Search in sources :

Example 16 with SpanMultiTermQueryWrapper

use of org.apache.lucene.search.spans.SpanMultiTermQueryWrapper in project lucene-solr by apache.

the class TestPayloadScoreQuery method testRewrite.

public void testRewrite() throws IOException {
    SpanMultiTermQueryWrapper xyz = new SpanMultiTermQueryWrapper(new WildcardQuery(new Term("field", "xyz*")));
    PayloadScoreQuery psq = new PayloadScoreQuery(xyz, new AveragePayloadFunction(), false);
    // if query wasn't rewritten properly, the query would have failed with "Rewrite first!"
    searcher.search(psq, 1);
}
Also used : WildcardQuery(org.apache.lucene.search.WildcardQuery) SpanMultiTermQueryWrapper(org.apache.lucene.search.spans.SpanMultiTermQueryWrapper) Term(org.apache.lucene.index.Term)

Example 17 with SpanMultiTermQueryWrapper

use of org.apache.lucene.search.spans.SpanMultiTermQueryWrapper in project lucene-solr by apache.

the class MultiTermHighlighting method extractAutomata.

/**
   * Extracts MultiTermQueries that match the provided field predicate.
   * Returns equivalent automata that will match terms.
   */
public static CharacterRunAutomaton[] extractAutomata(Query query, Predicate<String> fieldMatcher, boolean lookInSpan, Function<Query, Collection<Query>> preRewriteFunc) {
    // TODO Lucene needs a Query visitor API!  LUCENE-3041
    List<CharacterRunAutomaton> list = new ArrayList<>();
    Collection<Query> customSubQueries = preRewriteFunc.apply(query);
    if (customSubQueries != null) {
        for (Query sub : customSubQueries) {
            list.addAll(Arrays.asList(extractAutomata(sub, fieldMatcher, lookInSpan, preRewriteFunc)));
        }
    } else if (query instanceof BooleanQuery) {
        for (BooleanClause clause : (BooleanQuery) query) {
            if (!clause.isProhibited()) {
                list.addAll(Arrays.asList(extractAutomata(clause.getQuery(), fieldMatcher, lookInSpan, preRewriteFunc)));
            }
        }
    } else if (query instanceof ConstantScoreQuery) {
        list.addAll(Arrays.asList(extractAutomata(((ConstantScoreQuery) query).getQuery(), fieldMatcher, lookInSpan, preRewriteFunc)));
    } else if (query instanceof BoostQuery) {
        list.addAll(Arrays.asList(extractAutomata(((BoostQuery) query).getQuery(), fieldMatcher, lookInSpan, preRewriteFunc)));
    } else if (query instanceof DisjunctionMaxQuery) {
        for (Query sub : ((DisjunctionMaxQuery) query).getDisjuncts()) {
            list.addAll(Arrays.asList(extractAutomata(sub, fieldMatcher, lookInSpan, preRewriteFunc)));
        }
    } else if (lookInSpan && query instanceof SpanOrQuery) {
        for (Query sub : ((SpanOrQuery) query).getClauses()) {
            list.addAll(Arrays.asList(extractAutomata(sub, fieldMatcher, lookInSpan, preRewriteFunc)));
        }
    } else if (lookInSpan && query instanceof SpanNearQuery) {
        for (Query sub : ((SpanNearQuery) query).getClauses()) {
            list.addAll(Arrays.asList(extractAutomata(sub, fieldMatcher, lookInSpan, preRewriteFunc)));
        }
    } else if (lookInSpan && query instanceof SpanNotQuery) {
        list.addAll(Arrays.asList(extractAutomata(((SpanNotQuery) query).getInclude(), fieldMatcher, lookInSpan, preRewriteFunc)));
    } else if (lookInSpan && query instanceof SpanPositionCheckQuery) {
        list.addAll(Arrays.asList(extractAutomata(((SpanPositionCheckQuery) query).getMatch(), fieldMatcher, lookInSpan, preRewriteFunc)));
    } else if (lookInSpan && query instanceof SpanBoostQuery) {
        list.addAll(Arrays.asList(extractAutomata(((SpanBoostQuery) query).getQuery(), fieldMatcher, lookInSpan, preRewriteFunc)));
    } else if (lookInSpan && query instanceof SpanMultiTermQueryWrapper) {
        list.addAll(Arrays.asList(extractAutomata(((SpanMultiTermQueryWrapper<?>) query).getWrappedQuery(), fieldMatcher, lookInSpan, preRewriteFunc)));
    } else if (query instanceof PrefixQuery) {
        final PrefixQuery pq = (PrefixQuery) query;
        Term prefix = pq.getPrefix();
        if (fieldMatcher.test(prefix.field())) {
            list.add(new CharacterRunAutomaton(Operations.concatenate(Automata.makeString(prefix.text()), Automata.makeAnyString())) {

