plain; charset=utf-8');
echo 'false';
return VIEW_NONE;
}
$db->commit();
if ($redirect) {
Controller::redirect(SCRIPT_PATH . 'setting/feed');
} else {
header('Content-Type: text/plain; charset=utf-8');
echo 'true';
}
return VIEW_NONE;
}
<?php require_once dirname(dirname(__FILE__)) . '/webapp/config.php'; require_once 'FeedParser.php'; $uri = $_SERVER['argv'][1]; $feed = new FeedParser($uri); $res = $feed->parse(); print_r($feed->toArray());
<?php
require 'inc/config.php';
require 'inc/db.php';
require 'lib/FeedParser.class.php';
require 'lib/Parser.class.php';
echo "Loading feeds...";
$feed_parser = new FeedParser();
$i = 0;
$opml = simplexml_load_file('http://news.bbc.co.uk/rss/feeds.opml');
foreach ($opml->xpath('//outline') as $item) {
if ((string) $item['language'] == 'en-gb') {
$feed_parser->add((string) $item['xmlUrl']);
}
}
$feed_parser->add('http://www.guardian.co.uk/rss');
$feed_parser->add('http://www.guardian.co.uk/rssfeed/0,,11,00.xml');
$feed_parser->add('http://www.guardian.co.uk/rssfeed/0,,12,00.xml');
$feed_parser->add('http://www.guardian.co.uk/rssfeed/0,,5,00.xml');
$feed_parser->add('http://www.guardian.co.uk/rssfeed/0,,24,00.xml');
$feed_parser->add('http://www.guardian.co.uk/rssfeed/0,15065,19,00.xml');
$feed_parser->add('http://www.guardian.co.uk/rssfeed/0,,18,00.xml');
$feed_parser->add('http://www.guardian.co.uk/rssfeed/0,,7,00.xml');
$feed_parser->add('http://english.aljazeera.net/Services/Rss/?PostingId=2007731105943979989');
echo "DONE\n";
$parser_factory = ParserFactory::create();
function prompt($parser, $item, $category, $locations)
{
echo "\n" . $item->title() . "\n";
echo $locations[0]['name'] . ' <-> ' . $locations[1]['name'] . "\n";
if ($category) {
<?php
include 'FeedParser.php';
$Parser = new FeedParser();
$Parser->parse('http://www.sitepoint.com/rss.php');
$channels = $Parser->getChannels();
$items = $Parser->getItems();
?>
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>Testing the PHP Universal Feed Parser</title>
<style type="text/css">
body{
padding: 0px;
margin: 50px 150px;
border: 1px solid #ddd;
font-family: verdana, arial;
}
h1#title {
background-color: #eee;
border-bottom : 1px solid #ddd;
margin: 0px 0px 15px;
padding:10px;
text-align: center;
}
h1#title a{
font-size: 18px;
}
exit;
}
$latest_code = $tree['mybb']['version_code']['value'];
$latest_version = "<strong>" . $tree['mybb']['latest_version']['value'] . "</strong> (" . $latest_code . ")";
if ($latest_code > $mybb->version_code) {
$latest_version = "<span style=\"color: #C00;\">" . $latest_version . "</span>";
$version_warn = 1;
$updated_cache['latest_version'] = $latest_version;
$updated_cache['latest_version_code'] = $latest_code;
} else {
$version_warn = 0;
$latest_version = "<span style=\"color: green;\">" . $latest_version . "</span>";
}
$cache->update("update_check", $updated_cache);
require_once MYBB_ROOT . "inc/class_feedparser.php";
$feed_parser = new FeedParser();
$feed_parser->parse_feed("http://feeds.feedburner.com/MyBBDevelopmentBlog");
$table = new Table();
$table->construct_header($lang->your_version);
$table->construct_header($lang->latest_version);
$table->construct_cell("<strong>" . $mybb->version . "</strong> (" . $mybb->version_code . ")");
$table->construct_cell($latest_version);
$table->construct_row();
$table->output($lang->version_check);
if ($version_warn) {
$page->output_error("<p><em>{$lang->error_out_of_date}</em> {$lang->update_forum}</p>");
} else {
$page->output_success("<p><em>{$lang->success_up_to_date}</em></p>");
}
if ($feed_parser->error == '') {
foreach ($feed_parser->items as $item) {
function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false, $rss = false)
{
$debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
_debug_suppress(!$debug_enabled);
_debug("start", $debug_enabled);
$result = db_query("SELECT title FROM ttrss_feeds\n\t\t\tWHERE id = '{$feed}'");
$title = db_fetch_result($result, 0, "title");
// feed was batch-subscribed or something, we need to get basic info
// this is not optimal currently as it fetches stuff separately TODO: optimize
if ($title == "[Unknown]") {
_debug("setting basic feed info for {$feed}...");
set_basic_feed_info($feed);
}
$result = db_query("SELECT id,update_interval,auth_login,\n\t\t\tfeed_url,auth_pass,cache_images,\n\t\t\tmark_unread_on_update, owner_uid,\n\t\t\tpubsub_state, auth_pass_encrypted,\n\t\t\tfeed_language,\n\t\t\t(SELECT max(date_entered) FROM\n\t\t\t\tttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '{$feed}') AS last_article_timestamp\n\t\t\tFROM ttrss_feeds WHERE id = '{$feed}'");
if (db_num_rows($result) == 0) {
_debug("feed {$feed} NOT FOUND/SKIPPED", $debug_enabled);
return false;
}
$last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp"));
if (defined('_DISABLE_HTTP_304')) {
$last_article_timestamp = 0;
}
$owner_uid = db_fetch_result($result, 0, "owner_uid");
$mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, 0, "mark_unread_on_update"));
$pubsub_state = db_fetch_result($result, 0, "pubsub_state");
$auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, 0, "auth_pass_encrypted"));
db_query("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 ($auth_pass_encrypted) {
require_once "crypt.php";
$auth_pass = decrypt_string($auth_pass);
}
$cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
$fetch_url = db_fetch_result($result, 0, "feed_url");
$feed_language = db_escape_string(mb_strtolower(db_fetch_result($result, 0, "feed_language")));
if (!$feed_language) {
$feed_language = 'english';
}
$feed = db_escape_string($feed);
$date_feed_processed = date('Y-m-d H:i');
$cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml";
$pluginhost = new PluginHost();
$pluginhost->set_debug($debug_enabled);
$user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
$pluginhost->load(PLUGINS, PluginHost::KIND_ALL);
$pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid);
$pluginhost->load_data();
if ($rss && is_object($rss) && get_class($rss) == "FeedParser") {
_debug("using previously initialized parser object");
} else {
