Examples with SpanOrQuery org.apache.lucene.search.spans.SpanOrQuery used on opensource projects

Example 26 with SpanOrQuery

use of org.apache.lucene.search.spans.SpanOrQuery 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()]);
}

Example 27 with SpanOrQuery

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

the class SpanOrBuilder method getSpanQuery.

@Override
public SpanQuery getSpanQuery(Element e) throws ParserException {
    List<SpanQuery> clausesList = new ArrayList<>();
    for (Node kid = e.getFirstChild(); kid != null; kid = kid.getNextSibling()) {
        if (kid.getNodeType() == Node.ELEMENT_NODE) {
            SpanQuery clause = factory.getSpanQuery((Element) kid);
            clausesList.add(clause);
        }
    }
    SpanQuery[] clauses = clausesList.toArray(new SpanQuery[clausesList.size()]);
    SpanOrQuery soq = new SpanOrQuery(clauses);
    float boost = DOMUtils.getAttribute(e, "boost", 1.0f);
    return new SpanBoostQuery(soq, boost);
}

Example 28 with SpanOrQuery

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

the class SpanOrTermsBuilder method getSpanQuery.

@Override
public SpanQuery getSpanQuery(Element e) throws ParserException {
    String fieldName = DOMUtils.getAttributeWithInheritanceOrFail(e, "fieldName");
    String value = DOMUtils.getNonBlankTextOrFail(e);
    List<SpanQuery> clausesList = new ArrayList<>();
    try (TokenStream ts = analyzer.tokenStream(fieldName, value)) {
        TermToBytesRefAttribute termAtt = ts.addAttribute(TermToBytesRefAttribute.class);
        ts.reset();
        while (ts.incrementToken()) {
            SpanTermQuery stq = new SpanTermQuery(new Term(fieldName, BytesRef.deepCopyOf(termAtt.getBytesRef())));
            clausesList.add(stq);
        }
        ts.end();
        SpanOrQuery soq = new SpanOrQuery(clausesList.toArray(new SpanQuery[clausesList.size()]));
        float boost = DOMUtils.getAttribute(e, "boost", 1.0f);
        return new SpanBoostQuery(soq, boost);
    } catch (IOException ioe) {
        throw new ParserException("IOException parsing value:" + value);
    }
}

Example 29 with SpanOrQuery

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

the class TestQueryBuilder method testMultiWordSynonymsPhrase.

/** forms graph query */
public void testMultiWordSynonymsPhrase() throws Exception {
    SpanNearQuery expectedNear = SpanNearQuery.newOrderedNearQuery("field").addClause(new SpanTermQuery(new Term("field", "guinea"))).addClause(new SpanTermQuery(new Term("field", "pig"))).setSlop(0).build();
    SpanTermQuery expectedTerm = new SpanTermQuery(new Term("field", "cavy"));
    QueryBuilder queryBuilder = new QueryBuilder(new MockSynonymAnalyzer());
    assertEquals(new SpanOrQuery(new SpanQuery[] { expectedNear, expectedTerm }), queryBuilder.createPhraseQuery("field", "guinea pig"));
}

Example 30 with SpanOrQuery

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

the class TestMemoryIndexAgainstRAMDir method testPassesIfWrapped.

// LUCENE-3831
public void testPassesIfWrapped() throws IOException {
    RegexpQuery regex = new RegexpQuery(new Term("field", "worl."));
    SpanQuery wrappedquery = new SpanOrQuery(new SpanMultiTermQueryWrapper<>(regex));
    MemoryIndex mindex = randomMemoryIndex();
    mindex.addField("field", new MockAnalyzer(random()).tokenStream("field", "hello there"));
    // This passes though
    assertEquals(0, mindex.search(wrappedquery), 0.00001f);
    TestUtil.checkReader(mindex.createSearcher().getIndexReader());
}