/**
* Converts given $text into a URL slug consisting of URL valid characters.
* For non-Unicode setups this means character in the range a-z, numbers and _, for Unicode
* setups it means all characters except space, &, ;, /, :, =, ?, [, ], (, ), -.
*
* Invalid characters are converted to -.
*
* Example with a non-Unicode setup
*
* 'My car' => 'My-car'
* 'What is this?' => 'What-is-this'
* 'This & that' => 'This-that'
* 'myfile.tpl' => 'Myfile-tpl',
* 'øæå' => 'oeaeaa'
*
* @param string $text
* @param string $defaultText
* @param string|null $transformation
*
* @return string
*/
public function convert($text, $defaultText = '_1', $transformation = null)
{
if (!isset($transformation)) {
$transformation = $this->configuration['transformation'];
}
if (strlen($text) === 0) {
$text = $defaultText;
}
if (isset($this->configuration['transformationGroups'][$transformation]['commands']) && !empty($this->configuration['transformationGroups'][$transformation]['commands'])) {
$text = $this->transformationProcessor->transform($text, $this->configuration['transformationGroups'][$transformation]['commands']);
}
return $this->cleanupText($text, $this->configuration['transformationGroups'][$transformation]['cleanupMethod']);
}
/**
* Converts given $text into a URL slug consisting of URL valid characters.
* For non-Unicode setups this means character in the range a-z, numbers and _, for Unicode
* setups it means all characters except space, &, ;, /, :, =, ?, [, ], (, ), -
*
* Invalid characters are converted to -.
*
* Example with a non-Unicode setup
*
* 'My car' => 'My-car'
* 'What is this?' => 'What-is-this'
* 'This & that' => 'This-that'
* 'myfile.tpl' => 'Myfile-tpl',
* 'øæå' => 'oeaeaa'
*
* @param string $text
* @param string $defaultText
* @param string|null $transformation
*
* @return string
*/
public function convert($text, $defaultText = "_1", $transformation = null)
{
if (!isset($transformation)) {
$transformation = $this->configuration["transformation"];
}
if (strlen($text) === 0) {
$text = $defaultText;
}
if (isset($this->configuration["transformationGroups"][$transformation]["commands"]) && !empty($this->configuration["transformationGroups"][$transformation]["commands"])) {
$text = $this->transformationProcessor->transform($text, $this->configuration["transformationGroups"][$transformation]["commands"]);
}
return $this->cleanupText($text, $this->configuration["transformationGroups"][$transformation]["cleanupMethod"]);
}
/**
* Get subquery to select relevant word IDs
*
* @uses getStopWordThresholdValue() To get threshold for words we would like to ignore in query.
*
* @param \eZ\Publish\Core\Persistence\Database\SelectQuery $query
* @param string $string
*
* @return \eZ\Publish\Core\Persistence\Database\SelectQuery
*/
protected function getWordIdSubquery( SelectQuery $query, $string )
{
$subQuery = $query->subSelect();
$tokens = $this->tokenizeString(
$this->processor->transform( $string, $this->configuration['commands'] )
);
$wordExpressions = array();
foreach ( $tokens as $token )
{
$wordExpressions[] = $this->getWordExpression( $subQuery, $token );
}
$whereCondition = $subQuery->expr->lOr( $wordExpressions );
// If stop word threshold is below 100%, make it part of $whereCondition
if ( $this->configuration['stopWordThresholdFactor'] < 1 )
{
$whereCondition = $subQuery->expr->lAnd(
$whereCondition,
$subQuery->expr->lt(
$this->dbHandler->quoteColumn( 'object_count' ),
$subQuery->bindValue( $this->getStopWordThresholdValue() )
)
);
}
$subQuery
->select( $this->dbHandler->quoteColumn( 'id' ) )
->from( $this->dbHandler->quoteTable( 'ezsearch_word' ) )
->where( $whereCondition );
return $subQuery;
}
/**
* Get subquery to select relevant word IDs
*
* @param \eZ\Publish\Core\Persistence\Database\SelectQuery $query
* @param string $string
*
* @return \eZ\Publish\Core\Persistence\Database\SelectQuery
*/
protected function getWordIdSubquery(SelectQuery $query, $string)
{
$subQuery = $query->subSelect();
$tokens = $this->tokenizeString($this->processor->transform($string, $this->configuration['commands']));
$wordExpressions = array();
foreach ($tokens as $token) {
$wordExpressions[] = $this->getWordExpression($subQuery, $token);
}
$subQuery->select($this->dbHandler->quoteColumn('id'))->from($this->dbHandler->quoteTable('ezsearch_word'))->where($subQuery->expr->lAnd($subQuery->expr->lOr($wordExpressions), $subQuery->expr->lt($this->dbHandler->quoteColumn('object_count'), $subQuery->bindValue($this->configuration['searchThresholdValue']))));
return $subQuery;
}
/**
* Build WordIDArray and update ezsearch_word table.