$rss_hash = false;
$force_refetch = isset($_REQUEST["force_refetch"]);
foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) {
$feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, $last_article_timestamp, $auth_login, $auth_pass);
}
// try cache
if (!$feed_data && file_exists($cache_filename) && is_readable($cache_filename) && !$auth_login && !$auth_pass && filemtime($cache_filename) > time() - 30) {
_debug("using local cache [{$cache_filename}].", $debug_enabled);
@($feed_data = file_get_contents($cache_filename));
if ($feed_data) {
$rss_hash = sha1($feed_data);
}
} else {
_debug("local cache will not be used for this feed", $debug_enabled);
}
// fetch feed from source
if (!$feed_data) {
_debug("fetching [{$fetch_url}]...", $debug_enabled);
_debug("If-Modified-Since: " . gmdate('D, d M Y H:i:s \\G\\M\\T', $last_article_timestamp), $debug_enabled);
$feed_data = fetch_file_contents($fetch_url, false, $auth_login, $auth_pass, false, $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT, $force_refetch ? 0 : $last_article_timestamp);
global $fetch_curl_used;
if (!$fetch_curl_used) {
$tmp = @gzdecode($feed_data);
if ($tmp) {
$feed_data = $tmp;
}
}
$feed_data = trim($feed_data);
_debug("fetch done.", $debug_enabled);
// cache vanilla feed data for re-use
if ($feed_data && !$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) {
$new_rss_hash = sha1($feed_data);
if ($new_rss_hash != $rss_hash) {
_debug("saving {$cache_filename}", $debug_enabled);
@file_put_contents($cache_filename, $feed_data);
}
}
}
if (!$feed_data) {
global $fetch_last_error;
global $fetch_last_error_code;
_debug("unable to fetch: {$fetch_last_error} [{$fetch_last_error_code}]", $debug_enabled);
$error_escaped = '';
// If-Modified-Since
if ($fetch_last_error_code != 304) {
$error_escaped = db_escape_string($fetch_last_error);
} else {
_debug("source claims data not modified, nothing to do.", $debug_enabled);
}
db_query("UPDATE ttrss_feeds SET last_error = '{$error_escaped}',\n\t\t\t\t\t\tlast_updated = NOW() WHERE id = '{$feed}'");
return;
}
}
foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) {
$feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed);
}
// set last update to now so if anything *simplepie* crashes later we won't be
// continuously failing on the same feed
//db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'");
if (!$rss) {
$rss = new FeedParser($feed_data);
$rss->init();
}
// print_r($rss);
$feed = db_escape_string($feed);
if (!$rss->error()) {
// We use local pluginhost here because we need to load different per-user feed plugins
$pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss);
_debug("language: {$feed_language}", $debug_enabled);
_debug("processing feed data...", $debug_enabled);
// db_query("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)";
}
$result = db_query("SELECT owner_uid,favicon_avg_color,\n\t\t\t\t(favicon_last_checked IS NULL OR {$favicon_interval_qpart}) AS\n\t\t\t\t\t\tfavicon_needs_check\n\t\t\t\tFROM ttrss_feeds WHERE id = '{$feed}'");
$favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, "favicon_needs_check"));
$favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color");
$owner_uid = db_fetch_result($result, 0, "owner_uid");
$site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
_debug("site_url: {$site_url}", $debug_enabled);
_debug("feed_title: " . $rss->get_title(), $debug_enabled);
if ($favicon_needs_check || $force_refetch) {
/* terrible hack: if we crash on floicon shit here, we won't check
* the icon avgcolor again (unless the icon got updated) */
$favicon_file = ICONS_DIR . "/{$feed}.ico";
$favicon_modified = @filemtime($favicon_file);
_debug("checking favicon...", $debug_enabled);
check_feed_favicon($site_url, $feed);
$favicon_modified_new = @filemtime($favicon_file);
if ($favicon_modified_new > $favicon_modified) {
$favicon_avg_color = '';
}
if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') {
require_once "colors.php";
db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE\n\t\t\t\t\t\t\tid = '{$feed}'");
$favicon_color = db_escape_string(calculate_avg_color($favicon_file));
$favicon_colorstring = ",favicon_avg_color = '" . $favicon_color . "'";
} else {
if ($favicon_avg_color == 'fail') {
_debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled);
}
}
db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW()\n\t\t\t\t\t{$favicon_colorstring}\n\t\t\t\t\tWHERE id = '{$feed}'");
}
_debug("loading filters & labels...", $debug_enabled);
$filters = load_filters($feed, $owner_uid);
_debug("" . count($filters) . " filters loaded.", $debug_enabled);
$items = $rss->get_items();
if (!is_array($items)) {
_debug("no articles found.", $debug_enabled);
db_query("UPDATE ttrss_feeds\n\t\t\t\t\tSET last_updated = NOW(), last_error = '' WHERE id = '{$feed}'");
return;
// no articles
}
if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) {
_debug("checking for PUSH hub...", $debug_enabled);
$feed_hub_url = false;
$links = $rss->get_links('hub');
if ($links && is_array($links)) {
foreach ($links as $l) {
$feed_hub_url = $l;
break;
}
}
_debug("feed hub url: {$feed_hub_url}", $debug_enabled);
$feed_self_url = $fetch_url;
$links = $rss->get_links('self');
if ($links && is_array($links)) {
foreach ($links as $l) {
$feed_self_url = $l;
break;
}
}
_debug("feed self url = {$feed_self_url}");
if ($feed_hub_url && $feed_self_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 = $s->subscribe($feed_self_url);
_debug("feed hub url found, subscribe request sent. [rc={$rc}]", $debug_enabled);
db_query("UPDATE ttrss_feeds SET pubsub_state = 1\n\t\t\t\t\t\tWHERE id = '{$feed}'");
}
}
_debug("processing articles...", $debug_enabled);
$tstart = time();
foreach ($items as $item) {
if ($_REQUEST['xdebug'] == 3) {
print_r($item);
}
if (ini_get("max_execution_time") > 0 && time() - $tstart >= ini_get("max_execution_time") * 0.7) {
_debug("looks like there's too many articles to process at once, breaking out", $debug_enabled);
break;
}
$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());
}
