diff options
Diffstat (limited to 'phpBB')
| -rw-r--r-- | phpBB/posting.php | 276 | ||||
| -rw-r--r-- | phpBB/search.php | 193 |
2 files changed, 182 insertions, 287 deletions
diff --git a/phpBB/posting.php b/phpBB/posting.php index 9d6fb05405..1fa4eae645 100644 --- a/phpBB/posting.php +++ b/phpBB/posting.php @@ -112,48 +112,55 @@ function remove_common($percent, $word_id_list = array()) $row = $db->sql_fetchrow($result); - $common_threshold = floor($row['total_posts'] * $percent); - - $sql = "SELECT word_id - FROM " . SEARCH_MATCH_TABLE . " - $word_id_sql - GROUP BY word_id - HAVING COUNT(word_id) > $common_threshold"; - $result = $db->sql_query($sql); - if( !$result ) + if( $row['total_posts'] > 100 ) { - message_die(GENERAL_ERROR, "Couldn't obtain common word list", "", __LINE__, __FILE__, $sql); - } + $common_threshold = floor($row['total_posts'] * $percent); - if( $post_count = $db->sql_numrows($result) ) - { - $common_word_id_list = array(); - while( $row = $db->sql_fetchrow($result) ) + $sql = "SELECT word_id + FROM " . SEARCH_MATCH_TABLE . " + $word_id_sql + GROUP BY word_id + HAVING COUNT(word_id) > $common_threshold"; + $result = $db->sql_query($sql); + if( !$result ) { - $common_word_id_list[] = $row['word_id']; + message_die(GENERAL_ERROR, "Couldn't obtain common word list", "", __LINE__, __FILE__, $sql); } - $db->sql_freeresult($result); - - if(count($common_word_ids) != 0) + if( $word_count = $db->sql_numrows($result) ) { - $common_word_id_list = implode(", ", $common_word_id_list); - - $sql = "UPDATE " . SEARCH_WORD_TABLE . " - SET word_common = 1 - WHERE word_id IN ($common_word_id_list)"; - $result = $db->sql_query($sql); - if( !$result ) + $common_word_id_list = array(); + while( $row = $db->sql_fetchrow($result) ) { - message_die(GENERAL_ERROR, "Couldn't delete word list entry", "", __LINE__, __FILE__, $sql); + $common_word_id_list[] = $row['word_id']; } - $sql = "DELETE FROM " . SEARCH_WORD_MATCH . " - WHERE word_id IN ($common_word_id_list)"; - $result = $db->sql_query($sql); - if( !$result ) + $db->sql_freeresult($result); + + if( count($common_word_ids) != 0 ) + { + $common_word_id_list = implode(", ", $common_word_id_list); + + $sql = "UPDATE " . SEARCH_WORD_TABLE . " + SET word_common = " . TRUE . " + WHERE word_id IN ($common_word_id_list)"; + $result = $db->sql_query($sql); + if( !$result ) + { + message_die(GENERAL_ERROR, "Couldn't delete word list entry", "", __LINE__, __FILE__, $sql); + } + + $sql = "DELETE FROM " . SEARCH_WORD_MATCH . " + WHERE word_id IN ($common_word_id_list)"; + $result = $db->sql_query($sql); + if( !$result ) + { + message_die(GENERAL_ERROR, "Couldn't delete word match entry", "", __LINE__, __FILE__, $sql); + } + } + else { - message_die(GENERAL_ERROR, "Couldn't delete word match entry", "", __LINE__, __FILE__, $sql); + return 0; } } else @@ -161,120 +168,89 @@ function remove_common($percent, $word_id_list = array()) return 0; } } + else + { + return 0; + } - return $words_removed; + return $word_count; } -function remove_old_words($post_id) +function remove_unmatched_words() { - global $db, $phpbb_root_path, $board_config, $lang; - - $stopword_array = @file($phpbb_root_path . "language/lang_" . $board_config['default_lang'] . "/search_stopwords.txt"); - $synonym_array = @file($phpbb_root_path . "language/lang_" . $board_config['default_lang'] . "/search_synonyms.txt"); + global $db; - $sql = "SELECT post_text - FROM " . POSTS_TEXT_TABLE . " - WHERE post_id = $post_id"; - if( $result = $db->sql_query($sql) ) + switch(SQL_LAYER) { - $row = $db->sql_fetchrow($result); - - $search_text = clean_words($row['post_text'], $stopword_array, $synonym_array); - $search_matches = split_words($search_text); - - if( count($search_matches) ) - { - $word = array(); - $word_count = array(); - $phrase_string = $text; - - $sql_in = ""; - for ($j = 0; $j < count($search_matches); $j++) - { - $this_word = strtolower(trim($search_matches[$j])); - - if( empty($word_count[$this_word]) ) - { - $word_count[$this_word] = 1; - } + case 'postgresql': + $sql = "DELETE FROM " . SEARCH_WORD_TABLE . " + WHERE word_id NOT IN ( + SELECT word_id + FROM " . SEARCH_MATCH_TABLE . ") + GROUP BY word_id"; + $result = $db->sql_query($sql); + if( !$result ) + { + message_die(GENERAL_ERROR, "Couldn't delete old words from word table", __LINE__, __FILE__, $sql); + } - $new_word = true; - for($k = 0; $k < count($word); $k++) - { - if( $this_word == $word[$k] ) - { - $new_word = false; - $word_count[$this_word]++; - } - } + $unmatched_count = $db->sql_affectedrows(); - if( $new_word ) - { - $word[] = $this_word; - } - } + break; - for($j = 0; $j < count($word); $j++) + case 'oracle': + $sql = "DELETE FROM " . SEARCH_WORD_TABLE . " + WHERE word_id IN ( + SELECT w.word_id + FROM " . SEARCH_WORD_TABLE . " w, " . SEARCH_MATCH_TABLE . " m + WHERE w.word_id = m.word_id(+) + AND m.word_id IS NULL)"; + $result = $db->sql_query($sql); + if( !$result ) { - if( $word[$j] ) - { - if( $sql_in != "" ) - { - $sql_in .= ", "; - } - $sql_in .= "'" . $word[$j] . "'"; - } + message_die(GENERAL_ERROR, "Couldn't delete old words from word table", __LINE__, __FILE__, $sql); } - $sql = "SELECT word_id, word_text - FROM " . SEARCH_WORD_TABLE . " - WHERE word_text IN ($sql_in)"; + $unmatched_count = $db->sql_affectedrows(); + + break; + + case 'mssql': + case 'msaccess': + $sql = "DELETE FROM " . SEARCH_WORD_TABLE . " + WHERE word_id IN ( + SELECT w.word_id + FROM " . SEARCH_WORD_TABLE . " w + LEFT JOIN " . SEARCH_MATCH_TABLE . " m ON m.word_id = w.word_id + WHERE m.word_id IS NULL)"; $result = $db->sql_query($sql); if( !$result ) { - message_die(GENERAL_ERROR, "Couldn't select words", "", __LINE__, __FILE__, $sql); + message_die(GENERAL_ERROR, "Couldn't delete old words from word table", __LINE__, __FILE__, $sql); } - if( $word_check_count = $db->sql_numrows($result) ) - { - $check_words = $db->sql_fetchrowset($result); - - $word_id_sql = ""; - for($i = 0; $i < count($check_words); $i++ ) - { - if( $word_id_sql != "" ) - { - $word_id_sql .= ", "; - } - $word_id_sql .= $check_words[$i]['word_id']; - } + $unmatched_count = $db->sql_affectedrows(); - $sql = "SELECT word_id, COUNT(post_id) AS post_occur_count - FROM " . SEARCH_MATCH_TABLE . " - WHERE word_id IN ($word_id_sql) - GROUP BY word_id"; - if( !$result = $db->sql_query($sql) ) - { - message_die(GENERAL_ERROR, "Couldn't obtain search word sums", "", __LINE__, __FILE__, $sql); - } + break; - if( $post_count = $db->sql_numrows($result) ) + case 'mysql': + case 'mysql4': + $sql = "SELECT w.word_id + FROM " . SEARCH_WORD_TABLE . " w + LEFT JOIN " . SEARCH_MATCH_TABLE . " m ON m.word_id = w.word_id + WHERE m.word_id IS NULL"; + if( $result = $db->sql_query($sql) ) + { + if( $unmatched_count = $db->sql_numrows($result) ) { - $rowset = $db->sql_fetchrowset($result); - - $word_id_sql = ""; - for($i = 0; $i < $post_count; $i++) + $rowset = array(); + while( $row = $db->sql_fetchrow($result) ) { - if( $rowset[$i]['post_occur_count'] == 1 ) - { - if( $word_id_sql != "" ) - { - $word_id_sql .