lib/dbeng/dbfts_mysql.php
<?php
class dbfts_mysql extends dbfts_abs
{
// LIST OF STOP WORDS: http://dev.mysql.com/doc/refman/5.0/en/fulltext-stopwords.html
// Array copied from http://www.linein.org/blog/2008/11/10/mysql-full-text-stopwords-array/
private $stop_words = ['a\'s', 'able', 'about', 'above', 'according', 'accordingly', 'across', 'actually',
'after', 'afterwards', 'again', 'against', 'ain\'t', 'all', 'allow', 'allows', 'almost',
'alone', 'along', 'already', 'also', 'although', 'always', 'am', 'among', 'amongst', 'an',
'and', 'another', 'any', 'anybody', 'anyhow', 'anyone', 'anything', 'anyway', 'anyways',
'anywhere', 'apart', 'appear', 'appreciate', 'appropriate', 'are', 'aren\'t', 'around', 'as',
'aside', 'ask', 'asking', 'associated', 'at', 'available', 'away', 'awfully', 'be', 'became',
'because', 'become', 'becomes', 'becoming', 'been', 'before', 'beforehand', 'behind', 'being',
'believe', 'below', 'beside', 'besides', 'best', 'better', 'between', 'beyond', 'both', 'brief',
'but', 'by', 'c\'mon', 'c\'s', 'came', 'can', 'can\'t', 'cannot', 'cant', 'cause', 'causes',
'certain', 'certainly', 'changes', 'clearly', 'co', 'com', 'come', 'comes', 'concerning',
'consequently', 'consider', 'considering', 'contain', 'containing', 'contains', 'corresponding',
'could', 'couldn\'t', 'course', 'currently', 'definitely', 'described', 'despite', 'did', 'didn\'t',
'different', 'do', 'does', 'doesn\'t', 'doing', 'don\'t', 'done', 'down', 'downwards', 'during',
'each', 'edu', 'eg', 'eight', 'either', 'else', 'elsewhere', 'enough', 'entirely', 'especially',
'et', 'etc', 'even', 'ever', 'every', 'everybody', 'everyone', 'everything', 'everywhere', 'ex',
'exactly', 'example', 'except', 'far', 'few', 'fifth', 'first', 'five', 'followed', 'following',
'follows', 'for', 'former', 'formerly', 'forth', 'four', 'from', 'further', 'furthermore', 'get',
'gets', 'getting', 'given', 'gives', 'go', 'goes', 'going', 'gone', 'got', 'gotten', 'greetings',
'had', 'hadn\'t', 'happens', 'hardly', 'has', 'hasn\'t', 'have', 'haven\'t', 'having', 'he',
'he\'s', 'hello', 'help', 'hence', 'her', 'here', 'here\'s', 'hereafter', 'hereby', 'herein',
'hereupon', 'hers', 'herself', 'hi', 'him', 'himself', 'his', 'hither', 'hopefully', 'how',
'howbeit', 'however', 'i\'d', 'i\'ll', 'i\'m', 'i\'ve', 'ie', 'if', 'ignored', 'immediate',
'in', 'inasmuch', 'inc', 'indeed', 'indicate', 'indicated', 'indicates', 'inner', 'insofar',
'instead', 'into', 'inward', 'is', 'isn\'t', 'it', 'it\'d', 'it\'ll', 'it\'s', 'its', 'itself',
'just', 'keep', 'keeps', 'kept', 'know', 'knows', 'known', 'last', 'lately', 'later', 'latter',
'latterly', 'least', 'less', 'lest', 'let', 'let\'s', 'like', 'liked', 'likely', 'little',
'look', 'looking', 'looks', 'ltd', 'mainly', 'many', 'may', 'maybe', 'me', 'mean', 'meanwhile',
'merely', 'might', 'more', 'moreover', 'most', 'mostly', 'much', 'must', 'my', 'myself', 'name',
'namely', 'nd', 'near', 'nearly', 'necessary', 'need', 'needs', 'neither', 'never', 'nevertheless',
'new', 'next', 'nine', 'no', 'nobody', 'non', 'none', 'noone', 'nor', 'normally', 'not', 'nothing',
'novel', 'now', 'nowhere', 'obviously', 'of', 'off', 'often', 'oh', 'ok', 'okay', 'old', 'on', 'once',
'one', 'ones', 'only', 'onto', 'or', 'other', 'others', 'otherwise', 'ought', 'our', 'ours',
'ourselves', 'out', 'outside', 'over', 'overall', 'own', 'particular', 'particularly', 'per',
'perhaps', 'placed', 'please', 'plus', 'possible', 'presumably', 'probably', 'provides', 'que',