if (!$entry_guid) {
continue;
}
$entry_guid = "{$owner_uid},{$entry_guid}";
$entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid));
_debug("guid {$entry_guid} / {$entry_guid_hashed}", $debug_enabled);
$entry_timestamp = "";
$entry_timestamp = $item->get_date();
_debug("orig date: " . $item->get_date(), $debug_enabled);
if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
$entry_timestamp = time();
}
$entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
_debug("date {$entry_timestamp} [{$entry_timestamp_fmt}]", $debug_enabled);
// $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
// $entry_title = decode_numeric_entities($entry_title);
$entry_title = $item->get_title();
$entry_link = rewrite_relative_url($site_url, $item->get_link());
_debug("title {$entry_title}", $debug_enabled);
_debug("link {$entry_link}", $debug_enabled);
if (!$entry_title) {
$entry_title = date("Y-m-d H:i:s", $entry_timestamp);
}
$entry_content = $item->get_content();
if (!$entry_content) {
$entry_content = $item->get_description();
}
if ($_REQUEST["xdebug"] == 2) {
print "content: ";
print $entry_content;
print "\n";
}
$entry_comments = $item->get_comments_url();
$entry_author = $item->get_author();
$entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
$entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245));
$entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245));
$num_comments = (int) $item->get_comments_count();
_debug("author {$entry_author}", $debug_enabled);
_debug("num_comments: {$num_comments}", $debug_enabled);
_debug("looking for tags...", $debug_enabled);
// parse <category> entries into tags
$additional_tags = array();
$additional_tags_src = $item->get_categories();
if (is_array($additional_tags_src)) {
foreach ($additional_tags_src as $tobj) {
array_push($additional_tags, $tobj);
}
}
$entry_tags = array_unique($additional_tags);
for ($i = 0; $i < count($entry_tags); $i++) {
$entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
}
_debug("tags found: " . join(",", $entry_tags), $debug_enabled);
_debug("done collecting data.", $debug_enabled);
$result = db_query("SELECT id, content_hash, lang FROM ttrss_entries\n\t\t\t\t\tWHERE guid = '" . db_escape_string($entry_guid) . "' OR guid = '{$entry_guid_hashed}'");
if (db_num_rows($result) != 0) {
$base_entry_id = db_fetch_result($result, 0, "id");
$entry_stored_hash = db_fetch_result($result, 0, "content_hash");
$article_labels = get_article_labels($base_entry_id, $owner_uid);
$entry_language = db_fetch_result($result, 0, "lang");
} else {
$base_entry_id = false;
$entry_stored_hash = "";
$article_labels = array();
$entry_language = "";
}
$article = array("owner_uid" => $owner_uid, "guid" => $entry_guid, "guid_hashed" => $entry_guid_hashed, "title" => $entry_title, "content" => $entry_content, "link" => $entry_link, "labels" => $article_labels, "tags" => $entry_tags, "author" => $entry_author, "force_catchup" => false, "score_modifier" => 0, "language" => $entry_language, "feed" => array("id" => $feed, "fetch_url" => $fetch_url, "site_url" => $site_url));
$entry_plugin_data = "";
$entry_current_hash = calculate_article_hash($article, $pluginhost);
_debug("article hash: {$entry_current_hash} [stored={$entry_stored_hash}]", $debug_enabled);
if ($entry_current_hash == $entry_stored_hash && !isset($_REQUEST["force_rehash"])) {
_debug("stored article seems up to date [IID: {$base_entry_id}], updating timestamp only", $debug_enabled);
// 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("UPDATE ttrss_entries SET date_updated = NOW()\n\t\t\t\t\t\tWHERE id = '{$base_entry_id}'");
// if we allow duplicate posts, we have to continue to
// create the user entries for this feed
if (!get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
continue;
}
}
_debug("hash differs, applying plugin filters:", $debug_enabled);
foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) {
_debug("... " . get_class($plugin), $debug_enabled);
$start = microtime(true);
$article = $plugin->hook_article_filter($article);
_debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled);
$entry_plugin_data .= mb_strtolower(get_class($plugin)) . ",";
}
$entry_plugin_data = db_escape_string($entry_plugin_data);
_debug("plugin data: {$entry_plugin_data}", $debug_enabled);
// Workaround: 4-byte unicode requires utf8mb4 in MySQL. See https://tt-rss.org/forum/viewtopic.php?f=1&t=3377&p=20077#p20077
if (DB_TYPE == "mysql") {
foreach ($article as $k => $v) {
// i guess we'll have to take the risk of 4byte unicode labels & tags here
if (!is_array($article[$k])) {
$article[$k] = preg_replace('/[\\x{10000}-\\x{10FFFF}]/u', "�", $v);
}
}
}
$entry_tags = $article["tags"];
$entry_guid = db_escape_string($entry_guid);
$entry_title = db_escape_string($article["title"]);
$entry_author = db_escape_string($article["author"]);
$entry_link = db_escape_string($article["link"]);
$entry_content = $article["content"];
// escaped below
$entry_force_catchup = $article["force_catchup"];
$article_labels = $article["labels"];
$entry_score_modifier = (int) $article["score_modifier"];
$entry_language = db_escape_string($article["language"]);
if ($debug_enabled) {
_debug("article labels:", $debug_enabled);
print_r($article_labels);
}
_debug("force catchup: {$entry_force_catchup}");
if ($cache_images && is_writable(CACHE_DIR . '/images')) {
cache_images($entry_content, $site_url, $debug_enabled);
}
$entry_content = db_escape_string($entry_content, false);
db_query("BEGIN");
$result = db_query("SELECT id FROM\tttrss_entries\n\t\t\t\t\tWHERE (guid = '{$entry_guid}' OR guid = '{$entry_guid_hashed}')");
if (db_num_rows($result) == 0) {
_debug("base guid [{$entry_guid}] not found", $debug_enabled);
// base post entry does not exist, create it
