phrase.php

Bug tracker, and reporter.

<?php
/**
 * Zend Framework
 *
 * LICENSE
 *
 * This source file is subject to the new BSD license that is bundled
 * with this package in the file LICENSE.txt.
 * It is also available through the world-wide-web at this URL:
 * http://framework.zend.com/license/new-bsd
 * If you did not receive a copy of the license and are unable to
 * obtain it through the world-wide-web, please send an email
 * to [e-mail] so we can send you a copy immediately.
 *
 * @category   Zend
 * @package    Zend_Search_Lucene
 * @subpackage Search
 * @copyright  Copyright (c) 2005-2008 Zend Technologies USA Inc. (http://www.zend.com)
 * @license    http://framework.zend.com/license/new-bsd     New BSD License
 */


/**
 * Zend_Search_Lucene_Search_Query
 */
require_once 'Zend/Search/Lucene/Search/Query.php';

/**
 * Zend_Search_Lucene_Search_Weight_MultiTerm
 */
require_once 'Zend/Search/Lucene/Search/Weight/Phrase.php';


/**
 * A Query that matches documents containing a particular sequence of terms.
 *
 * @category   Zend
 * @package    Zend_Search_Lucene
 * @subpackage Search
 * @copyright  Copyright (c) 2005-2008 Zend Technologies USA Inc. (http://www.zend.com)
 * @license    http://framework.zend.com/license/new-bsd     New BSD License
 */
class Zend_Search_Lucene_Search_Query_Phrase extends Zend_Search_Lucene_Search_Query
{
    /**
     * Terms to find.
     * Array of Zend_Search_Lucene_Index_Term objects.
     *
     * @var array
     */
    private $_terms;

    /**
     * Term positions (relative positions of terms within the phrase).
     * Array of integers
     *
     * @var array
     */
    private $_offsets;

    /**
     * Sets the number of other words permitted between words in query phrase.
     * If zero, then this is an exact phrase search.  For larger values this works
     * like a WITHIN or NEAR operator.
     *
     * The slop is in fact an edit-distance, where the units correspond to
     * moves of terms in the query phrase out of position.  For example, to switch
     * the order of two words requires two moves (the first move places the words
     * atop one another), so to permit re-orderings of phrases, the slop must be
     * at least two.
     * More exact matches are scored higher than sloppier matches, thus search
     * results are sorted by exactness.
     *
     * The slop is zero by default, requiring exact matches.
     *
     * @var integer
     */
    private $_slop;

    /**
     * Result vector.
     *
     * @var array
     */
    private $_resVector = null;

    /**
     * Terms positions vectors.
     * Array of Arrays:
     * term1Id => (docId => array( pos1, pos2, ... ), ...)
     * term2Id => (docId => array( pos1, pos2, ... ), ...)
     *
     * @var array
     */
    private $_termsPositions = array();

    /**
     * Class constructor.  Create a new prase query.
     *
     * @param string $field    Field to search.
     * @param array  $terms    Terms to search Array of strings.
     * @param array  $offsets  Relative term positions. Array of integers.
     * @throws Zend_Search_Lucene_Exception
     */
    public function __construct($terms = null, $offsets = null, $field = null)
    {
        $this->_slop = 0;

        if (is_array($terms)) {
            $this->_terms = array();
            foreach ($terms as $termId => $termText) {
                $this->_terms[$termId] = ($field !== null)? new Zend_Search_Lucene_Index_Term($termText, $field):
                                                            new Zend_Search_Lucene_Index_Term($termText);
            }
        } else if ($terms === null) {
            $this->_terms = array();
        } else {
            throw new Zend_Search_Lucene_Exception('terms argument must be array of strings or null');
        }

        if (is_array($offsets)) {
            if (count($this->_terms) != count($offsets)) {
                throw new Zend_Search_Lucene_Exception('terms and offsets arguments must have the same size.');
            }
            $this->_offsets = $offsets;
        } else if ($offsets === null) {
            $this->_offsets = array();
            foreach ($this->_terms as $termId => $term) {
                $position = count($this->_offsets);
                $this->_offsets[$termId] = $position;
            }
        } else {
            throw new Zend_Search_Lucene_Exception('offsets argument must be array of strings or null');
        }
    }