'quite', 'qv', 'rather', 'rd', 're', 'really', 'reasonably', 'regarding', 'regardless', 'regards',
'relatively', 'respectively', 'right', 'said', 'same', 'saw', 'say', 'saying', 'says', 'second',
'secondly', 'see', 'seeing', 'seem', 'seemed', 'seeming', 'seems', 'seen', 'self', 'selves',
'sensible', 'sent', 'serious', 'seriously', 'seven', 'several', 'shall', 'she', 'should', 'shouldn\'t',
'since', 'six', 'so', 'some', 'somebody', 'somehow', 'someone', 'something', 'sometime', 'sometimes',
'somewhat', 'somewhere', 'soon', 'sorry', 'specified', 'specify', 'specifying', 'still', 'sub', 'such',
'sup', 'sure', 't\'s', 'take', 'taken', 'tell', 'tends', 'th', 'than', 'thank', 'thanks', 'thanx',
'that', 'that\'s', 'thats', 'the', 'their', 'theirs', 'them', 'themselves', 'then', 'thence', 'there',
'there\'s', 'thereafter', 'thereby', 'therefore', 'therein', 'theres', 'thereupon', 'these', 'they',
'they\'d', 'they\'ll', 'they\'re', 'they\'ve', 'think', 'third', 'this', 'thorough', 'thoroughly',
'those', 'though', 'three', 'through', 'throughout', 'thru', 'thus', 'to', 'together', 'too', 'took',
'toward', 'towards', 'tried', 'tries', 'truly', 'try', 'trying', 'twice', 'two', 'un', 'under',
'unfortunately', 'unless', 'unlikely', 'until', 'unto', 'up', 'upon', 'us', 'use', 'used',
'useful', 'uses', 'using', 'usually', 'value', 'various', 'very', 'via', 'viz', 'vs', 'want',
'wants', 'was', 'wasn\'t', 'way', 'we', 'we\'d', 'we\'ll', 'we\'re', 'we\'ve', 'welcome', 'well',
'went', 'were', 'weren\'t', 'what', 'what\'s', 'whatever', 'when', 'whence', 'whenever', 'where',
'where\'s', 'whereafter', 'whereas', 'whereby', 'wherein', 'whereupon', 'wherever', 'whether',
'which', 'while', 'whither', 'who', 'who\'s', 'whoever', 'whole', 'whom', 'whose', 'why', 'will',
'willing', 'wish', 'with', 'within', 'without', 'won\'t', 'wonder', 'would', 'would', 'wouldn\'t',
'yes', 'yet', 'you', 'you\'d', 'you\'ll', 'you\'re', 'you\'ve', 'your', 'yours', 'yourself',
'yourselves', 'zero', ];
/*
* Constructs a query part to match textfields. Abstracted so we can use
* a database specific FTS engine if one is provided by the DBMS
*/
public function createTextQuery($searchFields, $additionalFields)
{
SpotTiming::start(__CLASS__.'::'.__FUNCTION__);
/*
* Initialize some basic values which are used as return values to
* make sure always return a valid set
*/
$filterValueSql = [];
$sortFields = [];
$addFields = [];
/*
* MySQL's fultxt search has a minimum length of words for indexes. Per default this is
* a minimum word length of 4. This means that a searchstring like 'Top 40' will not
* be found because both 'Top' and '40' are shorter than 4 characters.
*
* We query the server setting, and if this is the case, we fall back to a basic LIKE
* search because it has no such limitation
*/
$serverSetting = $this->_db->arrayQuery("SHOW VARIABLES WHERE variable_name = 'ft_min_word_len'");
$minWordLen = $serverSetting[0]['Value'];
//var_dump($searchFields);
foreach ($searchFields as $searchItem) {
$hasTooShortWords = false;
$hasLongEnoughWords = false;
$hasStopWords = false;
$hasNoStopWords = false;
$hasSearchOpAsTerm = false;
$hasPhraseWithOnlyInvalids = false;
$searchMode = 'match-natural';
$searchValue = trim($searchItem['value']);
$field = $searchItem['fieldname'];
/*
* Look at each individual word. If it is shorter than $minWordLen, we have to perform
* a LIKE search as well
*
* We do not use splitWords() here, because we need to have the lengths of the
* individual words not of the phrase search.