$result = db_query("INSERT INTO ttrss_entries\n\t\t\t\t\t\t\t(title,\n\t\t\t\t\t\t\tguid,\n\t\t\t\t\t\t\tlink,\n\t\t\t\t\t\t\tupdated,\n\t\t\t\t\t\t\tcontent,\n\t\t\t\t\t\t\tcontent_hash,\n\t\t\t\t\t\t\tno_orig_date,\n\t\t\t\t\t\t\tdate_updated,\n\t\t\t\t\t\t\tdate_entered,\n\t\t\t\t\t\t\tcomments,\n\t\t\t\t\t\t\tnum_comments,\n\t\t\t\t\t\t\tplugin_data,\n\t\t\t\t\t\t\tlang,\n\t\t\t\t\t\t\tauthor)\n\t\t\t\t\t\tVALUES\n\t\t\t\t\t\t\t('{$entry_title}',\n\t\t\t\t\t\t\t'{$entry_guid_hashed}',\n\t\t\t\t\t\t\t'{$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'{$entry_current_hash}',\n\t\t\t\t\t\t\tfalse,\n\t\t\t\t\t\t\tNOW(),\n\t\t\t\t\t\t\t'{$date_feed_processed}',\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'{$entry_plugin_data}',\n\t\t\t\t\t\t\t'{$entry_language}',\n\t\t\t\t\t\t\t'{$entry_author}')");
} else {
$base_entry_id = db_fetch_result($result, 0, "id");
}
// now it should exist, if not - bad luck then
$result = db_query("SELECT id FROM ttrss_entries\n\t\t\t\t\tWHERE guid = '{$entry_guid}' OR guid = '{$entry_guid_hashed}'");
$entry_ref_id = 0;
$entry_int_id = 0;
if (db_num_rows($result) == 1) {
_debug("base guid found, checking for user record", $debug_enabled);
$ref_id = db_fetch_result($result, 0, "id");
$entry_ref_id = $ref_id;
/* $stored_guid = db_fetch_result($result, 0, "guid");
if ($stored_guid != $entry_guid_hashed) {
if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled);
db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE
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("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("article filters: ", $debug_enabled);
if (count($article_filters) != 0) {
print_r($article_filters);
}
}
if (find_article_filter($article_filters, "filter")) {
db_query("COMMIT");
// close transaction in progress
continue;
}
$score = calculate_article_score($article_filters) + $entry_score_modifier;
_debug("initial score: {$score} [including plugin modifier: {$entry_score_modifier}]", $debug_enabled);
$query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE\n\t\t\t\t\t\t\tref_id = '{$ref_id}' AND owner_uid = '{$owner_uid}'\n\t\t\t\t\t\t\t{$dupcheck_qpart}";
// if ($_REQUEST["xdebug"]) print "$query\n";
$result = db_query($query);
// okay it doesn't exist - create user entry
if (db_num_rows($result) == 0) {
_debug("user record not found, creating...", $debug_enabled);
if ($score >= -500 && !find_article_filter($article_filters, 'catchup') && !$entry_force_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')) {
$result = db_query("SELECT COUNT(*) AS similar FROM
ttrss_entries,ttrss_user_entries
WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD."
AND owner_uid = $owner_uid");
$ngram_similar = db_fetch_result($result, 0, "similar");
_debug("N-gram similar results: $ngram_similar", $debug_enabled);
if ($ngram_similar > 0) {
$unread = 'false';
}
} */
$last_marked = $marked == 'true' ? 'NOW()' : 'NULL';
$last_published = $published == 'true' ? 'NOW()' : 'NULL';
$result = db_query("INSERT INTO ttrss_user_entries\n\t\t\t\t\t\t\t\t(ref_id, owner_uid, feed_id, unread, last_read, marked,\n\t\t\t\t\t\t\t\tpublished, score, tag_cache, label_cache, uuid,\n\t\t\t\t\t\t\t\tlast_marked, last_published)\n\t\t\t\t\t\t\tVALUES ('{$ref_id}', '{$owner_uid}', '{$feed}', {$unread},\n\t\t\t\t\t\t\t\t{$last_read_qpart}, {$marked}, {$published}, '{$score}', '', '',\n\t\t\t\t\t\t\t\t'', {$last_marked}, {$last_published})");
if (PUBSUBHUBBUB_HUB && $published == 'true') {
$rss_link = get_self_url_prefix() . "/public.php?op=rss&id=-2&key=" . get_feed_access_key(-2, false, $owner_uid);
$p = new Publisher(PUBSUBHUBBUB_HUB);
/* $pubsub_result = */
$p->publish_update($rss_link);
}
$result = db_query("SELECT int_id FROM ttrss_user_entries WHERE\n\t\t\t\t\t\t\t\tref_id = '{$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 {
_debug("user record FOUND", $debug_enabled);
$entry_ref_id = db_fetch_result($result, 0, "ref_id");
$entry_int_id = db_fetch_result($result, 0, "int_id");
}
_debug("RID: {$entry_ref_id}, IID: {$entry_int_id}", $debug_enabled);
if (DB_TYPE == "pgsql") {
$tsvector_combined = db_escape_string(mb_substr($entry_title . ' ' . strip_tags($entry_content), 0, 1000000));
$tsvector_qpart = "tsvector_combined = to_tsvector('{$feed_language}', '{$tsvector_combined}'),";
} else {
$tsvector_qpart = "";
}
db_query("UPDATE ttrss_entries\n\t\t\t\t\t\tSET title = '{$entry_title}',\n\t\t\t\t\t\t\tcontent = '{$entry_content}',\n\t\t\t\t\t\t\tcontent_hash = '{$entry_current_hash}',\n\t\t\t\t\t\t\tupdated = '{$entry_timestamp_fmt}',\n\t\t\t\t\t\t\t{$tsvector_qpart}\n\t\t\t\t\t\t\tnum_comments = '{$num_comments}',\n\t\t\t\t\t\t\tplugin_data = '{$entry_plugin_data}',\n\t\t\t\t\t\t\tauthor = '{$entry_author}',\n\t\t\t\t\t\t\tlang = '{$entry_language}'\n\t\t\t\t\t\tWHERE id = '{$ref_id}'");
// update aux data
db_query("UPDATE ttrss_user_entries\n\t\t\t\t\t\t\tSET score = '{$score}' WHERE ref_id = '{$ref_id}'");
if ($mark_unread_on_update) {
db_query("UPDATE ttrss_user_entries\n\t\t\t\t\t\t\tSET last_read = null, unread = true WHERE ref_id = '{$ref_id}'");
}
}
db_query("COMMIT");
_debug("assigning labels [other]...", $debug_enabled);
foreach ($article_labels as $label) {
label_add_article($entry_ref_id, $label[1], $owner_uid);
}
_debug("assigning labels [filters]...", $debug_enabled);
assign_article_to_label_filters($entry_ref_id, $article_filters, $owner_uid, $article_labels);
_debug("looking for enclosures...", $debug_enabled);
// enclosures
$enclosures = array();
$encs = $item->get_enclosures();
if (is_array($encs)) {
foreach ($encs as $e) {
$e_item = array($e->link, $e->type, $e->length, $e->title, $e->width, $e->height);
array_push($enclosures, $e_item);
}
}
if ($debug_enabled) {
_debug("article enclosures:", $debug_enabled);
print_r($enclosures);
}
db_query("BEGIN");
// debugging
// db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'");
foreach ($enclosures as $enc) {
$enc_url = db_escape_string($enc[0]);
$enc_type = db_escape_string($enc[1]);
$enc_dur = db_escape_string($enc[2]);
$enc_title = db_escape_string($enc[3]);
$enc_width = intval($enc[4]);
$enc_height = intval($enc[5]);
$result = db_query("SELECT id FROM ttrss_enclosures\n\t\t\t\t\t\tWHERE content_url = '{$enc_url}' AND post_id = '{$entry_ref_id}'");
if (db_num_rows($result) == 0) {
