Example usage for org.apache.lucene.queryparser.classic MultiFieldQueryParser MultiFieldQueryParser

Introduction

In this page you can find the example usage for org.apache.lucene.queryparser.classic MultiFieldQueryParser MultiFieldQueryParser.

Prototype

public MultiFieldQueryParser(String[] fields, Analyzer analyzer) 

Source Link

Document

Creates a MultiFieldQueryParser.

Usage

From source file:de.anycook.db.lucene.FulltextIndex.java

License:Open Source License

public Set<String> search(String q) throws IOException {
    Set<String> recipes = new LinkedHashSet<>();
    String fields[] = new String[] { "description", "steps" };
    logger.debug(String.format("searching for %s", q));

    try (IndexReader reader = DirectoryReader.open(index)) {
        int hitsPerPage = 1000;
        IndexSearcher searcher = new IndexSearcher(reader);
        Query query = new MultiFieldQueryParser(fields, analyzer).parse(q);
        TopScoreDocCollector collector = TopScoreDocCollector.create(hitsPerPage, null);
        searcher.search(query, collector);

        ScoreDoc[] hits = collector.topDocs().scoreDocs;
        for (ScoreDoc hit : hits) {
            Document d = searcher.doc(hit.doc);
            recipes.add(d.get("title"));
        }/*w  w  w  .  jav  a  2  s  .  c  om*/

    } catch (CorruptIndexException | ParseException e) {
        logger.error(e);
    }

    logger.debug(String.format("found %d results", recipes.size()));
    return recipes;
}

---

From source file:de.elbe5.cms.search.SearchBean.java

License:Open Source License

public void searchContent(ContentSearchResultData result) {
    result.getResults().clear();//from  w  w w  .  ja v a  2 s .  c  om
    String[] fieldNames = result.getFieldNames();
    ScoreDoc[] hits = null;
    float maxScore = 0f;
    try {
        String indexPath = ApplicationPath.getAppPath() + "contentindex";
        ensureDirectory(indexPath);
        IndexReader reader = DirectoryReader.open(FSDirectory.open(Paths.get(indexPath)));
        IndexSearcher searcher = new IndexSearcher(reader);
        Analyzer analyzer = new StandardAnalyzer();
        MultiFieldQueryParser parser = new MultiFieldQueryParser(fieldNames, analyzer);
        String pattern = result.getPattern();
        pattern = pattern.trim();
        Query query = null;
        if (pattern.length() != 0) {
            query = parser.parse(pattern);
            //Log.log("Searching for: " + query.toString());
            TopDocs topDocs = searcher.search(query, result.getMaxSearchResults());
            hits = topDocs.scoreDocs;
            maxScore = topDocs.getMaxScore();
        }
        if (hits != null) {
            for (ScoreDoc hit : hits) {
                Document doc = searcher.doc(hit.doc);
                ContentSearchData data = null;
                String type = doc.get("type");
                switch (type) {
                case SiteSearchData.TYPE:
                    data = new SiteSearchData();
                    break;
                case PageSearchData.TYPE:
                    data = new PageSearchData();
                    break;
                case FileSearchData.TYPE:
                    data = new FileSearchData();
                    break;
                }
                assert (data != null);
                data.setDoc(doc);
                data.setScore(maxScore <= 1f ? hit.score : hit.score / maxScore);
                data.evaluateDoc();
                data.setContexts(query, analyzer);
                result.getResults().add(data);
            }
        }
        reader.close();
    } catch (Exception ignore) {
    }
}

---

From source file:de.elbe5.cms.search.SearchBean.java

License:Open Source License

public void searchUsers(UserSearchResultData result) {
    result.getResults().clear();/*from   www.j  a  va2 s .co  m*/
    String[] fieldNames = result.getFieldNames();
    ScoreDoc[] hits = null;
    float maxScore = 0f;
    try {
        String indexPath = ApplicationPath.getAppPath() + "userindex";
        ensureDirectory(indexPath);
        IndexReader reader = DirectoryReader.open(FSDirectory.open(Paths.get(indexPath)));
        IndexSearcher searcher = new IndexSearcher(reader);
        Analyzer analyzer = new StandardAnalyzer();
        MultiFieldQueryParser parser = new MultiFieldQueryParser(fieldNames, analyzer);
        String pattern = result.getPattern();
        pattern = pattern.trim();
        Query query = null;
        if (pattern.length() != 0) {
            query = parser.parse(pattern);
            //Log.log("Searching for: " + query.toString());
            TopDocs topDocs = searcher.search(query, result.getMaxSearchResults());
            hits = topDocs.scoreDocs;
            maxScore = topDocs.getMaxScore();
        }
        if (hits != null) {
            for (ScoreDoc hit : hits) {
                Document doc = searcher.doc(hit.doc);
                UserSearchData data = new UserSearchData();
                data.setDoc(doc);
                data.setScore(maxScore <= 1f ? hit.score : hit.score / maxScore);
                data.evaluateDoc();
                data.setContexts(query, analyzer);
                result.getResults().add(data);
            }
        }
        reader.close();
    } catch (Exception ignore) {
    }
}