                @Override
                public String toString() {
                    return pq.toString();
                }
            });
        }
    } else if (query instanceof FuzzyQuery) {
        final FuzzyQuery fq = (FuzzyQuery) query;
        if (fieldMatcher.test(fq.getField())) {
            String utf16 = fq.getTerm().text();
            int[] termText = new int[utf16.codePointCount(0, utf16.length())];
            for (int cp, i = 0, j = 0; i < utf16.length(); i += Character.charCount(cp)) {
                termText[j++] = cp = utf16.codePointAt(i);
            }
            int termLength = termText.length;
            int prefixLength = Math.min(fq.getPrefixLength(), termLength);
            String suffix = UnicodeUtil.newString(termText, prefixLength, termText.length - prefixLength);
            LevenshteinAutomata builder = new LevenshteinAutomata(suffix, fq.getTranspositions());
            String prefix = UnicodeUtil.newString(termText, 0, prefixLength);
            Automaton automaton = builder.toAutomaton(fq.getMaxEdits(), prefix);
            list.add(new CharacterRunAutomaton(automaton) {

                @Override
                public String toString() {
                    return fq.toString();
                }
            });
        }
    } else if (query instanceof TermRangeQuery) {
        final TermRangeQuery tq = (TermRangeQuery) query;
        if (fieldMatcher.test(tq.getField())) {
            final CharsRef lowerBound;
            if (tq.getLowerTerm() == null) {
                lowerBound = null;
            } else {
                lowerBound = new CharsRef(tq.getLowerTerm().utf8ToString());
            }
            final CharsRef upperBound;
            if (tq.getUpperTerm() == null) {
                upperBound = null;
            } else {
                upperBound = new CharsRef(tq.getUpperTerm().utf8ToString());
            }
            final boolean includeLower = tq.includesLower();
            final boolean includeUpper = tq.includesUpper();
            final CharsRef scratch = new CharsRef();
            @SuppressWarnings("deprecation") final Comparator<CharsRef> comparator = CharsRef.getUTF16SortedAsUTF8Comparator();
            // this is *not* an automaton, but its very simple
            list.add(new CharacterRunAutomaton(Automata.makeEmpty()) {

                @Override
                public boolean run(char[] s, int offset, int length) {
                    scratch.chars = s;
                    scratch.offset = offset;
                    scratch.length = length;
                    if (lowerBound != null) {
                        int cmp = comparator.compare(scratch, lowerBound);
                        if (cmp < 0 || (!includeLower && cmp == 0)) {
                            return false;
                        }
                    }
                    if (upperBound != null) {
                        int cmp = comparator.compare(scratch, upperBound);
                        if (cmp > 0 || (!includeUpper && cmp == 0)) {
                            return false;
                        }
                    }
                    return true;
                }

                @Override
                public String toString() {
                    return tq.toString();
                }
            });
        }
    } else if (query instanceof AutomatonQuery) {
        final AutomatonQuery aq = (AutomatonQuery) query;
        if (fieldMatcher.test(aq.getField())) {
            list.add(new CharacterRunAutomaton(aq.getAutomaton()) {