= ", "; - } - $word_id_sql .= $rowset[$i]['word_id']; - } + $rowset[] = $row['word_id']; } + $word_id_sql = implode(", ", $rowset); + if( $word_id_sql ) { $sql = "DELETE FROM " . SEARCH_WORD_TABLE . " @@ -285,24 +261,21 @@ function remove_old_words($post_id) message_die(GENERAL_ERROR, "Couldn't delete word list entry", "", __LINE__, __FILE__, $sql); } } + else + { + return 0; + } } - - $sql = "DELETE FROM " . SEARCH_MATCH_TABLE . " - WHERE post_id = $post_id"; - $result = $db->sql_query($sql); - if( !$result ) + else { - message_die(GENERAL_ERROR, "Couldn't delete word match entry for this post", "", __LINE__, __FILE__, $sql); + return 0; } } - } - } - else - { - message_die(GENERAL_ERROR, "Couldn't obtain post text", "", __LINE__, __FILE__, $sql); + + break; } - return; + return $unmatched_count; } function add_search_words($post_id, $post_text, $post_title = "") @@ -1658,13 +1631,21 @@ if( ( $submit || $confirm || $mode == "delete" ) && !$error ) if( $delete || $mode == "delete" ) { - remove_old_words($post_id); + $sql = "DELETE FROM " . SEARCH_MATCH_TABLE . " + WHERE post_id = $post_id"; + $result = $db->sql_query($sql, BEGIN_TRANSACTION); + if( !$result ) + { + message_die(GENERAL_ERROR, "Couldn't delete word match entry for this post", "", __LINE__, __FILE__, $sql); + } + + remove_unmatched_words(); - $sql = "DELETE FROM " . POSTS_TEXT_TABLE . " + $sql = "DELETE FROM " . POSTS_TEXT_TABLE . " WHERE post_id = $post_id"; - if($db->sql_query($sql, BEGIN_TRANSACTION)) + if( $db->sql_query($sql) ) { - $sql = "DELETE FROM " . POSTS_TABLE . " + $sql = "DELETE FROM " . POSTS_TABLE . " WHERE post_id = $post_id"; if($db->sql_query($sql)) { @@ -1676,7 +1657,7 @@ if( ( $submit || $confirm || $mode == "delete" ) && !$error ) // if( $db->sql_query($sql) ) { - $sql = "DELETE FROM " . TOPICS_TABLE . " + $sql = "DELETE FROM " . TOPICS_TABLE . " WHERE topic_id = $topic_id"; if( $db->sql_query($sql) ) { @@ -1882,12 +1863,18 @@ if( ( $submit || $confirm || $mode == "delete" ) && !$error ) } } - remove_old_words($post_id); + $sql = "DELETE FROM " . SEARCH_MATCH_TABLE . " + WHERE post_id = $post_id"; + $result = $db->sql_query($sql, BEGIN_TRANSACTION); + if( !$result ) + { + message_die(GENERAL_ERROR, "Couldn't delete word match entry for this post", "", __LINE__, __FILE__, $sql); + } $sql = "UPDATE " . POSTS_TABLE . " SET bbcode_uid = '$bbcode_uid', enable_bbcode = $bbcode_on, enable_html = $html_on, enable_smilies = $smilies_on, enable_sig = $attach_sig" . $edited_sql . " WHERE post_id = $post_id"; - if($db->sql_query($sql, BEGIN_TRANSACTION)) + if($db->sql_query($sql)) { $sql = "UPDATE " . POSTS_TEXT_TABLE . " SET post_text = '$post_message', post_subject = '$post_subject' @@ -1897,7 +1884,8 @@ if( ( $submit || $confirm || $mode == "delete" ) && !$error ) { if( $db->sql_query($sql) ) { - add_search_words($post_id, stripslashes($post_message)); + add_search_words($post_id, stripslashes($post_message)); + remove_unmatched_words(); // // Update topics table here @@ -2020,8 +2008,8 @@ if( ( $submit || $confirm || $mode == "delete" ) && !