---

From source file:de.unihildesheim.iw.lucene.query.TryExactTermsQuery.java

License:Open Source License

/**
 * New instance using the supplied query.
 *
 * @param analyzer Query analyzer/*from w w  w .  ja v a 2 s . c  o  m*/
 * @param queryStr Query string
 * @param fields Fields to query
 * @throws ParseException Thrown, if the query could not be parsed
 */
public TryExactTermsQuery(@NotNull final Analyzer analyzer, @NotNull final String queryStr,
        @NotNull final String... fields) throws ParseException {
    if (fields.length == 0) {
        throw new IllegalArgumentException("Empty fields list.");
    }
    if (StringUtils.isStrippedEmpty(queryStr)) {
        throw new IllegalArgumentException("Empty query.");
    }

    this.queryTerms = QueryUtils.tokenizeQueryString(queryStr, analyzer);

    final QueryParser qParser = new MultiFieldQueryParser(fields, analyzer);

    this.query = new BooleanQuery();
    this.uniqueQueryTerms = new HashSet<>(this.queryTerms);
    for (final String term : this.uniqueQueryTerms) {
        @SuppressWarnings("ObjectAllocationInLoop")
        final BooleanClause bc = new BooleanClause(qParser.parse(QueryParserBase.escape(term)), Occur.SHOULD);
        this.query.add(bc);
    }
    this.query.setMinimumNumberShouldMatch(this.uniqueQueryTerms.size());
    if (LOG.isDebugEnabled()) {
        LOG.debug("TEQ {} uQt={}", this.query, this.uniqueQueryTerms);
    }
}

---

From source file:edu.cmu.lti.oaqa.baseqa.concept.rerank.scorers.LuceneConceptScorer.java

License:Apache License

@Override
public boolean initialize(ResourceSpecifier aSpecifier, Map<String, Object> aAdditionalParams)
        throws ResourceInitializationException {
    super.initialize(aSpecifier, aAdditionalParams);
    hits = Integer.class.cast(getParameterValue("hits"));
    // query constructor
    String stoplistPath = String.class.cast(getParameterValue("stoplist-path"));
    try {/*  www.j  av  a  2 s .  c  o m*/
        stoplist = Resources.readLines(getClass().getResource(stoplistPath), UTF_8).stream().map(String::trim)
                .collect(toSet());
    } catch (IOException e) {
        throw new ResourceInitializationException(e);
    }
    // load index parameters
    idFieldName = String.class.cast(getParameterValue("id-field"));
    sourceFieldName = String.class.cast(getParameterValue("source-field"));
    //noinspection unchecked
    fields = Iterables.toArray((Iterable<String>) getParameterValue("fields"), String.class);
    String uriPrefixPath = String.class.cast(getParameterValue("uri-prefix"));
    try {
        uriPrefix = Resources.readLines(getClass().getResource(uriPrefixPath), UTF_8).stream()
                .map(line -> line.split("\t")).collect(toMap(segs -> segs[0], segs -> segs[1]));
    } catch (IOException e) {
        throw new ResourceInitializationException(e);
    }
    String index = String.class.cast(getParameterValue("index"));
    // create lucene
    Analyzer analyzer = new StandardAnalyzer();
    parser = new MultiFieldQueryParser(fields, analyzer);
    try {
        reader = DirectoryReader.open(FSDirectory.open(Paths.get(index)));
    } catch (IOException e) {
        throw new ResourceInitializationException(e);
    }
    searcher = new IndexSearcher(reader);
    return true;
}

---

From source file:edu.cmu.lti.oaqa.baseqa.concept.retrieval.LuceneConceptRetrievalExecutor.java