db_query("INSERT INTO ttrss_enclosures\n\t\t\t\t\t\t\t(content_url, content_type, title, duration, post_id, width, height) VALUES\n\t\t\t\t\t\t\t('{$enc_url}', '{$enc_type}', '{$enc_title}', '{$enc_dur}', '{$entry_ref_id}', {$enc_width}, {$enc_height})");
}
}
db_query("COMMIT");
// check for manual tags (we have to do it here since they're loaded from filters)
foreach ($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('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("filtered article tags:", $debug_enabled);
print_r($filtered_tags);
}
// Save article tags in the database
if (count($filtered_tags) > 0) {
db_query("BEGIN");
foreach ($filtered_tags as $tag) {
$tag = sanitize_tag($tag);
$tag = db_escape_string($tag);
if (!tag_is_valid($tag)) {
continue;
}
$result = db_query("SELECT id FROM ttrss_tags\n\t\t\t\t\t\t\tWHERE tag_name = '{$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("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(",", $tags_to_cache));
db_query("UPDATE ttrss_user_entries\n\t\t\t\t\t\tSET tag_cache = '{$tags_str}' WHERE ref_id = '{$entry_ref_id}'\n\t\t\t\t\t\tAND owner_uid = {$owner_uid}");
db_query("COMMIT");
}
_debug("article processed", $debug_enabled);
}
_debug("purging feed...", $debug_enabled);
purge_feed($feed, 0, $debug_enabled);
db_query("UPDATE ttrss_feeds\n\t\t\t\tSET last_updated = NOW(), last_error = '' WHERE id = '{$feed}'");
// db_query("COMMIT");
} else {
$error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
_debug("fetch error: {$error_msg}", $debug_enabled);
if (count($rss->errors()) > 1) {
foreach ($rss->errors() as $error) {
_debug("+ {$error}");
}
}
db_query("UPDATE ttrss_feeds SET last_error = '{$error_msg}',\n\t\t\t\tlast_updated = NOW() WHERE id = '{$feed}'");
unset($rss);
}
_debug("done", $debug_enabled);
return $rss;
}
function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false)
{
$debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
_debug("start", $debug_enabled);
$result = db_query("SELECT id,update_interval,auth_login,\n\t\t\tfeed_url,auth_pass,cache_images,last_updated,\n\t\t\tmark_unread_on_update, owner_uid,\n\t\t\tpubsub_state, auth_pass_encrypted,\n\t\t\t(SELECT max(date_entered) FROM\n\t\t\t\tttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '{$feed}') AS last_article_timestamp\n\t\t\tFROM ttrss_feeds WHERE id = '{$feed}'");
if (db_num_rows($result) == 0) {
_debug("feed {$feed} NOT FOUND/SKIPPED", $debug_enabled);
return false;
}
$last_updated = db_fetch_result($result, 0, "last_updated");
$last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp"));
if (defined('_DISABLE_HTTP_304')) {
$last_article_timestamp = 0;
}
$owner_uid = db_fetch_result($result, 0, "owner_uid");
$mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, 0, "mark_unread_on_update"));
$pubsub_state = db_fetch_result($result, 0, "pubsub_state");
$auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, 0, "auth_pass_encrypted"));
db_query("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 ($auth_pass_encrypted) {
require_once "crypt.php";
$auth_pass = decrypt_string($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);
$date_feed_processed = date('Y-m-d H:i');
$cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml";
$pluginhost = new PluginHost();
$pluginhost->set_debug($debug_enabled);
$user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
$pluginhost->load(PLUGINS, PluginHost::KIND_ALL);
$pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid);
$pluginhost->load_data();
$rss = false;
$rss_hash = false;
$force_refetch = isset($_REQUEST["force_refetch"]);
if (file_exists($cache_filename) && is_readable($cache_filename) && !$auth_login && !$auth_pass && filemtime($cache_filename) > time() - 30) {
_debug("using local cache.", $debug_enabled);
@($feed_data = file_get_contents($cache_filename));
if ($feed_data) {
$rss_hash = sha1($feed_data);
}
} else {
_debug("local cache will not be used for this feed", $debug_enabled);
}
if (!$rss) {
foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) {
$feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed);
}
if (!$feed_data) {
_debug("fetching [{$fetch_url}]...", $debug_enabled);
_debug("If-Modified-Since: " . gmdate('D, d M Y H:i:s \\G\\M\\T', $last_article_timestamp), $debug_enabled);
$feed_data = fetch_file_contents($fetch_url, false, $auth_login, $auth_pass, false, $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT, $force_refetch ? 0 : $last_article_timestamp);
global $fetch_curl_used;
if (!$fetch_curl_used) {
$tmp = @gzdecode($feed_data);
if ($tmp) {
$feed_data = $tmp;
}
}
$feed_data = trim($feed_data);
_debug("fetch done.", $debug_enabled);
/* if ($feed_data) {
$error = verify_feed_xml($feed_data);
if ($error) {
_debug("error verifying XML, code: " . $error->code, $debug_enabled);
if ($error->code == 26) {
_debug("got error 26, trying to decode entities...", $debug_enabled);
$feed_data = html_entity_decode($feed_data, ENT_COMPAT, 'UTF-8');
$error = verify_feed_xml($feed_data);
if ($error) $feed_data = '';
}
}
} */
}
if (!$feed_data) {
global $fetch_last_error;
global $fetch_last_error_code;
_debug("unable to fetch: {$fetch_last_error} [{$fetch_last_error_code}]", $debug_enabled);
$error_escaped = '';
// If-Modified-Since
if ($fetch_last_error_code != 304) {
$error_escaped = db_escape_string($fetch_last_error);
} else {
_debug("source claims data not modified, nothing to do.", $debug_enabled);
}
db_query("UPDATE ttrss_feeds SET last_error = '{$error_escaped}',\n\t\t\t\t\t\tlast_updated = NOW() WHERE id = '{$feed}'");
return;
}
}
foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) {
$feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed);
}
// set last update to now so if anything *simplepie* crashes later we won't be
// continuously failing on the same feed
//db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'");
if (!$rss) {
$rss = new FeedParser($feed_data);
$rss->init();
}
// print_r($rss);
$feed = db_escape_string($feed);
if (!$rss->error()) {
// cache data for later
if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) {
$new_rss_hash = sha1($rss_data);