                @Override
                public String toString() {
                    return aq.toString();
                }
            });
        }
    }
    return list.toArray(new CharacterRunAutomaton[list.size()]);
}
Also used : BooleanQuery(org.apache.lucene.search.BooleanQuery) Query(org.apache.lucene.search.Query) AutomatonQuery(org.apache.lucene.search.AutomatonQuery) SpanPositionCheckQuery(org.apache.lucene.search.spans.SpanPositionCheckQuery) SpanBoostQuery(org.apache.lucene.search.spans.SpanBoostQuery) SpanNotQuery(org.apache.lucene.search.spans.SpanNotQuery) ConstantScoreQuery(org.apache.lucene.search.ConstantScoreQuery) SpanNearQuery(org.apache.lucene.search.spans.SpanNearQuery) PrefixQuery(org.apache.lucene.search.PrefixQuery) FuzzyQuery(org.apache.lucene.search.FuzzyQuery) DisjunctionMaxQuery(org.apache.lucene.search.DisjunctionMaxQuery) BooleanQuery(org.apache.lucene.search.BooleanQuery) BoostQuery(org.apache.lucene.search.BoostQuery) TermRangeQuery(org.apache.lucene.search.TermRangeQuery) SpanOrQuery(org.apache.lucene.search.spans.SpanOrQuery) CharacterRunAutomaton(org.apache.lucene.util.automaton.CharacterRunAutomaton) ArrayList(java.util.ArrayList) SpanBoostQuery(org.apache.lucene.search.spans.SpanBoostQuery) BoostQuery(org.apache.lucene.search.BoostQuery) SpanNotQuery(org.apache.lucene.search.spans.SpanNotQuery) Comparator(java.util.Comparator) AutomatonQuery(org.apache.lucene.search.AutomatonQuery) ConstantScoreQuery(org.apache.lucene.search.ConstantScoreQuery) SpanMultiTermQueryWrapper(org.apache.lucene.search.spans.SpanMultiTermQueryWrapper) CharacterRunAutomaton(org.apache.lucene.util.automaton.CharacterRunAutomaton) Automaton(org.apache.lucene.util.automaton.Automaton) DisjunctionMaxQuery(org.apache.lucene.search.DisjunctionMaxQuery) LevenshteinAutomata(org.apache.lucene.util.automaton.LevenshteinAutomata) TermRangeQuery(org.apache.lucene.search.TermRangeQuery) Term(org.apache.lucene.index.Term) SpanOrQuery(org.apache.lucene.search.spans.SpanOrQuery) CharsRef(org.apache.lucene.util.CharsRef) BooleanClause(org.apache.lucene.search.BooleanClause) SpanPositionCheckQuery(org.apache.lucene.search.spans.SpanPositionCheckQuery) PrefixQuery(org.apache.lucene.search.PrefixQuery) SpanBoostQuery(org.apache.lucene.search.spans.SpanBoostQuery) FuzzyQuery(org.apache.lucene.search.FuzzyQuery) SpanNearQuery(org.apache.lucene.search.spans.SpanNearQuery)

Example 18 with SpanMultiTermQueryWrapper

use of org.apache.lucene.search.spans.SpanMultiTermQueryWrapper in project lucene-solr by apache.

the class TestUnifiedHighlighterMTQ method testSpanPositionCheck.

