-// Clean up an entry (posting), remove HTML, BBcode, stopwords, etc

-function clean_words($entry, &$search, &$replace)

-{

- // Weird, $init_match doesn't work with static when double quotes (") are used...

- static $init_match = array('^', '$', '&', '(', ')', '<', '>', '`', "'", '|', ',', '@', '_', '?', '%');

- static $init_replace = array(" ", " ", " ", " ", " ", " ", " ", " ", "", " ", " ", " ", " ", " ", " ");

-

- static $later_match = array("-", "~", "+", ".", "[", "]", "{", "}", ":", "\\", "/", "=", "#", "\"", ";", "*", "!");

- static $later_replace = array(" ", " ", " ", " ", " ", " ", " ", " ", " ", " " , " ", " ", " ", " ", " ", " ", " ");

-

- static $sgml_match = array("&nbsp;", "&szlig;", "&agrave;", "&aacute;", "&acirc;", "&atilde;", "&auml;", "&aring;", "&aelig;", "&ccedil;", "&egrave;", "&eacute;", "&ecirc;", "&euml;", "&igrave;", "&iacute;", "&icirc;", "&iuml;", "&eth;", "&ntilde;", "&ograve;", "&oacute;", "&ocirc;", "&otilde;", "&ouml;", "&oslash;", "&ugrave;", "&uacute;", "&ucirc;", "&uuml;", "&yacute;", "&thorn;", "&yuml;");

- static $sgml_replace = array(" ", "s", "a", "a", "a", "a", "a", "a", "a", "c", "e", "e", "e", "e", "i", "i", "i", "i", "o", "n", "o", "o", "o", "o", "o", "o", "u", "u", "u", "u", "y", "t", "y");

-

- static $accent_match = array("ß", "à", "á", "â", "ã", "ä", "å", "æ", "ç", "è", "é", "ê", "ë", "ì", "í", "î", "ï", "ð", "ñ", "ò", "ó", "ô", "õ", "ö", "ø", "ù", "ú", "û", "ü", "ý", "þ", "ÿ");

- static $accent_replace = array("s", "a", "a", "a", "a", "a", "a", "a", "c", "e", "e", "e", "e", "i", "i", "i", "i", "o", "n", "o", "o", "o", "o", "o", "o", "u", "u", "u", "u", "y", "t", "y");

-

- $entry = " " . strip_tags(strtolower($entry)) . " ";

-

- $entry = str_replace($sgml_match, $sgml_match, $entry);

- $entry = str_replace($accent_match, $accent_replace, $entry);

-

- // Replace line endings by a space

- $entry = preg_replace("/[\n\r]/is", " ", $entry);

- // Remove URL's

- $entry = preg_replace("/\b[a-z0-9]+:\/\/[a-z0-9\.\-]+(\/[a-z0-9\?\.%_\-\+=&\/]+)?/si", " ", $entry);

-

- // Filter out strange characters like ^, $, &, change "it's" to "its"

- // str_replace with arrays is buggy in some PHP versions so traverse the arrays manually ;(

- for($i = 0; $i < count($init_match); $i++)

- {

- $entry = str_replace($init_match[$i], $init_replace[$i], $entry);

- }

-

- // Quickly remove BBcode.

- $entry = preg_replace("/\[code:[0-9]+:[0-9a-z]{10,}\].*?\[\/code:[0-9]+:[0-9a-z]{10,}\]/is", " ", $entry);

- $entry = preg_replace("/\[img\].*?\[\/img\]/is", " ", $entry);

- $entry = preg_replace("/\[\/?[a-z\*=\+\-]+[0-9a-z]?(\:[a-z0-9]+)?:[a-z0-9]{10,}(\:[a-z0-9]+)?=?.*?\]/si", " ", $entry);

- // URLs

- $entry = preg_replace("/\[\/?[a-z\*]+[=\+\-]?[0-9a-z]+?:[a-z0-9]{10,}[=.*?]?\]/si", " ", $entry);

- $entry = preg_replace("/\[\/?url(=.*?)?\]/si", " ", $entry);

- // Numbers

- $entry = preg_replace("/\b[0-9]+\b/si", " ", $entry);

- // HTML entities like &1234;

- $entry = preg_replace("/\b&[a-z]+;\b/is", " ", $entry);

- // 'words' that consist of <2 or >50 characters are removed.

- $entry = preg_replace("/\b[a-z0-9]{1,2}?\b/si", " ", $entry);

- $entry = preg_replace("/\b[a-z0-9]{50,}?\b/si", " ", $entry);

-

- // Remove some more strange characters

- for($i = 0; $i < count($later_match); $i++)

- {

- $entry = str_replace($later_match[$i], $later_replace[$i], $entry);

- }

-

- // Remove stopwords

- $entry = preg_replace($search, $replace, $entry);

-

- return $entry;

-}

-

-function split_words(&$entry)

-{

- preg_match_all("/\b(\w[\w']*\w+|\w+?)\b/", $entry, $split_entries);

-

- return $split_entries[1];

-}

-

-function remove_common($percent, $delete_common = 0)

-{

- global $db;

-

- $sql = "SELECT COUNT(DISTINCT post_id) as total_posts

- FROM " . SEARCH_MATCH_TABLE;

- $result = $db->sql_query($sql);

- if( !$result )

- {

- $error = $db->sql_error();

- die("Couldn't get maximum post ID :: " . $sql . " :: " . $error['message']);

- }

- $total_posts = $db->sql_fetchrow($result);

- $total_posts = $total_posts['total_posts'];

-

- $common_threshold = floor($total_posts * ( $percent / 100 ));

-

- $sql = "SELECT word_id

- FROM " . SEARCH_MATCH_TABLE . "

- GROUP BY word_id

- HAVING count(word_id) > $common_threshold";

- $result = $db->sql_query($sql);

- if( !$result )

- {

- $error = $db->sql_error();

- die("Couldn't obtain common word list :: " . $sql . " :: " . $error['message']);

- }

- $common_words = $db->sql_numrows($result);

-

- while($row = $db->sql_fetchrow($result))

- {

- $common_word_ids[] = $row['word_id'];

- }

- $db->sql_freeresult($result);

-

- if(count($common_word_ids) != 0)

- {

- $common_word_ids = implode(',',$common_word_ids);

- }

- else

- {

- // We didn't remove any common words

- return 0;

- }

-

- $sql = "UPDATE ". SEARCH_WORD_TABLE ."

- SET word_common = 1

- WHERE word_id IN ($common_word_ids)";

- $result = $db->sql_query($sql);

- if( !$result )

- {

- $error = $db->sql_error();

- die("Couldn't delete word list entry :: " . $sql . " :: " . $error['message']);

- }

-

- if( $delete_common)

- {

- $sql = "DELETE FROM ".SEARCH_MATCH_TABLE."

- WHERE word_id IN ($common_word_ids)";

- $result = $db->sql_query($sql);

- if( !$result )

- {

- $error = $db->sql_error();

- die("Couldn't delete word match entry :: " . $sql . " :: " . $error['message']);

- }

- }

-

- return $common_words;

-}

-

- print "Removed ". remove_common($common_percent, 1) ." words that where too common.<br>";