if ($new_rss_hash != $rss_hash && count($rss->get_items()) > 0) {
_debug("saving {$cache_filename}", $debug_enabled);
@file_put_contents($cache_filename, $feed_data);
}
}
// We use local pluginhost here because we need to load different per-user feed plugins
$pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss);
_debug("processing feed data...", $debug_enabled);
// db_query("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)";
}
$result = db_query("SELECT title,site_url,owner_uid,favicon_avg_color,\n\t\t\t\t(favicon_last_checked IS NULL OR {$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");
$orig_site_url = db_fetch_result($result, 0, "site_url");
$favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, "favicon_needs_check"));
$favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color");
$owner_uid = db_fetch_result($result, 0, "owner_uid");
$site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
_debug("site_url: {$site_url}", $debug_enabled);
_debug("feed_title: " . $rss->get_title(), $debug_enabled);
if ($favicon_needs_check || $force_refetch) {
/* terrible hack: if we crash on floicon shit here, we won't check
* the icon avgcolor again (unless the icon got updated) */
$favicon_file = ICONS_DIR . "/{$feed}.ico";
$favicon_modified = @filemtime($favicon_file);
_debug("checking favicon...", $debug_enabled);
check_feed_favicon($site_url, $feed);
$favicon_modified_new = @filemtime($favicon_file);
if ($favicon_modified_new > $favicon_modified) {
$favicon_avg_color = '';
}
if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') {
require_once "colors.php";
db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE\n\t\t\t\t\t\t\tid = '{$feed}'");
$favicon_color = db_escape_string(calculate_avg_color($favicon_file));
$favicon_colorstring = ",favicon_avg_color = '" . $favicon_color . "'";
} else {
if ($favicon_avg_color == 'fail') {
_debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled);
}
}
db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW()\n\t\t\t\t\t{$favicon_colorstring}\n\t\t\t\t\tWHERE id = '{$feed}'");
}
if (!$registered_title || $registered_title == "[Unknown]") {
$feed_title = db_escape_string($rss->get_title());
if ($feed_title) {
_debug("registering title: {$feed_title}", $debug_enabled);
db_query("UPDATE ttrss_feeds SET\n\t\t\t\t\t\ttitle = '{$feed_title}' WHERE id = '{$feed}'");
}
}
if ($site_url && $orig_site_url != $site_url) {
db_query("UPDATE ttrss_feeds SET\n\t\t\t\t\tsite_url = '{$site_url}' WHERE id = '{$feed}'");
}
_debug("loading filters & labels...", $debug_enabled);
$filters = load_filters($feed, $owner_uid);
$labels = get_all_labels($owner_uid);
_debug("" . count($filters) . " filters loaded.", $debug_enabled);
$items = $rss->get_items();
if (!is_array($items)) {
_debug("no articles found.", $debug_enabled);
db_query("UPDATE ttrss_feeds\n\t\t\t\t\tSET last_updated = NOW(), last_error = '' WHERE id = '{$feed}'");
return;
// no articles
}
if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) {
_debug("checking for PUSH hub...", $debug_enabled);
$feed_hub_url = false;
$links = $rss->get_links('hub');
if ($links && is_array($links)) {
foreach ($links as $l) {
$feed_hub_url = $l;
break;
}
}
_debug("feed hub url: {$feed_hub_url}", $debug_enabled);
if ($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);
_debug("feed hub url found, subscribe request sent.", $debug_enabled);
db_query("UPDATE ttrss_feeds SET pubsub_state = 1\n\t\t\t\t\t\tWHERE id = '{$feed}'");
}
}
_debug("processing articles...", $debug_enabled);
foreach ($items as $item) {
if ($_REQUEST['xdebug'] == 3) {
print_r($item);
}
$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());
}
_debug("f_guid {$entry_guid}", $debug_enabled);
if (!$entry_guid) {
continue;
}
$entry_guid = "{$owner_uid},{$entry_guid}";
$entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid));
_debug("guid {$entry_guid} / {$entry_guid_hashed}", $debug_enabled);
$entry_timestamp = "";
$entry_timestamp = $item->get_date();
_debug("orig date: " . $item->get_date(), $debug_enabled);
if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
$entry_timestamp = time();
$no_orig_date = 'true';
} else {
$no_orig_date = 'false';
}
$entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
_debug("date {$entry_timestamp} [{$entry_timestamp_fmt}]", $debug_enabled);
// $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
// $entry_title = decode_numeric_entities($entry_title);
$entry_title = $item->get_title();
$entry_link = rewrite_relative_url($site_url, $item->get_link());
_debug("title {$entry_title}", $debug_enabled);
_debug("link {$entry_link}", $debug_enabled);
if (!$entry_title) {
$entry_title = date("Y-m-d H:i:s", $entry_timestamp);
}
$entry_content = $item->get_content();
if (!$entry_content) {
$entry_content = $item->get_description();
}
if ($_REQUEST["xdebug"] == 2) {
print "content: ";
print $entry_content;
print "\n";
}
$entry_comments = $item->get_comments_url();
$entry_author = $item->get_author();
$entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
$entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245));
$entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245));
$num_comments = (int) $item->get_comments_count();
_debug("author {$entry_author}", $debug_enabled);
_debug("num_comments: {$num_comments}", $debug_enabled);
_debug("looking for tags...", $debug_enabled);
// parse <category> entries into tags
$additional_tags = array();
$additional_tags_src = $item->get_categories();
if (is_array($additional_tags_src)) {
foreach ($additional_tags_src as $tobj) {
array_push($additional_tags, $tobj);
}
}
$entry_tags = array_unique($additional_tags);
for ($i = 0; $i < count($entry_tags); $i++) {
$entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
}
_debug("tags found: " . join(",", $entry_tags), $debug_enabled);
_debug("done collecting data.", $debug_enabled);
// TODO: less memory-hungry implementation
_debug("applying plugin filters..", $debug_enabled);
// FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?)
