$count = 0;
while ($line = db_fetch_assoc($result)) {
- $subscribers = db_escape_string($line["subscribers"]);
- $feed_url = db_escape_string($line["feed_url"]);
- $title = db_escape_string($line["title"]);
- $site_url = db_escape_string($line["site_url"]);
+ $subscribers = db_escape_string($link, $line["subscribers"]);
+ $feed_url = db_escape_string($link, $line["feed_url"]);
+ $title = db_escape_string($link, $line["title"]);
+ $site_url = db_escape_string($link, $line["site_url"]);
$tmp_result = db_query($link, "SELECT subscribers FROM
ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'");
}
$result = db_query($link, "SELECT id,update_interval,auth_login,
- feed_url,auth_pass,cache_images,last_updated,cache_content,
- mark_unread_on_update, owner_uid, update_on_checksum_change,
+ feed_url,auth_pass,cache_images,last_updated,
+ mark_unread_on_update, owner_uid,
pubsub_state
FROM ttrss_feeds WHERE id = '$feed'");
$owner_uid = db_fetch_result($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()
$auth_pass = db_fetch_result($result, 0, "auth_pass");
$cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
- $cache_content = sql_bool_to_bool(db_fetch_result($result, 0, "cache_content"));
$fetch_url = db_fetch_result($result, 0, "feed_url");
- $feed = db_escape_string($feed);
+ $feed = db_escape_string($link, $feed);
- if ($auth_login && $auth_pass ){
+ /* if ($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;
$simplepie_cache_dir = CACHE_DIR . "/simplepie";
+ $date_feed_processed = date('Y-m-d H:i');
+
if (!is_dir($simplepie_cache_dir)) {
mkdir($simplepie_cache_dir);
}
+ $feed_data = fetch_file_contents($fetch_url, false,
+ $auth_login, $auth_pass, false, $no_cache ? 15 : 45);
+
+ if (!$feed_data) {
+ global $fetch_last_error;
+
+ if ($debug_enabled) {
+ _debug("update_rss_feed: unable to fetch: $fetch_last_error");
+ }
+
+ $error_escaped = db_escape_string($link, $fetch_last_error);
+
+ db_query($link,
+ "UPDATE ttrss_feeds SET last_error = '$error_escaped',
+ last_updated = NOW() WHERE id = '$feed'");
+
+ return;
+ }
+
+ $pluginhost = new PluginHost($link);
+ $pluginhost->set_debug($debug_enabled);
+ $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid);
+
+ $pluginhost->load(PLUGINS, $pluginhost::KIND_ALL);
+ $pluginhost->load($user_plugins, $pluginhost::KIND_USER, $owner_uid);
+ $pluginhost->load_data();
+
+ foreach ($pluginhost->get_hooks($pluginhost::HOOK_FEED_FETCHED) as $plugin) {
+ $feed_data = $plugin->hook_feed_fetched($feed_data);
+ }
+
+ if ($debug_enabled) {
+ _debug("update_rss_feed: fetch done, parsing...");
+ }
+
$rss = new SimplePie();
- $rss->set_useragent(SELF_USER_AGENT);
- $rss->set_timeout($no_cache ? 15 : 60);
- $rss->set_feed_url($fetch_url);
+ $rss->set_sanitize_class("SanitizeDummy");
+ // simplepie ignores the above and creates default sanitizer anyway,
+ // so let's override it...
+ $rss->sanitize = new SanitizeDummy();
$rss->set_output_encoding('UTF-8');
- //$rss->force_feed(true);
+ $rss->set_raw_data($feed_data);
if ($debug_enabled) {
_debug("feed update interval (sec): " .