*
* There is one exception, if a phrase search contains ONLY stop words, we cannot
* use the FTS at all so we should be aware of that.
*/
$tempSearchValue = str_replace(['+', '-', '"', '(', ')', 'AND', 'NOT', 'OR'], '', $searchValue);
$termList = explode(' ', $tempSearchValue);
foreach ($termList as $term) {
if ((strlen($term) < $minWordLen) && (strlen($term) > 0)) {
$hasTooShortWords = true;
} // if
if (strlen($term) >= $minWordLen) {
$hasLongEnoughWords = true;
} // if
/*
* If the term is a stopword (things like: the, it, ...) we have to
* fallback to a like search as well.
*/
if (in_array(strtolower($term), $this->stop_words) !== false) {
$hasStopWords = true;
} else {
$hasNoStopWords = true;
}// if
} // foreach
/*
* remove any double whitespace because else the MySQL matcher will never
* find anything
*/
$searchValue = str_replace(' ', ' ', $searchValue);
/*
* MySQL has several types of searches - both boolean and natural matching for
* FTS is possible.
*
* We try some heuristics to select the most appropriate type of search.
* If a word starts with either an '+' or an '-', we switch to boolean match
*/
$termList = $this->splitWords($searchValue);
foreach ($termList as $term) {
/*
* We strip some characters because these are valid, but
* can cause the system to not recognize them as operators.
*
* The string: "(<test)" is such an example -- we only check
* the first character so we need to remove the parenthesis.
*/
$strippedTerm = trim($term, "()'\"");
/*
* If after stripping the term of these characters, no string
* is left, make sure we just abort the matching
*/
if (strlen($strippedTerm) < 1) {
continue;
} // if
/*
* + and - are only allowed at the beginning of the search to
* enforce it as an search operator. If they are in the
* words themselves, we fall back to LIKE
*/
if ((strpos($strippedTerm, '-') > 0) || (strpos($strippedTerm, '+') > 0) || (strpos($strippedTerm, '/') > 0)) {
$hasSearchOpAsTerm = true;
} // if
/*
* When there are boolean operators in the string, it's an
* boolean search
*/
if (strlen($strippedTerm[0]) > 0) {
if (strpos('+-~<>', $strippedTerm[0]) !== false) {
$searchMode = 'match-boolean';
$strippedTerm = trim($strippedTerm, '+-');
} // if
}
if (strlen(substr($strippedTerm, -1)) > 0) {
if (strpos('*', substr($strippedTerm, -1)) !== false) {
$searchMode = 'match-boolean';
$strippedTerm = trim($strippedTerm, '*');
} // if
}
if (strlen(substr($term, -1)) > 0) {
if (strpos('"', substr($term, -1)) !== false) {
$searchMode = 'match-boolean';
} // if
}
/*
* We get the complete phrase here, we need to look into
* the phrase terms, because if it only contains invalid terms
* (eg: only stopwords, only shortwords, or a combination thereof),
* we must disable the FTS completely.
*
*/
if ((!$hasPhraseWithOnlyInvalids) && ($term[0] == '"')) {
$tmpFoundValidTerms = false;
$tmpTermList = explode(' ', $strippedTerm);
foreach ($tmpTermList as $tmpTerm) {
if (strlen($tmpTerm) >= $minWordLen) {
if (in_array(strtolower($tmpTerm), $this->stop_words) === false) {
$tmpFoundValidTerms = true;
} // if
} // if
} // foreach
if (!$tmpFoundValidTerms) {
$hasPhraseWithOnlyInvalids = true;
} // if
} // if
} // foreach
// Actually determine the searchmode
/*
* Test cases:
*
* 9th Company
* Ubuntu 11
* Top 40
* "Top 40"
* South Park
* Sex and the city
* Rio
* "sex and the city 2"
* Just Go With It (fallback naar like, enkel stopwoorden of te kort)
* "Just Go With It" (fallback naar like, en quotes gestripped)
* +"taken 2" +(2012) (fallback naar like, en quotes gestripped - enkel stop woorden maar operators)
* +empire +sun
* x-art (like search because it contains an -)
* 50/50 (like search because it contains an /)
* Arvo -Lamentate (natural without like)