$result = db_query("SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries\n\t\t\t\t\tWHERE ref_id = id AND (guid = '" . db_escape_string($entry_guid) . "' OR guid = '{$entry_guid_hashed}') AND owner_uid = {$owner_uid}");
if (db_num_rows($result) != 0) {
$entry_plugin_data = db_fetch_result($result, 0, "plugin_data");
$stored_article = array("title" => db_fetch_result($result, 0, "title"), "content" => db_fetch_result($result, 0, "content"), "link" => db_fetch_result($result, 0, "link"), "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")), "author" => db_fetch_result($result, 0, "author"));
} else {
$entry_plugin_data = "";
$stored_article = array();
}
$article = array("owner_uid" => $owner_uid, "guid" => $entry_guid, "title" => $entry_title, "content" => $entry_content, "link" => $entry_link, "tags" => $entry_tags, "plugin_data" => $entry_plugin_data, "author" => $entry_author, "stored" => $stored_article);
foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) {
$article = $plugin->hook_article_filter($article);
}
$entry_tags = $article["tags"];
$entry_guid = db_escape_string($entry_guid);
$entry_title = db_escape_string($article["title"]);
$entry_author = db_escape_string($article["author"]);
$entry_link = db_escape_string($article["link"]);
$entry_plugin_data = db_escape_string($article["plugin_data"]);
$entry_content = $article["content"];
// escaped below
_debug("plugin data: {$entry_plugin_data}", $debug_enabled);
if ($cache_images && is_writable(CACHE_DIR . '/images')) {
cache_images($entry_content, $site_url, $debug_enabled);
}
$entry_content = db_escape_string($entry_content, false);
$content_hash = "SHA1:" . sha1($entry_content);
db_query("BEGIN");
$result = db_query("SELECT id FROM\tttrss_entries\n\t\t\t\t\tWHERE (guid = '{$entry_guid}' OR guid = '{$entry_guid_hashed}')");
if (db_num_rows($result) == 0) {
_debug("base guid [{$entry_guid}] not found", $debug_enabled);
// base post entry does not exist, create it
$result = db_query("INSERT INTO ttrss_entries\n\t\t\t\t\t\t\t(title,\n\t\t\t\t\t\t\tguid,\n\t\t\t\t\t\t\tlink,\n\t\t\t\t\t\t\tupdated,\n\t\t\t\t\t\t\tcontent,\n\t\t\t\t\t\t\tcontent_hash,\n\t\t\t\t\t\t\tno_orig_date,\n\t\t\t\t\t\t\tdate_updated,\n\t\t\t\t\t\t\tdate_entered,\n\t\t\t\t\t\t\tcomments,\n\t\t\t\t\t\t\tnum_comments,\n\t\t\t\t\t\t\tplugin_data,\n\t\t\t\t\t\t\tauthor)\n\t\t\t\t\t\tVALUES\n\t\t\t\t\t\t\t('{$entry_title}',\n\t\t\t\t\t\t\t'{$entry_guid_hashed}',\n\t\t\t\t\t\t\t'{$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\t'{$date_feed_processed}',\n\t\t\t\t\t\t\t'{$entry_comments}',\n\t\t\t\t\t\t\t'{$num_comments}',\n\t\t\t\t\t\t\t'{$entry_plugin_data}',\n\t\t\t\t\t\t\t'{$entry_author}')");
$article_labels = array();
} 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("UPDATE ttrss_entries SET date_updated = NOW()\n\t\t\t\t\t\tWHERE id = '{$base_entry_id}'");
$article_labels = get_article_labels($base_entry_id, $owner_uid);
}
// now it should exist, if not - bad luck then
$result = db_query("SELECT\n\t\t\t\t\t\tid,content_hash,no_orig_date,title,plugin_data,guid,\n\t\t\t\t\t\t" . SUBSTRING_FOR_DATE . "(date_updated,1,19) as date_updated,\n\t\t\t\t\t\t" . SUBSTRING_FOR_DATE . "(updated,1,19) as updated,\n\t\t\t\t\t\tnum_comments\n\t\t\t\t\tFROM\n\t\t\t\t\t\tttrss_entries\n\t\t\t\t\tWHERE guid = '{$entry_guid}' OR guid = '{$entry_guid_hashed}'");
$entry_ref_id = 0;
$entry_int_id = 0;
if (db_num_rows($result) == 1) {
_debug("base guid found, checking for user record", $debug_enabled);
// 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"));
$orig_plugin_data = db_fetch_result($result, 0, "plugin_data");
$ref_id = db_fetch_result($result, 0, "id");
$entry_ref_id = $ref_id;
/* $stored_guid = db_fetch_result($result, 0, "guid");
if ($stored_guid != $entry_guid_hashed) {
if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled);
db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE
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("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("article filters: ", $debug_enabled);
if (count($article_filters) != 0) {
print_r($article_filters);
}
}
if (find_article_filter($article_filters, "filter")) {
db_query("COMMIT");
// close transaction in progress
continue;
}
$score = calculate_article_score($article_filters);
_debug("initial score: {$score}", $debug_enabled);
$query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE\n\t\t\t\t\t\t\tref_id = '{$ref_id}' AND owner_uid = '{$owner_uid}'\n\t\t\t\t\t\t\t{$dupcheck_qpart}";
// if ($_REQUEST["xdebug"]) print "$query\n";
$result = db_query($query);
// okay it doesn't exist - create user entry
if (db_num_rows($result) == 0) {
_debug("user record not found, creating...", $debug_enabled);
if ($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')) {
$result = db_query("SELECT COUNT(*) AS similar FROM\n\t\t\t\t\t\t\t\t\tttrss_entries,ttrss_user_entries\n\t\t\t\t\t\t\t\tWHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'\n\t\t\t\t\t\t\t\t\tAND similarity(title, '{$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");