public void testSpanPositionCheck() throws Exception {
    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, indexAnalyzer);
    Field body = new Field("body", "", fieldType);
    Document doc = new Document();
    doc.add(body);
    body.setStringValue("This is a test.");
    iw.addDocument(doc);
    body.setStringValue("Test a one sentence document.");
    iw.addDocument(doc);
    IndexReader ir = iw.getReader();
    iw.close();
    IndexSearcher searcher = newSearcher(ir);
    UnifiedHighlighter highlighter = new UnifiedHighlighter(searcher, indexAnalyzer);
    SpanQuery childQuery = new SpanMultiTermQueryWrapper<>(new WildcardQuery(new Term("body", "te*")));
    Query query = new SpanFirstQuery(childQuery, 1000000);
    TopDocs topDocs = searcher.search(query, 10, Sort.INDEXORDER);
    assertEquals(2, topDocs.totalHits);
    String[] snippets = highlighter.highlight("body", query, topDocs);
    assertEquals(2, snippets.length);
    assertEquals("This is a <b>test</b>.", snippets[0]);
    assertEquals("<b>Test</b> a one sentence document.", snippets[1]);
    ir.close();
}
Also used : IndexSearcher(org.apache.lucene.search.IndexSearcher) WildcardQuery(org.apache.lucene.search.WildcardQuery) SpanMultiTermQueryWrapper(org.apache.lucene.search.spans.SpanMultiTermQueryWrapper) Query(org.apache.lucene.search.Query) PhraseQuery(org.apache.lucene.search.PhraseQuery) RegexpQuery(org.apache.lucene.search.RegexpQuery) SpanFirstQuery(org.apache.lucene.search.spans.SpanFirstQuery) PrefixQuery(org.apache.lucene.search.PrefixQuery) FuzzyQuery(org.apache.lucene.search.FuzzyQuery) MatchAllDocsQuery(org.apache.lucene.search.MatchAllDocsQuery) WildcardQuery(org.apache.lucene.search.WildcardQuery) DisjunctionMaxQuery(org.apache.lucene.search.DisjunctionMaxQuery) SpanTermQuery(org.apache.lucene.search.spans.SpanTermQuery) SpanBoostQuery(org.apache.lucene.search.spans.SpanBoostQuery) SpanNotQuery(org.apache.lucene.search.spans.SpanNotQuery) ConstantScoreQuery(org.apache.lucene.search.ConstantScoreQuery) SpanQuery(org.apache.lucene.search.spans.SpanQuery) SpanNearQuery(org.apache.lucene.search.spans.SpanNearQuery) TermQuery(org.apache.lucene.search.TermQuery) BooleanQuery(org.apache.lucene.search.BooleanQuery) BoostQuery(org.apache.lucene.search.BoostQuery) TermRangeQuery(org.apache.lucene.search.TermRangeQuery) SpanOrQuery(org.apache.lucene.search.spans.SpanOrQuery) SpanFirstQuery(org.apache.lucene.search.spans.SpanFirstQuery) Term(org.apache.lucene.index.Term) Document(org.apache.lucene.document.Document) SpanQuery(org.apache.lucene.search.spans.SpanQuery) TopDocs(org.apache.lucene.search.TopDocs) Field(org.apache.lucene.document.Field) IndexReader(org.apache.lucene.index.IndexReader) RandomIndexWriter(org.apache.lucene.index.RandomIndexWriter)

Example 19 with SpanMultiTermQueryWrapper

use of org.apache.lucene.search.spans.SpanMultiTermQueryWrapper in project lucene-solr by apache.

the class TestUnifiedHighlighterMTQ method testCustomSpanQueryHighlighting.

public void testCustomSpanQueryHighlighting() throws Exception {
    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, indexAnalyzer);
    Document doc = new Document();
    doc.add(new Field("body", "alpha bravo charlie delta echo foxtrot golf hotel india juliet", fieldType));
    doc.add(newTextField("id", "id", Field.Store.YES));
    iw.addDocument(doc);
    IndexReader ir = iw.getReader();
    iw.close();
    IndexSearcher searcher = newSearcher(ir);
    UnifiedHighlighter highlighter = new UnifiedHighlighter(searcher, indexAnalyzer) {