$rss->set_cache_duration($cache_age);
}
- $rss->init();
+ @$rss->init();
// print_r($rss);
- if ($debug_enabled) {
- _debug("update_rss_feed: fetch done, parsing...");
- }
-
- $feed = db_escape_string($feed);
+ $feed = db_escape_string($link, $feed);
if (!$rss->error()) {
- global $pluginhost;
+ // 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", $rss);
if ($debug_enabled) {
$favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
}
- $result = db_query($link, "SELECT title,icon_url,site_url,owner_uid,
+ $result = db_query($link, "SELECT title,site_url,owner_uid,
(favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
favicon_needs_check
FROM ttrss_feeds WHERE id = '$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");
- $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 250));
+ $site_url = db_escape_string($link, mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
if ($debug_enabled) {
_debug("update_rss_feed: checking favicon...");
if (!$registered_title || $registered_title == "[Unknown]") {
- $feed_title = db_escape_string($rss->get_title());
+ $feed_title = db_escape_string($link, $rss->get_title());
if ($debug_enabled) {
_debug("update_rss_feed: registering title: $feed_title");
site_url = '$site_url' WHERE id = '$feed'");
}
- $icon_url = db_escape_string(mb_substr(
- rewrite_relative_url($fetch_url, $rss->get_image_url()), 0, 250));
-
- if ($icon_url && $orig_icon_url != $icon_url) {
- db_query($link, "UPDATE ttrss_feeds SET icon_url = '$icon_url' WHERE id = '$feed'");
- }
-
if ($debug_enabled) {
_debug("update_rss_feed: loading filters & labels...");
}
}
foreach ($items as $item) {
- if ($_REQUEST['xdebug'] == 2) {
+ if ($_REQUEST['xdebug'] == 3) {
print_r($item);
}
if (!$entry_guid) $entry_guid = $item->get_link();
if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
- if ($cache_content) {
- $entry_guid = "ccache:$entry_guid";
- }
-
- if ($auth_login || $auth_pass) {
- $entry_guid = "auth,$owner_uid:$entry_guid";
- }
-
if ($debug_enabled) {
_debug("update_rss_feed: guid $entry_guid");
}
if (!$entry_guid) continue;
+ $entry_guid = "$owner_uid,$entry_guid";
+
$entry_timestamp = "";
$entry_timestamp = strtotime($item->get_date());
- if ($entry_timestamp == -1 || !$entry_timestamp) {
+ if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
$entry_timestamp = time();
$no_orig_date = 'true';
} else {
$entry_title = $item->get_title();
- $entry_link = rewrite_relative_url($site_url, htmlspecialchars_decode($item->get_link()));
+ $entry_link = rewrite_relative_url($site_url, $item->get_link());
if ($debug_enabled) {
_debug("update_rss_feed: title $entry_title");
$entry_content = $item->get_content();
if (!$entry_content) $entry_content = $item->get_description();
- if ($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_cached_content = "";
-
$entry_comments = $item->data["comments"];
if ($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);
+ $entry_author = db_escape_string($link, $entry_author);
}
- $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 250));
-
- $result = db_query($link, "SELECT id FROM ttrss_entries
- WHERE guid = '$entry_guid'");
+ $entry_guid = db_escape_string($link, mb_substr($entry_guid, 0, 245));
- $entry_comments = db_escape_string(mb_substr($entry_comments, 0, 250));
- $entry_author = db_escape_string(mb_substr($entry_author, 0, 250));
+ $entry_comments = db_escape_string($link, mb_substr($entry_comments, 0, 245));
+ $entry_author = db_escape_string($link, mb_substr($entry_author, 0, 245));
$num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments');
}
if ($debug_enabled) {
- _debug("update_rss_feed: done collecting data [TITLE:$entry_title]");
+ _debug("update_rss_feed: done collecting data.");
}
// TODO: less memory-hungry implementation
_debug("update_rss_feed: applying plugin filters..");
}
- $article = array("owner_uid" => $owner_uid,
+ // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?)
