]> git.wh0rd.org - tt-rss.git/blobdiff - include/rssfuncs.php
add ability to auto-assign articles to labels (bump schema)
[tt-rss.git] / include / rssfuncs.php
index 47e0c68f32612c98f4261bf4c2a7e63175ad9ff1..1e2feb3cba46ed611cdcc548ee35a0bf9d7e3723 100644 (file)
@@ -1,7 +1,7 @@
 <?php
        define('DAEMON_UPDATE_LOGIN_LIMIT', 30);
        define('DAEMON_FEED_LIMIT', 100);
-       define('DAEMON_SLEEP_INTERVAL', 120);
+       define('DAEMON_SLEEP_INTERVAL', 60);
 
        function update_feedbrowser_cache($link) {
 
 
        } // function update_daemon_common
 
-       function fetch_twitter_rss($link, $url, $owner_uid) {
-
-               require_once 'lib/tmhoauth/tmhOAuth.php';
-               require_once "lib/magpierss/rss_fetch.inc";
-               require_once 'lib/magpierss/rss_utils.inc';
-
-               $result = db_query($link, "SELECT twitter_oauth FROM ttrss_users
-                       WHERE id = $owner_uid");
-
-               $access_token = json_decode(db_fetch_result($result, 0, 'twitter_oauth'), true);
-               $url_escaped = db_escape_string($url);
-
-               if ($access_token) {
-
-                       $tmhOAuth = new tmhOAuth(array(
-                               'consumer_key'    => CONSUMER_KEY,
-                               'consumer_secret' => CONSUMER_SECRET,
-                               'user_token' => $access_token['oauth_token'],
-                               'user_secret' => $access_token['oauth_token_secret'],
-                       ));
-
-                       $code = $tmhOAuth->request('GET', $url,
-                               convertUrlQuery(parse_url($url, PHP_URL_QUERY)));
-
-                       if ($code == 200) {
-
-                               $content = $tmhOAuth->response['response'];
-
-                               define('MAGPIE_CACHE_ON', false);
-
-                               $rss = new MagpieRSS($content, MAGPIE_OUTPUT_ENCODING,
-                                       MAGPIE_INPUT_ENCODING, MAGPIE_DETECT_ENCODING );
-
-                               return $rss;
-
-                       } else {
-
-                               db_query($link, "UPDATE ttrss_feeds
-                                       SET last_error = 'OAuth authorization failed ($code).'
-                                       WHERE feed_url = '$url_escaped' AND owner_uid = $owner_uid");
-                       }
-
-               } else {
-
-                       db_query($link, "UPDATE ttrss_feeds
-                               SET last_error = 'OAuth information not found.'
-                               WHERE feed_url = '$url_escaped' AND owner_uid = $owner_uid");
-
-                       return false;
-               }
-       }
-
        function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false,
                $override_url = false) {
 
                $cache_age = (is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ?
                        -1 : get_feed_update_interval($link, $feed) * 60;
 
-               if ($update_method == 3) {
-                       $rss = fetch_twitter_rss($link, $fetch_url, $owner_uid);
-               } else if ($update_method == 1) {
+               if ($update_method == 1) {
 
                        define('MAGPIE_CACHE_AGE', $cache_age);
                        define('MAGPIE_CACHE_ON', !$no_cache);
-                       define('MAGPIE_FETCH_TIME_OUT', 60);
+                       define('MAGPIE_FETCH_TIME_OUT', $no_cache ? 15 : 60);
                        define('MAGPIE_CACHE_DIR', CACHE_DIR . "/magpie");
 
                        $rss = @fetch_rss($fetch_url);
 
                        $rss = new SimplePie();
                        $rss->set_useragent(SELF_USER_AGENT);
-#                      $rss->set_timeout(10);
+                       $rss->set_timeout($no_cache ? 15 : 60);
                        $rss->set_feed_url($fetch_url);
                        $rss->set_output_encoding('UTF-8');
                        //$rss->force_feed(true);
                        }
 
                        if ($debug_enabled) {
-                               _debug("update_rss_feed: loading filters...");
+                               _debug("update_rss_feed: loading filters & labels...");
                        }
 
                        $filters = load_filters($link, $feed, $owner_uid);
+                       $labels = get_all_labels($link, $owner_uid);
 
-//                     if ($debug_enabled) {
-//                             print_r($filters);
-//                     }
+                       if ($debug_enabled) {
+                               //print_r($filters);
+                               _debug("update_rss_feed: " . count($filters) . " filters loaded.");
+                       }
 