        @Override
        protected List<Query> preMultiTermQueryRewrite(Query query) {
            if (query instanceof MyWrapperSpanQuery) {
                return Collections.singletonList(((MyWrapperSpanQuery) query).originalQuery);
            }
            return null;
        }
    };
    int docId = searcher.search(new TermQuery(new Term("id", "id")), 1).scoreDocs[0].doc;
    WildcardQuery wildcardQuery = new WildcardQuery(new Term("body", "foxtr*"));
    SpanMultiTermQueryWrapper wildcardQueryWrapper = new SpanMultiTermQueryWrapper<>(wildcardQuery);
    SpanQuery wrappedQuery = new MyWrapperSpanQuery(wildcardQueryWrapper);
    BooleanQuery query = new BooleanQuery.Builder().add(wrappedQuery, BooleanClause.Occur.SHOULD).build();
    int[] docIds = new int[] { docId };
    String[] snippets = highlighter.highlightFields(new String[] { "body" }, query, docIds, new int[] { 2 }).get("body");
    assertEquals(1, snippets.length);
    assertEquals("alpha bravo charlie delta echo <b>foxtrot</b> golf hotel india juliet", snippets[0]);
    ir.close();
}
Also used : IndexSearcher(org.apache.lucene.search.IndexSearcher) SpanTermQuery(org.apache.lucene.search.spans.SpanTermQuery) TermQuery(org.apache.lucene.search.TermQuery) WildcardQuery(org.apache.lucene.search.WildcardQuery) BooleanQuery(org.apache.lucene.search.BooleanQuery) Query(org.apache.lucene.search.Query) PhraseQuery(org.apache.lucene.search.PhraseQuery) RegexpQuery(org.apache.lucene.search.RegexpQuery) SpanFirstQuery(org.apache.lucene.search.spans.SpanFirstQuery) PrefixQuery(org.apache.lucene.search.PrefixQuery) FuzzyQuery(org.apache.lucene.search.FuzzyQuery) MatchAllDocsQuery(org.apache.lucene.search.MatchAllDocsQuery) WildcardQuery(org.apache.lucene.search.WildcardQuery) DisjunctionMaxQuery(org.apache.lucene.search.DisjunctionMaxQuery) SpanTermQuery(org.apache.lucene.search.spans.SpanTermQuery) SpanBoostQuery(org.apache.lucene.search.spans.SpanBoostQuery) SpanNotQuery(org.apache.lucene.search.spans.SpanNotQuery) ConstantScoreQuery(org.apache.lucene.search.ConstantScoreQuery) SpanQuery(org.apache.lucene.search.spans.SpanQuery) SpanNearQuery(org.apache.lucene.search.spans.SpanNearQuery) TermQuery(org.apache.lucene.search.TermQuery) BooleanQuery(org.apache.lucene.search.BooleanQuery) BoostQuery(org.apache.lucene.search.BoostQuery) TermRangeQuery(org.apache.lucene.search.TermRangeQuery) SpanOrQuery(org.apache.lucene.search.spans.SpanOrQuery) SpanMultiTermQueryWrapper(org.apache.lucene.search.spans.SpanMultiTermQueryWrapper) Term(org.apache.lucene.index.Term) Document(org.apache.lucene.document.Document) SpanQuery(org.apache.lucene.search.spans.SpanQuery) Field(org.apache.lucene.document.Field) IndexReader(org.apache.lucene.index.IndexReader) RandomIndexWriter(org.apache.lucene.index.RandomIndexWriter)

Example 20 with SpanMultiTermQueryWrapper

use of org.apache.lucene.search.spans.SpanMultiTermQueryWrapper in project lucene-solr by apache.

the class TestUnifiedHighlighterMTQ method testSpanNot.