_debug("N-gram similar results: {$ngram_similar}", $debug_enabled);
if ($ngram_similar > 0) {
$unread = 'false';
}
}
$last_marked = $marked == 'true' ? 'NOW()' : 'NULL';
$last_published = $published == 'true' ? 'NOW()' : 'NULL';
$result = db_query("INSERT INTO ttrss_user_entries\n\t\t\t\t\t\t\t\t(ref_id, owner_uid, feed_id, unread, last_read, marked,\n\t\t\t\t\t\t\t\tpublished, score, tag_cache, label_cache, uuid,\n\t\t\t\t\t\t\t\tlast_marked, last_published)\n\t\t\t\t\t\t\tVALUES ('{$ref_id}', '{$owner_uid}', '{$feed}', {$unread},\n\t\t\t\t\t\t\t\t{$last_read_qpart}, {$marked}, {$published}, '{$score}', '', '',\n\t\t\t\t\t\t\t\t'', {$last_marked}, {$last_published})");
if (PUBSUBHUBBUB_HUB && $published == 'true') {
$rss_link = get_self_url_prefix() . "/public.php?op=rss&id=-2&key=" . get_feed_access_key(-2, false, $owner_uid);
$p = new Publisher(PUBSUBHUBBUB_HUB);
$pubsub_result = $p->publish_update($rss_link);
}
$result = db_query("SELECT int_id FROM ttrss_user_entries WHERE\n\t\t\t\t\t\t\t\tref_id = '{$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 {
_debug("user record FOUND", $debug_enabled);
$entry_ref_id = db_fetch_result($result, 0, "ref_id");
$entry_int_id = db_fetch_result($result, 0, "int_id");
}
_debug("RID: {$entry_ref_id}, IID: {$entry_int_id}", $debug_enabled);
$post_needs_update = false;
$update_insignificant = false;
if ($orig_num_comments != $num_comments) {
$post_needs_update = true;
$update_insignificant = true;
}
if ($entry_plugin_data != $orig_plugin_data) {
$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("post {$entry_guid_hashed} needs update...", $debug_enabled);
}
// print "<!-- post $orig_title needs update : $post_needs_update -->";
db_query("UPDATE ttrss_entries\n\t\t\t\t\t\t\tSET title = '{$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\t\tplugin_data = '{$entry_plugin_data}'\n\t\t\t\t\t\t\tWHERE id = '{$ref_id}'");
if (!$update_insignificant) {
if ($mark_unread_on_update) {
db_query("UPDATE ttrss_user_entries\n\t\t\t\t\t\t\t\t\tSET last_read = null, unread = true WHERE ref_id = '{$ref_id}'");
}
}
}
}
db_query("COMMIT");
_debug("assigning labels...", $debug_enabled);
assign_article_to_label_filters($entry_ref_id, $article_filters, $owner_uid, $article_labels);
_debug("looking for enclosures...", $debug_enabled);
// enclosures
$enclosures = array();
$encs = $item->get_enclosures();
if (is_array($encs)) {
foreach ($encs as $e) {
$e_item = array($e->link, $e->type, $e->length);
array_push($enclosures, $e_item);
}
}
if ($debug_enabled) {
_debug("article enclosures:", $debug_enabled);
print_r($enclosures);
}
db_query("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]);
$result = db_query("SELECT id FROM ttrss_enclosures\n\t\t\t\t\t\tWHERE content_url = '{$enc_url}' AND post_id = '{$entry_ref_id}'");
if (db_num_rows($result) == 0) {
db_query("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("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('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("filtered article tags:", $debug_enabled);
print_r($filtered_tags);
}
// Save article tags in the database
if (count($filtered_tags) > 0) {
db_query("BEGIN");
foreach ($filtered_tags as $tag) {
$tag = sanitize_tag($tag);
$tag = db_escape_string($tag);
if (!tag_is_valid($tag)) {
continue;
}
$result = db_query("SELECT id FROM ttrss_tags\n\t\t\t\t\t\t\tWHERE tag_name = '{$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("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(",", $tags_to_cache));
db_query("UPDATE ttrss_user_entries\n\t\t\t\t\t\tSET tag_cache = '{$tags_str}' WHERE ref_id = '{$entry_ref_id}'\n\t\t\t\t\t\tAND owner_uid = {$owner_uid}");
db_query("COMMIT");
}
if (get_pref("AUTO_ASSIGN_LABELS", $owner_uid, false)) {
_debug("auto-assigning labels...", $debug_enabled);
foreach ($labels as $label) {
$caption = preg_quote($label["caption"]);
if ($caption && preg_match("/\\b{$caption}\\b/i", "{$tags_str} " . strip_tags($entry_content) . " {$entry_title}")) {
if (!labels_contains_caption($article_labels, $caption)) {
label_add_article($entry_ref_id, $caption, $owner_uid);
}
}
}
}
_debug("article processed", $debug_enabled);
}
_debug("purging feed...", $debug_enabled);
purge_feed($feed, 0, $debug_enabled);
db_query("UPDATE ttrss_feeds\n\t\t\t\tSET last_updated = NOW(), last_error = '' WHERE id = '{$feed}'");
// db_query("COMMIT");
} else {
$error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
_debug("error fetching feed: {$error_msg}", $debug_enabled);
db_query("UPDATE ttrss_feeds SET last_error = '{$error_msg}',\n\t\t\t\t\tlast_updated = NOW() WHERE id = '{$feed}'");
}
unset($rss);
_debug("done", $debug_enabled);
}
<?php
/*
*
* Basic Blogger Data Dump
* This file will download your blogger feed and save to log file
*
*/
include 'class.feedparser.php';
$feed = new FeedParser();
$limit = 1;
$user = '******';
$data = $feed->blogger($user, $limit);
$fh = fopen('blogger.log', 'w');
fwrite($fh, print_r($data, 1));
fclose($fh);
echo "Downloaded " . count($data['posts']) . " posts\n";
echo "Bandwidth Usage: {$feed->bandwidth}\n";
echo "Requests: {$feed->requests}\n";
echo "\n";