    /**
     * Set slop
     *
     * @param integer $slop
     */
    public function setSlop($slop)
    {
        $this->_slop = $slop;
    }


    /**
     * Get slop
     *
     * @return integer
     */
    public function getSlop()
    {
        return $this->_slop;
    }


    /**
     * Adds a term to the end of the query phrase.
     * The relative position of the term is specified explicitly or the one immediately
     * after the last term added.
     *
     * @param Zend_Search_Lucene_Index_Term $term
     * @param integer $position
     */
    public function addTerm(Zend_Search_Lucene_Index_Term $term, $position = null) {
        if ((count($this->_terms) != 0)&&(end($this->_terms)->field != $term->field)) {
            throw new Zend_Search_Lucene_Exception('All phrase terms must be in the same field: ' .
                                                   $term->field . ':' . $term->text);
        }

        $this->_terms[] = $term;
        if ($position !== null) {
            $this->_offsets[] = $position;
        } else if (count($this->_offsets) != 0) {
            $this->_offsets[] = end($this->_offsets) + 1;
        } else {
            $this->_offsets[] = 0;
        }
    }


    /**
     * Re-write query into primitive queries in the context of specified index
     *
     * @param Zend_Search_Lucene_Interface $index
     * @return Zend_Search_Lucene_Search_Query
     */
    public function rewrite(Zend_Search_Lucene_Interface $index)
    {
        if (count($this->_terms) == 0) {
            return new Zend_Search_Lucene_Search_Query_Empty();
        } else if ($this->_terms[0]->field !== null) {
            return $this;
        } else {
            $query = new Zend_Search_Lucene_Search_Query_Boolean();
            $query->setBoost($this->getBoost());

            foreach ($index->getFieldNames(true) as $fieldName) {
                $subquery = new Zend_Search_Lucene_Search_Query_Phrase();
                $subquery->setSlop($this->getSlop());

                foreach ($this->_terms as $termId => $term) {
                    $qualifiedTerm = new Zend_Search_Lucene_Index_Term($term->text, $fieldName);

                    $subquery->addTerm($qualifiedTerm, $this->_offsets[$termId]);
                }

                $query->addSubquery($subquery);
            }

            return $query;
        }
    }

    /**
     * Optimize query in the context of specified index
     *
     * @param Zend_Search_Lucene_Interface $index
     * @return Zend_Search_Lucene_Search_Query
     */
    public function optimize(Zend_Search_Lucene_Interface $index)
    {
        // Check, that index contains all phrase terms
        foreach ($this->_terms as $term) {
            if (!$index->hasTerm($term)) {
                return new Zend_Search_Lucene_Search_Query_Empty();
            }
        }

        if (count($this->_terms) == 1) {
            // It's one term query
            $optimizedQuery = new Zend_Search_Lucene_Search_Query_Term(reset($this->_terms));
            $optimizedQuery->setBoost($this->getBoost());

            return $optimizedQuery;
        }

        if (count($this->_terms) == 0) {
            return new Zend_Search_Lucene_Search_Query_Empty();
        }


        return $this;
    }

    /**
     * Returns query term
     *
     * @return array
     */
    public function getTerms()
    {
        return $this->_terms;
    }


    /**
     * Set weight for specified term
     *
     * @param integer $num
     * @param Zend_Search_Lucene_Search_Weight_Term $weight
     */
    public function setWeight($num, $weight)
    {
        $this->_weights[$num] = $weight;
    }


    /**
     * Constructs an appropriate Weight implementation for this query.
     *
     * @param Zend_Search_Lucene_Interface $reader
     * @return Zend_Search_Lucene_Search_Weight
     */
    public function createWeight(Zend_Search_Lucene_Interface $reader)
    {
        $this->_weight = new Zend_Search_Lucene_Search_Weight_Phrase($this, $reader);
        return $this->_weight;
    }


    /**