*
* Ported from the legacy code
*
* @see https://github.com/ezsystems/ezpublish-legacy/blob/master/kernel/search/plugins/ezsearchengine/ezsearchengine.php#L155
*
* @param array $indexArrayOnlyWords words for object to add
*
* @return array wordIDArray
*/
private function buildWordIDArray(array $indexArrayOnlyWords)
{
$wordCount = count($indexArrayOnlyWords);
$wordIDArray = [];
$wordArray = [];
// store the words in the index and remember the ID
$this->dbHandler->beginTransaction();
for ($arrayCount = 0; $arrayCount < $wordCount; $arrayCount += 500) {
// Fetch already indexed words from database
$wordArrayChuck = array_slice($indexArrayOnlyWords, $arrayCount, 500);
$wordRes = $this->searchIndex->getWords($wordArrayChuck);
// Build a has of the existing words
$wordResCount = count($wordRes);
$existingWordArray = [];
for ($i = 0; $i < $wordResCount; ++$i) {
$wordIDArray[] = $wordRes[$i]['id'];
$existingWordArray[] = $wordRes[$i]['word'];
$wordArray[$wordRes[$i]['word']] = $wordRes[$i]['id'];
}
// Update the object count of existing words by one
if (count($wordIDArray) > 0) {
$this->searchIndex->incrementWordObjectCount($wordIDArray);
}
// Insert if there is any news words
$newWordArray = array_diff($wordArrayChuck, $existingWordArray);
if (count($newWordArray) > 0) {
$this->searchIndex->addWords($newWordArray);
$newWordRes = $this->searchIndex->getWords($newWordArray);
$newWordCount = count($newWordRes);
for ($i = 0; $i < $newWordCount; ++$i) {
$wordLowercase = $this->transformationProcessor->transformByGroup($newWordRes[$i]['word'], 'lowercase');
$wordArray[$wordLowercase] = $newWordRes[$i]['id'];
}
}
}
$this->dbHandler->commit();
return $wordArray;
}
/**
* Downcases a given string using string transformation processor.
*
* @param string $string
*
* @return string
*/
protected function lowerCase($string)
{
return $this->transformationProcessor->transformByGroup($string, "lowercase");
}
/**
* Construct instance of TransformationProcessor\DefinitionBased.
*
* Through the $ruleFiles array, a list of files with full text
* transformation rules is given. These files are parsed by
* {@link \eZ\Publish\Core\Persistence\TransformationProcessor\DefinitionBased\Parser}
* and then used for normalization in the full text search.
*
* @param \eZ\Publish\Core\Persistence\TransformationProcessor\DefinitionBased\Parser $parser
* @param \eZ\Publish\Core\Persistence\TransformationProcessor\PcreCompiler $compiler
* @param array $ruleFiles
*
* @return \eZ\Publish\Core\Persistence\TransformationProcessor\DefinitionBased
*/
public function __construct(Parser $parser, PcreCompiler $compiler, array $ruleFiles = array())
{
parent::__construct($compiler, $ruleFiles);
$this->parser = $parser;
}
/**
* Constructor.
*
* @param \eZ\Publish\Core\Persistence\TransformationProcessor\PcreCompiler $compiler
* @param string $installDir Base dir for rule loading
* @param array $ruleFiles
*/
public function __construct(PcreCompiler $compiler, array $ruleFiles = array())
{
parent::__construct($compiler, $ruleFiles);
}