                        if ($use_simplepie) {
                                $iterator = $rss->get_items();
                                        //print_r($entry_tags);
                                }
 
-                               # sanitize content
-
-                               $entry_content = sanitize_article_content($entry_content);
-                               $entry_title = sanitize_article_content($entry_title);
-
                                if ($debug_enabled) {
                                        _debug("update_rss_feed: done collecting data [TITLE:$entry_title]");
                                }
                                                }
                                        }
 
+                                       $article_labels = get_article_labels($link, $entry_ref_id);
+
                                        if (find_article_filter($article_filters, "filter")) {
                                                db_query($link, "COMMIT"); // close transaction in progress
                                                continue;
                                        _debug("update_rss_feed: assigning labels...");
                                }
 
-                               assign_article_to_labels($link, $entry_ref_id, $article_filters,
-                                       $owner_uid);
+                               assign_article_to_label_filters($link, $entry_ref_id, $article_filters,
+                                       $owner_uid, $article_labels);
 
                                if ($debug_enabled) {
                                        _debug("update_rss_feed: looking for enclosures...");
                                        db_query($link, "COMMIT");
                                }
 
+                               if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) {
+                                       if ($debug_enabled) {
+                                               _debug("update_rss_feed: auto-assigning labels...");
+                                       }
+
+                                       foreach ($labels as $label) {
+                                               $caption = $label["caption"];
+
+                                               if (preg_match("/\b$caption\b/i", "$tags_str $entry_content $entry_title")) {
+                                                       if (!labels_contains_caption($article_labels, $caption)) {
+                                                               label_add_article($link, $entry_ref_id, $caption, $owner_uid);
+                                                       }
+                                               }
+                                       }
+                               }
+
                                if ($debug_enabled) {
                                        _debug("update_rss_feed: article processed");
                                }
 
                return $params;
        }
+
+       function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
+               $matches = array();
+
+               foreach ($filters as $filter) {
+                       $match_any_rule = $filter["match_any_rule"];
+                       $filter_match = false;
+
+                       foreach ($filter["rules"] as $rule) {
+                               $match = false;
+                               $reg_exp = $rule["reg_exp"];
+
+                               if (!$reg_exp)
+                                       continue;
+
+                               switch ($rule["type"]) {
+                               case "title":
+                                       $match = @preg_match("/$reg_exp/i", $title);
+                                       break;
+                               case "content":
+                                       $match = @preg_match("/$reg_exp/i", $content);
+                                       break;
+                               case "both":
+                                       $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $title));
+                                       break;
+                               case "link":
+                                       $match = @preg_match("/$reg_exp/i", $link);
+                                       break;
+                               case "author":
+                                       $match = @preg_match("/$reg_exp/i", $author);
+                                       break;
+                               case "tag":
+                                       $tag_string = join(",", $tags);
+                                       $match = @preg_match("/$reg_exp/i", $tag_string);
+                                       break;
+                               }
+
+                               if ($match_any_rule) {
+                                       if ($match) {
+                                               $filter_match = true;
+                                               break;
+                                       }
+                               } else {
+                                       $filter_match = $match;
+                                       if (!$match) {
+                                               break;
+                                       }
+                               }
+                       }
+
+                       if ($filter_match) {
+                               foreach ($filter["actions"] AS $action) {
+                                       array_push($matches, $action);
+                               }
+                       }
+               }
+
+               return $matches;
+       }
+
+       function find_article_filter($filters, $filter_name) {
+               foreach ($filters as $f) {
+                       if ($f["type"] == $filter_name) {
+                               return $f;
+                       };
+               }
+               return false;
+       }
+
+       function find_article_filters($filters, $filter_name) {
+               $results = array();
+
+               foreach ($filters as $f) {
+                       if ($f["type"] == $filter_name) {
+                               array_push($results, $f);
+                       };
+               }
+               return $results;
+       }
+
+       function calculate_article_score($filters) {
+               $score = 0;
+
+               foreach ($filters as $f) {
+                       if ($f["type"] == "score") {
+                               $score += $f["param"];
+                       };
+               }
+               return $score;
+       }
+
+       function labels_contains_caption($labels, $caption) {
+               foreach ($labels as $label) {
+                       if ($label[1] == $caption) {
+                               return true;
+                       }
+               }
+
+               return false;
+       }
+
+       function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) {
+               foreach ($filters as $f) {
+                       if ($f["type"] == "label") {
+                               if (!labels_contains_caption($article_labels, $f["param"])) {
+                                       label_add_article($link, $id, $f["param"], $owner_uid);
+                               }
+                       }
+               }
+       }
 ?>