$error ) } else { - remove_old_words($post_id); add_search_words($post_id, stripslashes($post_message)); + remove_unmatched_words(); if( $db->sql_query($sql, END_TRANSACTION) ) { diff --git a/phpBB/search.php b/phpBB/search.php index 91eb6b6b13..2db6986e37 100644 --- a/phpBB/search.php +++ b/phpBB/search.php @@ -88,74 +88,6 @@ function split_words(&$entry) return $split_entries[1]; } - -function arrayintersect($array1, $array2) -{ - // - // make sure arguments are actually arrays - // - if( !is_array($array1) || !is_array($array2) ) - { - return false; - } - - if( sizeof($array2) > sizeof($array1) ) - { - $temp = $array1; - $array1 = $array2; - $array2 = $temp; - - unset($temp); - } - - sort($array1); - rsort($array2); - - // the greatest element in array2 - $eg2 = $array2[0]; - - for($i = 0; $i < sizeof($array1); $i++) - { - $e1 = $array1[$i]; - - for($j = 0; $j < sizeof($array2); $j++) - { - $e2 = $array2[$j]; - - if( $e1 > $e2 ) - { - // - // we have passed the match - // - if( $e2 == $eg2 ) - { - break(2); - } - else - { - break; - } - } - else if( $e1 == $e2 ) - { - // - // we found a match - // - $retArray[] = $e1; - - break; - } - } - } - return $retArray; -} - -function inarray($needle, $haystack) -{ - for($i = 0; $i < count($haystack) && $haystack[$i] != $needle; $i++); - - return ( $i != count($haystack) ); -} // // End of functions defns // ---------------------- @@ -352,7 +284,9 @@ else if( $query_keywords != "" || $query_author != "" || $search_id ) $split_search = array(); $split_search = split_words($cleaned_search); + $word_count = 0; $word_match = array(); + $result_list = array(); $current_match_type = "and"; for($i = 0; $i < count($split_search); $i++) @@ -376,20 +310,7 @@ else if( $query_keywords != "" || $query_author != "" || $search_id ) $current_match_type = "and"; } - $word_match[$current_match_type][] = $split_search[$i]; - } - } - - @reset($word_match); - - $word_count = 0; - $result_list = array(); - - while( list($match_type, $match_word_list) = each($word_match) ) - { - for($i = 0; $i < count($match_word_list); $i++ ) - { - $match_word = str_replace("*", "%", $match_word_list[$i]); + $match_word = str_replace("*", "%", $split_search[$i]); $sql = "SELECT m.post_id FROM " . SEARCH_WORD_TABLE . " w, " . SEARCH_MATCH_TABLE . " m @@ -399,49 +320,30 @@ else if( $query_keywords != "" || $query_author != "" || $search_id ) $result = $db->sql_query($sql); if( !$result ) { - message_die(GENERAL_ERROR, "Couldn't matched posts", "", __LINE__, __FILE__, $sql); + message_die(GENERAL_ERROR, "Couldn't obtain matched posts list", "", __LINE__, __FILE__, $sql); } $row = array(); - while( $temp_row = $db->sql_fetchrow($result) ) { $row['' . $temp_row['post_id'] . ''] = 1; - } - - @reset($row); - while( list($post_id, $match_count) = each($row) ) - { if( !$word_count ) { - $result_list['' . $post_id . ''] = $match_count; - } - else if( $match_type == "and" ) - { - $result_list['' . $post_id . ''] = ( $result_list['' . $post_id . ''] ) ? $result_list['' . $post_id . ''] + intval($match_count) : 0; + $result_list['' . $temp_row['post_id'] . ''] = 1; } - else if( $match_type == "or" ) + else if( $current_match_type == "or" ) { - if( $result_list['' . $post_id . ''] ) - { - $result_list['' . $post_id . ''] += intval($match_count); - } - else - { - $result_list['' . $post_id . ''] = 0; - $result_list['' . $post_id . ''] += intval($match_count); - } + $result_list['' . $temp_row['post_id'] . ''] = 1; } - else if( $match_type == "not" ) + else if( $current_match_type == "not" ) { - $result_list['' . $post_id . ''] = 0; + $result_list['' . $temp_row['post_id'] . ''] = 0; } } - if( $match_type == "and" && $word_count ) + if( $current_match_type == "and" && $word_count ) { - @reset($row); @reset($result_list); while( list($post_id, $match_count) = each($result_list) ) @@ -494,28 +396,42 @@ else if( $query_keywords != "" || $query_author != "" || $search_id ) } else { - $sql = "SELECT topic_id - FROM " . POSTS_TABLE . " - WHERE post_id IN ($sql_post_id_in) - GROUP BY topic_id"; - $result = $db->sql_query($sql); - if( !$result ) + switch(SQL_LAYER) { - message_die(GENERAL_ERROR, "Couldn't matched posts", "", __LINE__, __FILE__, $sql); - } + case 'mysql': + case 'mysql4': + $sql = "SELECT topic_id + FROM " . POSTS_TABLE . " + WHERE post_id IN ($sql_post_id_in) + GROUP BY topic_id"; + $result = $db->sql_query($sql); + if( !$result ) + { + message_die(GENERAL_ERROR, "Couldn't matched posts", "", __LINE__, __FILE__, $sql); + } - $sql_post_id_in = ""; - while( $row = $db->sql_fetchrow($result) ) - { - if( $sql_post_id_in != "" ) - { - $sql_post_id_in .= ", "; - } - $sql_post_id_in .= $row['topic_id']; - } + $sql_post_id_in = ""; + while( $row = $db->sql_fetchrow($result) ) + { + if( $sql_post_id_in != "" ) + { + $sql_post_id_in .= ", "; + } + $sql_post_id_in .= $row['topic_id']; + } - $search_sql .= "t.topic_id IN ($sql_post_id_in) "; - + $search_sql .= "t.topic_id IN ($sql_post_id_in) "; + + break; + + default: + $search_sql .= "t.topic_id IN ( + SELECT topic_id + FROM " . POSTS_TABLE . " + WHERE post_id IN ($sql_post_id_in) + GROUP BY topic_id )"; + break; + } } } // @@ -611,11 +527,7 @@ else if( $query_keywords != "" || $query_author != "" || $search_id ) } $total_match_count = $db->sql_numrows($result); -/* if( $total_match_count > 500 ) - { - message_die(GENERAL_MESSAGE, $lang['Too_many_results']);//"Your search returned too many matches, refine your search criteria and try again"; - } -*/ + $searchset = $db->sql_fetchrowset($result); $db->sql_freeresult($result); @@ -868,25 +780,21 @@ else if( $query_keywords != "" || $query_author != "" || $search_id ) { if($return_chars != 0 ) { - if($return_chars != -1) - { - $message = (strlen($message) > $return_chars) ? substr($message, 0, $return_chars) . " ..." : $message; - } - + $bbcode_uid = $searchset[$i]['bbcode_uid']; + // // If the board has HTML off but the post has HTML // on then we process it, else leave it alone // if( $return_chars != -1 ) { - $message = preg_replace("#<([\/]?.*?)>#is", "<\\1>", $message); - $message = preg_replace("/[img\:[0-9a-z\:]+\].*?\[\/img\:[0-9a-z\:]+\]/si", "", $message); - $message = preg_replace("/[\/[a-z\*]+\:[0-9a-z\:]+\]/si", "", $message); + $message = (strlen($message) > $return_chars) ? substr($message, 0, $return_chars) . " ..." : $message; + $message = strip_tags($message); + $message = preg_replace("/\[.*?:$bbcode_uid:?.*?\]/si", "", $message); + $message = preg_replace("/\[url\]|\[\/url\]/si", "", $message); } else { - $bbcode_uid = $searchset[$i]['bbcode_uid']; - $user_sig = $searchset[$i]['user_sig']; $user_sig_bbcode_uid = $searchset[$i]['user_sig_bbcode_uid']; @@ -965,7 +873,6 @@ else if( $query_keywords != "" || $query_author != "" || $search_id ) { $message = ""; - if( count($orig_word) ) { $topic_title = preg_replace($orig_word, $replacement_word, $searchset[$i]['topic_title']); |