/**
* Query specific matches highlighting
*
* @param Highlighter $highlighter Highlighter object (also contains doc for highlighting)
*/
protected function _highlightMatches(Highlighter $highlighter)
{
$words = array();
$prefix = Index\Term::getPrefix($this->_term->text, $this->_prefixLength);
$prefixByteLength = strlen($prefix);
$prefixUtf8Length = Index\Term::getLength($prefix);
$termLength = Index\Term::getLength($this->_term->text);
$termRest = substr($this->_term->text, $prefixByteLength);
// we calculate length of the rest in bytes since levenshtein() is not UTF-8 compatible
$termRestLength = strlen($termRest);
$scaleFactor = 1 / (1 - $this->_minimumSimilarity);
$docBody = $highlighter->getDocument()->getFieldUtf8Value('body');
$tokens = Lucene\Analysis\Analyzer\Analyzer::getDefault()->tokenize($docBody, 'UTF-8');
foreach ($tokens as $token) {
$termText = $token->getTermText();
if (substr($termText, 0, $prefixByteLength) == $prefix) {
// Calculate similarity
$target = substr($termText, $prefixByteLength);
$maxDistance = isset($this->_maxDistances[strlen($target)]) ? $this->_maxDistances[strlen($target)] : $this->_calculateMaxDistance($prefixUtf8Length, $termRestLength, strlen($target));
if ($termRestLength == 0) {
// we don't have anything to compare. That means if we just add
// the letters for current term we get the new word
$similarity = $prefixUtf8Length == 0 ? 0 : 1 - strlen($target) / $prefixUtf8Length;
} elseif (strlen($target) == 0) {
$similarity = $prefixUtf8Length == 0 ? 0 : 1 - $termRestLength / $prefixUtf8Length;
} elseif ($maxDistance < abs($termRestLength - strlen($target))) {
//just adding the characters of term to target or vice-versa results in too many edits
//for example "pre" length is 3 and "prefixes" length is 8. We can see that
//given this optimal circumstance, the edit distance cannot be less than 5.
//which is 8-3 or more precisesly abs(3-8).
//if our maximum edit distance is 4, then we can discard this word
//without looking at it.
$similarity = 0;
} else {
$similarity = 1 - levenshtein($termRest, $target) / ($prefixUtf8Length + min($termRestLength, strlen($target)));
}
if ($similarity > $this->_minimumSimilarity) {
$words[] = $termText;
}
}
}
$highlighter->highlight($words);
}
/**
* Query specific matches highlighting
*
* @param Highlighter $highlighter Highlighter object (also contains doc for highlighting)
*/
protected function _highlightMatches(Highlighter $highlighter)
{
$words = array();
if ($this->_signs === null) {
foreach ($this->_terms as $term) {
$words[] = $term->text;
}
} else {
foreach ($this->_signs as $id => $sign) {
if ($sign !== false) {
$words[] = $this->_terms[$id]->text;
}
}
}
$highlighter->highlight($words);
}
/**
* Query specific matches highlighting
*
* @param Highlighter $highlighter Highlighter object (also contains doc for highlighting)
*/
protected function _highlightMatches(Highlighter $highlighter)
{
$words = array();
$matchExpression = '/^' . str_replace(array('\\?', '\\*'), array('.', '.*'), preg_quote($this->_pattern->text, '/')) . '$/';
ErrorHandler::start(E_WARNING);
$result = preg_match('/\\pL/u', 'a');
ErrorHandler::stop();
if ($result == 1) {
// PCRE unicode support is turned on
// add Unicode modifier to the match expression
$matchExpression .= 'u';
}
$docBody = $highlighter->getDocument()->getFieldUtf8Value('body');
$tokens = Analyzer::getDefault()->tokenize($docBody, 'UTF-8');
foreach ($tokens as $token) {
if (preg_match($matchExpression, $token->getTermText()) === 1) {
$words[] = $token->getTermText();
}
}
$highlighter->highlight($words);
}
/**
* Query specific matches highlighting
*
* @param Highlighter $highlighter Highlighter object (also contains doc for highlighting)
*/
protected function _highlightMatches(Highlighter $highlighter)
{
$words = array();
$docBody = $highlighter->getDocument()->getFieldUtf8Value('body');
$tokens = Lucene\Analysis\Analyzer\Analyzer::getDefault()->tokenize($docBody, 'UTF-8');
$lowerTermText = $this->_lowerTerm !== null ? $this->_lowerTerm->text : null;
$upperTermText = $this->_upperTerm !== null ? $this->_upperTerm->text : null;
if ($this->_inclusive) {
foreach ($tokens as $token) {
$termText = $token->getTermText();
if (($lowerTermText == null || $lowerTermText <= $termText) && ($upperTermText == null || $termText <= $upperTermText)) {
$words[] = $termText;
}
}
} else {
foreach ($tokens as $token) {
$termText = $token->getTermText();
if (($lowerTermText == null || $lowerTermText < $termText) && ($upperTermText == null || $termText < $upperTermText)) {
$words[] = $termText;
}
}
}
$highlighter->highlight($words);
}
/**
* Query specific matches highlighting
*
* @param Highlighter $highlighter Highlighter object (also contains doc for highlighting)
*/
protected function _highlightMatches(Highlighter $highlighter)
{
$highlighter->highlight($this->_term->text);
}
/**
* Query specific matches highlighting
*
* @param Highlighter $highlighter Highlighter object (also contains doc for highlighting)
*/
protected function _highlightMatches(Highlighter $highlighter)
{
$words = array();
foreach ($this->_terms as $term) {
$words[] = $term->text;
}
$highlighter->highlight($words);
}
/**
* Query specific matches highlighting
*
* @param Highlighter $highlighter Highlighter object (also contains doc for highlighting)
*/
protected function _highlightMatches(Highlighter $highlighter)
{
/** Skip fields detection. We don't need it, since we expect all fields presented in the HTML body and don't differentiate them */
/** Skip exact term matching recognition, keyword fields highlighting is not supported */
/** Skip wildcard queries recognition. Supported wildcards are removed by text analyzer */
// tokenize phrase using current analyzer and process it as a phrase query
$tokens = Analyzer::getDefault()->tokenize($this->_phrase, $this->_phraseEncoding);
if (count($tokens) == 0) {
// Do nothing
return;
}
if (count($tokens) == 1) {
$highlighter->highlight($tokens[0]->getTermText());
return;
}
//It's non-trivial phrase query
$words = array();
foreach ($tokens as $token) {
$words[] = $token->getTermText();
}
$highlighter->highlight($words);
}