+ $result = db_query($link, "SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries
+ WHERE ref_id = id AND guid = '".db_escape_string($link, $entry_guid)."' 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, // read only
+ "guid" => $entry_guid, // read only
"title" => $entry_title,
"content" => $entry_content,
"link" => $entry_link,
"tags" => $entry_tags,
- "author" => $entry_author);
+ "plugin_data" => $entry_plugin_data,
+ "author" => $entry_author,
+ "stored" => $stored_article);
- global $pluginhost;
foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) {
$article = $plugin->hook_article_filter($article);
}
$entry_tags = $article["tags"];
- $entry_content = db_escape_string($article["content"], false);
- $entry_title = db_escape_string($article["title"]);
- $entry_author = db_escape_string($article["author"]);
- $entry_link = db_escape_string($article["link"]);
+ $entry_guid = db_escape_string($link, $entry_guid);
+ $entry_title = db_escape_string($link, $article["title"]);
+ $entry_author = db_escape_string($link, $article["author"]);
+ $entry_link = db_escape_string($link, $article["link"]);
+ $entry_plugin_data = db_escape_string($link, $article["plugin_data"]);
+ $entry_content = $article["content"]; // escaped below
- $content_hash = "SHA1:" . sha1(strip_tags($entry_content));
- db_query($link, "BEGIN");
+ if ($debug_enabled) {
+ _debug("update_rss_feed: plugin data: $entry_plugin_data");
+ }
- if (db_num_rows($result) == 0) {
+ if ($cache_images && is_writable(CACHE_DIR . '/images'))
+ cache_images($entry_content, $site_url, $debug_enabled);
- if ($debug_enabled) {
- _debug("update_rss_feed: base guid not found");
- }
+ $entry_content = db_escape_string($link, $entry_content, false);
- if ($cache_content) {
- if ($debug_enabled) {
- _debug("update_rss_feed: caching content (initial)...");
- }
+ $content_hash = "SHA1:" . sha1($entry_content);
- $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass);
+ db_query($link, "BEGIN");
- if ($cache_images && is_writable(CACHE_DIR . '/images'))
- $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled);
+ $result = db_query($link, "SELECT id FROM ttrss_entries
+ WHERE guid = '$entry_guid'");
+
+ if (db_num_rows($result) == 0) {
- $entry_cached_content = db_escape_string($entry_cached_content, false);
+ if ($debug_enabled) {
+ _debug("update_rss_feed: base guid [$entry_guid] not found");
}
// base post entry does not exist, create it
date_entered,
comments,
num_comments,
+ plugin_data,
author)
VALUES
('$entry_title',
'$entry_timestamp_fmt',
'$entry_content',
'$content_hash',
- '$entry_cached_content',
+ '',
$no_orig_date,
NOW(),
- NOW(),
+ '$date_feed_processed',
'$entry_comments',
'$num_comments',
+ '$entry_plugin_data',
'$entry_author')");
$article_labels = array();
// now it should exist, if not - bad luck then
$result = db_query($link, "SELECT
- id,content_hash,no_orig_date,title,
+ id,content_hash,no_orig_date,title,plugin_data,
".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated,
".SUBSTRING_FOR_DATE."(updated,1,19) as updated,
- num_comments, cached_content
+ num_comments
FROM
ttrss_entries
WHERE guid = '$entry_guid'");
if (db_num_rows($result) == 1) {
if ($debug_enabled) {
- _debug("update_rss_feed: base guid found, checking for user record");
+ _debug("update_rss_feed: base guid [$entry_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");
- $orig_cached_content = trim(db_fetch_result($result, 0, "cached_content"));
$orig_date_updated = strtotime(db_fetch_result($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;
}
}
+ $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL';
+ $last_published = ($published == 'true') ? 'NOW()' : 'NULL';
+
$result = db_query($link,
"INSERT INTO ttrss_user_entries
(ref_id, owner_uid, feed_id, unread, last_read, marked,
- published, score, tag_cache, label_cache, uuid)
+ published, score, tag_cache, label_cache, uuid,
+ last_marked, last_published)
VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
- $last_read_qpart, $marked, $published, '$score', '', '', '')");
+ $last_read_qpart, $marked, $published, '$score', '', '',
+ '', $last_marked, $last_published)");
if (PUBSUBHUBBUB_HUB && $published == 'true') {
$rss_link = get_self_url_prefix() .