License:Apache License

@Override
public void initialize(UimaContext context) throws ResourceInitializationException {
    super.initialize(context);
    hits = UimaContextHelper.getConfigParameterIntValue(context, "hits", 100);
    // query constructor
    constructor = UimaContextHelper.createObjectFromConfigParameter(context, "query-string-constructor",
            "query-string-constructor-params", BooleanBagOfPhraseQueryStringConstructor.class,
            QueryStringConstructor.class);
    // lucene//w  w w .j a v  a  2s.com
    Analyzer analyzer = UimaContextHelper.createObjectFromConfigParameter(context, "query-analyzer",
            "query-analyzer-params", StandardAnalyzer.class, Analyzer.class);
    String[] fields = UimaContextHelper.getConfigParameterStringArrayValue(context, "fields");
    parser = new MultiFieldQueryParser(fields, analyzer);
    String index = UimaContextHelper.getConfigParameterStringValue(context, "index");
    try {
        reader = DirectoryReader.open(FSDirectory.open(Paths.get(index)));
    } catch (IOException e) {
        throw new ResourceInitializationException(e);
    }
    searcher = new IndexSearcher(reader);
    idFieldName = UimaContextHelper.getConfigParameterStringValue(context, "id-field", null);
    nameFieldName = UimaContextHelper.getConfigParameterStringValue(context, "name-field", null);
    sourceFieldName = UimaContextHelper.getConfigParameterStringValue(context, "source-field", null);
    String uriPrefixPath = UimaContextHelper.getConfigParameterStringValue(context, "uri-prefix");
    try {
        uriPrefix = Resources.readLines(getClass().getResource(uriPrefixPath), UTF_8).stream()
                .map(line -> line.split("\t")).collect(toMap(segs -> segs[0], segs -> segs[1]));
    } catch (IOException e) {
        throw new ResourceInitializationException(e);
    }
}

---

From source file:edu.cmu.lti.oaqa.baseqa.document.rerank.scorers.LuceneDocumentScorer.java

License:Apache License

private void search(String queryString, String conf) throws RuntimeException {
    if (queryString.trim().isEmpty())
        return;/*from w  ww.  j  a v a2  s. c o m*/
    ScoreDoc[] results;
    try {
        QueryParser parser = new MultiFieldQueryParser(fields, analyzer);
        Query query = parser.parse(queryString);
        results = searcher.search(query, hits).scoreDocs;
    } catch (ParseException | IOException e) {
        throw new RuntimeException(e);
    }
    for (int i = 0; i < results.length; i++) {
        try {
            int doc = results[i].doc;
            String uri = uriPrefix + reader.document(doc).get(idFieldName);
            if (!uri2conf2rank.contains(uri, conf) || uri2conf2rank.get(uri, conf) > i) {
                synchronizedPut(uri2conf2rank, uri, conf, i);
            }
            double score = results[i].score;
            if (!uri2conf2score.contains(uri, conf) || uri2conf2score.get(uri, conf) < score) {
                synchronizedPut(uri2conf2score, uri, conf, score);
            }
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
    }
}

---

From source file:edu.cmu.lti.oaqa.baseqa.document.retrieval.LuceneDocumentRetrievalExecutor.java

License:Apache License

@Override
public void initialize(UimaContext context) throws ResourceInitializationException {
    super.initialize(context);
    hits = UimaContextHelper.getConfigParameterIntValue(context, "hits", 100);
    // query constructor
    constructor = UimaContextHelper.createObjectFromConfigParameter(context, "query-string-constructor",
            "query-string-constructor-params", BooleanBagOfPhraseQueryStringConstructor.class,
            QueryStringConstructor.class);
    // lucene//from  w w  w  .j av  a  2s . c o  m
    Analyzer analyzer = UimaContextHelper.createObjectFromConfigParameter(context, "query-analyzer",
            "query-analyzer-params", StandardAnalyzer.class, Analyzer.class);
    String[] fields = UimaContextHelper.getConfigParameterStringArrayValue(context, "fields");
    parser = new MultiFieldQueryParser(fields, analyzer);
    String index = UimaContextHelper.getConfigParameterStringValue(context, "index");
    try {
        reader = DirectoryReader.open(FSDirectory.open(Paths.get(index)));
    } catch (IOException e) {
        throw new ResourceInitializationException(e);
    }
    searcher = new IndexSearcher(reader);
    idFieldName = UimaContextHelper.getConfigParameterStringValue(context, "id-field", null);
    titleFieldName = UimaContextHelper.getConfigParameterStringValue(context, "title-field", null);
    textFieldName = UimaContextHelper.getConfigParameterStringValue(context, "text-field", null);
    uriPrefix = UimaContextHelper.getConfigParameterStringValue(context, "uri-prefix", null);
}

---

From source file:edu.usc.ir.geo.gazetteer.GeoNameResolver.java

License:Apache License

private HashMap<String, List<Location>> resolveEntities(List<String> locationNames, int count,
        IndexReader reader) throws IOException {
    if (locationNames.size() >= 200)
        hitsPerPage = 5; // avoid heavy computation
    IndexSearcher searcher = new IndexSearcher(reader);
    Query q = null;//from  w ww . j  a  v a2s .  com

    HashMap<String, List<Location>> allCandidates = new HashMap<String, List<Location>>();

    for (String name : locationNames) {

        if (!allCandidates.containsKey(name)) {
            try {
                //query is wrapped in additional quotes (") to avoid query tokenization on space
                q = new MultiFieldQueryParser(new String[] { FIELD_NAME_NAME, FIELD_NAME_ALTERNATE_NAMES },
                        analyzer).parse(String.format("\"%s\"", name));