public void testSpanNot() throws Exception {
    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, indexAnalyzer);
    Field body = new Field("body", "", fieldType);
    Document doc = new Document();
    doc.add(body);
    body.setStringValue("This is a test.");
    iw.addDocument(doc);
    body.setStringValue("Test a one sentence document.");
    iw.addDocument(doc);
    IndexReader ir = iw.getReader();
    iw.close();
    IndexSearcher searcher = newSearcher(ir);
    UnifiedHighlighter highlighter = new UnifiedHighlighter(searcher, indexAnalyzer);
    SpanQuery include = new SpanMultiTermQueryWrapper<>(new WildcardQuery(new Term("body", "te*")));
    SpanQuery exclude = new SpanTermQuery(new Term("body", "bogus"));
    Query query = new SpanNotQuery(include, exclude);
    TopDocs topDocs = searcher.search(query, 10, Sort.INDEXORDER);
    assertEquals(2, topDocs.totalHits);
    String[] snippets = highlighter.highlight("body", query, topDocs);
    assertEquals(2, snippets.length);
    assertEquals("This is a <b>test</b>.", snippets[0]);
    assertEquals("<b>Test</b> a one sentence document.", snippets[1]);
    ir.close();
}
Also used : IndexSearcher(org.apache.lucene.search.IndexSearcher) WildcardQuery(org.apache.lucene.search.WildcardQuery) SpanMultiTermQueryWrapper(org.apache.lucene.search.spans.SpanMultiTermQueryWrapper) Query(org.apache.lucene.search.Query) PhraseQuery(org.apache.lucene.search.PhraseQuery) RegexpQuery(org.apache.lucene.search.RegexpQuery) SpanFirstQuery(org.apache.lucene.search.spans.SpanFirstQuery) PrefixQuery(org.apache.lucene.search.PrefixQuery) FuzzyQuery(org.apache.lucene.search.FuzzyQuery) MatchAllDocsQuery(org.apache.lucene.search.MatchAllDocsQuery) WildcardQuery(org.apache.lucene.search.WildcardQuery) DisjunctionMaxQuery(org.apache.lucene.search.DisjunctionMaxQuery) SpanTermQuery(org.apache.lucene.search.spans.SpanTermQuery) SpanBoostQuery(org.apache.lucene.search.spans.SpanBoostQuery) SpanNotQuery(org.apache.lucene.search.spans.SpanNotQuery) ConstantScoreQuery(org.apache.lucene.search.ConstantScoreQuery) SpanQuery(org.apache.lucene.search.spans.SpanQuery) SpanNearQuery(org.apache.lucene.search.spans.SpanNearQuery) TermQuery(org.apache.lucene.search.TermQuery) BooleanQuery(org.apache.lucene.search.BooleanQuery) BoostQuery(org.apache.lucene.search.BoostQuery) TermRangeQuery(org.apache.lucene.search.TermRangeQuery) SpanOrQuery(org.apache.lucene.search.spans.SpanOrQuery) Term(org.apache.lucene.index.Term) Document(org.apache.lucene.document.Document) SpanQuery(org.apache.lucene.search.spans.SpanQuery) SpanNotQuery(org.apache.lucene.search.spans.SpanNotQuery) TopDocs(org.apache.lucene.search.TopDocs) Field(org.apache.lucene.document.Field) SpanTermQuery(org.apache.lucene.search.spans.SpanTermQuery) IndexReader(org.apache.lucene.index.IndexReader) RandomIndexWriter(org.apache.lucene.index.RandomIndexWriter)

Aggregations

SpanMultiTermQueryWrapper (org.apache.lucene.search.spans.SpanMultiTermQueryWrapper)31 Term (org.apache.lucene.index.Term)27 SpanQuery (org.apache.lucene.search.spans.SpanQuery)19 WildcardQuery (org.apache.lucene.search.WildcardQuery)15 RegexpQuery (org.apache.lucene.search.RegexpQuery)14 Test (org.junit.Test)14 SpanNearQuery (org.apache.lucene.search.spans.SpanNearQuery)12 SpanTermQuery (org.apache.lucene.search.spans.SpanTermQuery)12 KrillIndex (de.ids_mannheim.korap.KrillIndex)11 SpanMultipleDistanceQuery (de.ids_mannheim.korap.query.SpanMultipleDistanceQuery)11 SpanClassQuery (de.ids_mannheim.korap.query.SpanClassQuery)10 SpanOrQuery (org.apache.lucene.search.spans.SpanOrQuery)10 Query (org.apache.lucene.search.Query)9 BooleanQuery (org.apache.lucene.search.BooleanQuery)8 BoostQuery (org.apache.lucene.search.BoostQuery)8 TermQuery (org.apache.lucene.search.TermQuery)8 SpanBoostQuery (org.apache.lucene.search.spans.SpanBoostQuery)8 ConstantScoreQuery (org.apache.lucene.search.ConstantScoreQuery)7 PrefixQuery (org.apache.lucene.search.PrefixQuery)7 TermRangeQuery (org.apache.lucene.search.TermRangeQuery)7