* +"Phantom" +(2013) <- Shouldn't use a LIKE per se
* "The Top" <- Should use a LIKE as its only keywords
* +"Warehouse 13" +S04 <- Shouldn't use a LIKE per se
*/
if ($hasPhraseWithOnlyInvalids) {
$searchMode = 'normal';
} elseif (($hasTooShortWords || $hasStopWords) && ($hasLongEnoughWords || $hasNoStopWords) && (!$hasSearchOpAsTerm)) {
if (($hasStopWords && !$hasNoStopWords) || ($hasTooShortWords && !$hasLongEnoughWords)) {
$searchMode = 'normal';
} else {
$searchMode = 'both-'.$searchMode;
} // else
} elseif ((($hasTooShortWords || $hasStopWords) && (!$hasLongEnoughWords && !$hasNoStopWords)) || $hasSearchOpAsTerm) {
$searchMode = 'normal';
} // else
/*
echo 'hasStopWords : ' . (int) $hasStopWords . '<br>';
echo 'hasLongEnoughWords : ' . (int) $hasLongEnoughWords . '<br>';
echo 'hasTooShortWords : ' . (int) $hasTooShortWords . '<br>';
echo 'hasNoStopWords : ' . (int) $hasNoStopWords . '<br>';
echo 'hasSearchOpAsTerm : ' . (int) $hasSearchOpAsTerm . '<br>';
echo 'hasPhraseWithOnlyInvalids : ' . (int) $hasPhraseWithOnlyInvalids . '<br>';
echo 'searchmode : ' . $searchMode . '<br>';
die();
*/
/*
* Start constructing the query. Sometimes we construct the query
* both with a LIKE and with a MATCH statement
*/
$queryPart = [];
$matchPart = '';
if (($searchMode == 'normal') || ($searchMode == 'both-match-natural') || ($searchMode == 'both-match-boolean')) {
$splitted = $this->splitWords($searchValue);
foreach ($splitted as $splittedTerm) {
/*
* If the term contains an boolean operator in the beginning,
* strip it
*/
$filteredTerm = trim($splittedTerm, '"');
$filteredTerm = ltrim($filteredTerm, '+-~<>');
$filteredTerm = rtrim($filteredTerm, '*');
if (!empty($filteredTerm)) {
$filteredTerm = str_replace(' ', '_', $filteredTerm);
$filteredTerm = stripslashes($filteredTerm);
$filteredTerm = str_replace('"', '', $filteredTerm);
$filteredTerm = str_replace('+', '', $filteredTerm);
$queryPart[] = ' '.$field.' LIKE '.$this->_db->safe('%'.$filteredTerm.'%');
} // if
} // foreach
} // if
if (($searchMode == 'match-natural') || ($searchMode == 'both-match-natural')) {
/* Natural language mode always defaults in MySQL 5.0 en 5.1, but cannot be explicitly defined in MySQL 5.0 */
$matchPart = ' MATCH('.$field.') AGAINST ('.$this->_db->safe($searchValue).')';
$queryPart[] = $matchPart;
} // if
/*
* Boolean searches with required or missing terms, will never match if the terms are
* stopwords because stopwords are not in the index and cannot be found
*/
if (($searchMode == 'match-boolean') || ($searchMode == 'both-match-boolean')) {
$matchPart = ' MATCH('.$field.') AGAINST ('.$this->_db->safe($searchValue).' IN BOOLEAN MODE)';
$queryPart[] = $matchPart;
} // if
/*
* Add the textqueries with an AND per search term
*/
$filterValueSql[] = ' ('.implode(' AND ', $queryPart).') ';
/*
* We add these extended textqueries as a column to the filterlist
* and use it as a relevance column. This allows us to sort on
* relevance
*/
if ($searchMode != 'normal') {
/*
* if we get multiple textsearches, we sort them per order
* in the system
*/
$tmpSortCounter = count($additionalFields) + count($addFields);
$addFields[] = $matchPart.' AS searchrelevancy'.$tmpSortCounter;
$sortFields[] = ['field' => 'searchrelevancy'.$tmpSortCounter,
'direction' => 'DESC',
'autoadded' => true,
'friendlyname' => null, ];
} // if
} // foreach
SpotTiming::stop(__CLASS__.'::'.__FUNCTION__, [$filterValueSql, $addFields, $sortFields]);
//var_dump($filterValueSql);
//var_dump($addFields);
//var_dump($sortFields);
//die();
return ['filterValueSql' => $filterValueSql,
'additionalTables' => [],
'additionalFields' => $addFields,
'sortFields' => $sortFields, ];
}
// createTextQuery()
} // dbfts_mysql