sleep(1); // prevent flood (FIXME make this an option?)
}
+ require_once "digest.php";
+
// Send feed digests by email if needed.
send_headlines_digests($link, $debug);
}
$result = db_query($link, "SELECT id,update_interval,auth_login,
- feed_url,auth_pass,cache_images,last_updated,cache_content,
+ feed_url,auth_pass,cache_images,last_updated,
mark_unread_on_update, owner_uid, update_on_checksum_change,
pubsub_state
FROM ttrss_feeds WHERE id = '$feed'");
$auth_pass = db_fetch_result($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);
$rss->set_timeout($no_cache ? 15 : 60);
$rss->set_feed_url($fetch_url);
$rss->set_output_encoding('UTF-8');
- //$rss->force_feed(true);
+ $rss->force_feed(true);
if ($debug_enabled) {
_debug("feed update interval (sec): " .
if (!$rss->error()) {
+ // We use local pluginhost here because we need to load different per-user feed plugins
+ $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid);
+
+ $pluginhost = new PluginHost($link);
+
+ $pluginhost->load(PLUGINS, $pluginhost::KIND_ALL);
+ $pluginhost->load($user_plugins, $pluginhost::KIND_USER, $owner_uid);
+ $pluginhost->load_data();
+
+ $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss);
+
if ($debug_enabled) {
_debug("update_rss_feed: processing feed data...");
}
$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));
+ $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
if ($debug_enabled) {
_debug("update_rss_feed: checking favicon...");
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 ($debug_enabled) {
_debug("update_rss_feed: loading filters & labels...");
}
if (!$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());
}
$entry_title = $item->get_title();
- $entry_link = rewrite_relative_url($site_url, $item->get_link());
+
+ $entry_link = rewrite_relative_url($site_url, htmlspecialchars_decode($item->get_link()));
if ($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()) {
$entry_author = db_escape_string($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'");
+ $entry_guid = db_escape_string(mb_substr($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));
+ $entry_comments = db_escape_string(mb_substr($entry_comments, 0, 245));
+ $entry_author = db_escape_string(mb_substr($entry_author, 0, 245));
$num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments');
_debug("update_rss_feed: looking for tags [2]...");
}
- /* taaaags */
- // <a href="..." rel="tag">Xorg</a>, //
-
- $entry_tags = null;
-
- preg_match_all("/<a.*?rel=['\"]tag['\"].*?\>([^<]+)<\/a>/i",
- $entry_content, $entry_tags);
-
- $entry_tags = $entry_tags[1];
-
- $entry_tags = array_merge($entry_tags, $additional_tags);
- $entry_tags = array_unique($entry_tags);
+ $entry_tags = array_unique($additional_tags);
for ($i = 0; $i < count($entry_tags); $i++)
$entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
}
// TODO: less memory-hungry implementation
- global $pluginhost;
if ($debug_enabled) {
_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($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" => $stored_article);
foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) {
$article = $plugin->hook_article_filter($article);
}
- $entry_title = $article["title"];
- $entry_content = $article["content"];
$entry_tags = $article["tags"];
- $entry_author = $article["author"];
+ $entry_guid = db_escape_string($entry_guid);
+ $entry_content = db_escape_string($article["content"], false);
+ $entry_title = db_escape_string($article["title"]);
+ $entry_author = db_escape_string($article["author"]);
+ $entry_link = db_escape_string($article["link"]);
+ $entry_plugin_data = db_escape_string($article["plugin_data"]);
+
+ if ($debug_enabled) {
+ _debug("update_rss_feed: plugin data: $entry_plugin_data");
+ }
- $entry_content = db_escape_string($entry_content, false);
- $entry_title = db_escape_string($entry_title);
- $entry_author = db_escape_string($entry_author);
- $entry_link = db_escape_string($entry_link);
+ if ($cache_images && is_writable(CACHE_DIR . '/images'))
+ $entry_content = cache_images($entry_content, $site_url, $debug_enabled);
- $content_hash = "SHA1:" . sha1(strip_tags($entry_content));
+ $content_hash = "SHA1:" . sha1($entry_content);
db_query($link, "BEGIN");
+ $result = db_query($link, "SELECT id FROM ttrss_entries
+ WHERE guid = '$entry_guid'");
+
if (db_num_rows($result) == 0) {
if ($debug_enabled) {
- _debug("update_rss_feed: base guid not found");
- }
-
- if ($cache_content) {
- if ($debug_enabled) {
- _debug("update_rss_feed: caching content (initial)...");
- }
-
- $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass);
-
- if ($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);
+ _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',
+ '',
$no_orig_date,
NOW(),
NOW(),
'$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;
$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) != $entry_title) {
db_query($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) {
} else {
- $error_msg = mb_substr($rss->error(), 0, 250);
+ $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
if ($debug_enabled) {
_debug("update_rss_feed: error fetching feed: $error_msg");
}
- $error_msg = db_escape_string($error_msg);
-
db_query($link,
"UPDATE ttrss_feeds SET last_error = '$error_msg',
last_updated = NOW() WHERE id = '$feed'");
return "";
}
+
+ function make_guid_from_title($title) {
+ return preg_replace("/[ \"\',.:;]/", "-",
+ mb_strtolower(strip_tags($title), 'utf-8'));
+ }
+
+
?>
git.wh0rd.org / tt-rss.git / blobdiff