                //sort descending on population
                SortField populationSort = new SortedNumericSortField(FIELD_NAME_POPULATION,
                        SortField.Type.LONG, true);

                Sort sort = new Sort(populationSort);
                //Fetch 3 times desired values, these will be sorted on code and only desired number will be kept
                ScoreDoc[] hits = searcher.search(q, hitsPerPage * 3, sort).scoreDocs;

                List<Location> topHits = new ArrayList<Location>();

                for (int i = 0; i < hits.length; ++i) {
                    Location tmpLocObj = new Location();

                    int docId = hits[i].doc;
                    Document d;
                    try {
                        d = searcher.doc(docId);
                        tmpLocObj.setName(d.get(FIELD_NAME_NAME));
                        tmpLocObj.setLongitude(d.get(FIELD_NAME_LONGITUDE));
                        tmpLocObj.setLatitude(d.get(FIELD_NAME_LATITUDE));
                        //If alternate names are empty put name as actual name
                        //This covers missing data and equals weight for later computation
                        if (d.get(FIELD_NAME_ALTERNATE_NAMES).isEmpty()) {
                            tmpLocObj.setAlternateNames(d.get(FIELD_NAME_NAME));
                        } else {
                            tmpLocObj.setAlternateNames(d.get(FIELD_NAME_ALTERNATE_NAMES));
                        }
                        tmpLocObj.setCountryCode(d.get(FIELD_NAME_COUNTRY_CODE));
                        tmpLocObj.setAdmin1Code(d.get(FIELD_NAME_ADMIN1_CODE));
                        tmpLocObj.setAdmin2Code(d.get(FIELD_NAME_ADMIN2_CODE));
                        tmpLocObj.setFeatureCode(d.get(FIELD_NAME_FEATURE_CODE));

                    } catch (IOException e) {
                        e.printStackTrace();
                    }
                    topHits.add(tmpLocObj);
                }
                //Picking hitsPerPage number of locations from feature code sorted list 
                allCandidates.put(name, pickTopSortedByCode(topHits, hitsPerPage));
            } catch (org.apache.lucene.queryparser.classic.ParseException e) {
                e.printStackTrace();
            }
        }
    }

    HashMap<String, List<Location>> resolvedEntities = new HashMap<String, List<Location>>();
    pickBestCandidates(resolvedEntities, allCandidates, count);
    return resolvedEntities;
}

---

From source file:eu.eexcess.wikipedialocal.recommender.PartnerConnector.java

License:Apache License

@Override
public ResultList queryPartnerNative(PartnerConfiguration partnerConfiguration, SecureUserProfile userProfile,
        PartnerdataLogger dataLogger) throws IOException {
    partnerConfig = partnerConfiguration;
    ResultList resultList = new ResultList();

    Analyzer analyzer = new EnglishAnalyzer();
    File directoryPath = new File(PartnerConfigurationEnum.CONFIG.getPartnerConfiguration().searchEndpoint);
    Directory directory = FSDirectory.open(directoryPath);
    IndexReader indexReader = IndexReader.open(directory);
    IndexSearcher indexSearcher = new IndexSearcher(indexReader);
    QueryParser queryParser = new MultiFieldQueryParser(FIELD_CONTENTS, analyzer);
    queryParser.setDefaultOperator(Operator.AND);
    String queryString = PartnerConfigurationEnum.CONFIG.getQueryGenerator().toQuery(userProfile);
    Query query = null;//from w  ww .  ja v a 2s.  c  o m
    try {
        query = queryParser.parse(queryString);
    } catch (ParseException e) {

        //logger.log(Level.SEVERE, "could not parse input query", e);
    }
    if (userProfile.numResults == null)
        userProfile.numResults = 10;
    TopDocs topDocs = indexSearcher.search(query, userProfile.numResults);
    for (ScoreDoc sDocs : topDocs.scoreDocs) {
        Result result = new Result();
        org.apache.lucene.document.Document doc = indexSearcher.doc(sDocs.doc);
        if (doc != null) {
            IndexableField title = doc.getField("title");
            IndexableField sectionTitle = doc.getField("sectionTitle");
            IndexableField category = doc.getField("category");
            IndexableField sectionText = doc.getField("sectionText");
            if (sectionText != null)
                result.description = sectionText.stringValue();
            if (category != null) {
                result.facets.type = category.stringValue();
                result.facets.provider = "Category";
            }
            if (title != null && sectionTitle != null)
                result.title = title.stringValue() + " - " + sectionTitle.stringValue();
            resultList.results.add(result);
        }
    }

    resultList.totalResults = topDocs.totalHits;
    return resultList;
}

---

• «
• 1
• 2
• 3
• 4
• 5
• 6
• »