$post_needs_update = false;
$update_insignificant = false;
- $cached_content_needs_update = false;
if ($orig_num_comments != $num_comments) {
$post_needs_update = true;
$update_insignificant = true;
}
- if ($content_hash != $orig_content_hash) {
+ if ($entry_plugin_data != $orig_plugin_data) {
$post_needs_update = true;
- $update_insignificant = false;
- $cached_content_needs_update = true;
+ $update_insignificant = true;
}
- if ($cache_content) {
- if ($debug_enabled) {
- _debug("update_rss_feed: caching content because original checksum changed...");
- }
-
- $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass);
-
- if ($entry_cached_content) {
- if ($cache_images && is_writable(CACHE_DIR . '/images'))
- $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled);
-
- $entry_cached_content = db_escape_string($entry_cached_content, false);
- $post_needs_update = true;
- } else {
- $entry_cached_content = db_escape_string($orig_cached_content);
- }
- } else {
- $entry_cached_content = db_escape_string($orig_cached_content);
+ if ($content_hash != $orig_content_hash) {
+ $post_needs_update = true;
+ $update_insignificant = false;
}
- if (db_escape_string($orig_title) != $entry_title) {
+ if (db_escape_string($link, $orig_title) != $entry_title) {
$post_needs_update = true;
$update_insignificant = false;
}
db_query($link, "UPDATE ttrss_entries
SET title = '$entry_title', content = '$entry_content',
content_hash = '$content_hash',
- cached_content = '$entry_cached_content',
updated = '$entry_timestamp_fmt',
- num_comments = '$num_comments'
+ num_comments = '$num_comments',
+ plugin_data = '$entry_plugin_data'
WHERE id = '$ref_id'");
if (!$update_insignificant) {
if ($mark_unread_on_update) {
db_query($link, "UPDATE ttrss_user_entries
SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
- } else if ($update_on_checksum_change) {
- db_query($link, "UPDATE ttrss_user_entries
- SET last_read = null WHERE ref_id = '$ref_id'
- AND unread = false");
}
}
}
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]);
+ $enc_url = db_escape_string($link, $enc[0]);
+ $enc_type = db_escape_string($link, $enc[1]);
+ $enc_dur = db_escape_string($link, $enc[2]);
$result = db_query($link, "SELECT id FROM ttrss_enclosures
WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
foreach ($filtered_tags as $tag) {
$tag = sanitize_tag($tag);
- $tag = db_escape_string($tag);
+ $tag = db_escape_string($link, $tag);
if (!tag_is_valid($tag)) continue;
$tags_to_cache = array_unique($tags_to_cache);
- $tags_str = db_escape_string(join(",", $tags_to_cache));
+ $tags_str = db_escape_string($link, join(",", $tags_to_cache));
db_query($link, "UPDATE ttrss_user_entries
SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
}
foreach ($labels as $label) {
- $caption = $label["caption"];
+ $caption = preg_quote($label["caption"]);
- if (preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
+ if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
if (!labels_contains_caption($article_labels, $caption)) {
label_add_article($link, $entry_ref_id, $caption, $owner_uid);
}
} else {
- $error_msg = db_escape_string(mb_substr($rss->error(), 0, 250));
+ $error_msg = db_escape_string($link, mb_substr($rss->error(), 0, 245));
if ($debug_enabled) {
_debug("update_rss_feed: error fetching feed: $error_msg");
$node = $doc->getElementsByTagName('body')->item(0);
- return $doc->saveXML($node, LIBXML_NOEMPTYTAG);
+ return $doc->saveXML($node);
}
function expire_lock_files($debug) {
foreach ($filters as $filter) {
$match_any_rule = $filter["match_any_rule"];
+ $inverse = $filter["inverse"];
$filter_match = false;
foreach ($filter["rules"] as $rule) {
$match = false;
$reg_exp = $rule["reg_exp"];
+ $rule_inverse = $rule["inverse"];
if (!$reg_exp)
continue;
break;
}
+ if ($rule_inverse) $match = !$match;
+
if ($match_any_rule) {
if ($match) {
$filter_match = true;
}
}
+ if ($inverse) $filter_match = !$filter_match;
+
if ($filter_match) {
foreach ($filter["actions"] AS $action) {
array_push($matches, $action);
}
}
- function cache_content($link, $url, $login, $pass) {
-
- $content = fetch_file_contents($url, $login, $pass);
-
- if ($content) {
- $doc = new DOMDocument();
- @$doc->loadHTML($content);
- $xpath = new DOMXPath($doc);
-
- $node = $doc->getElementsByTagName('body')->item(0);
+ function make_guid_from_title($title) {
+ return preg_replace("/[ \"\',.:;]/", "-",
+ mb_strtolower(strip_tags($title), 'utf-8'));
+ }
- if ($node) {
- $content = $doc->saveXML($node, LIBXML_NOEMPTYTAG);
- return $content;
- }
- }
-
- return "";
- }
?>