Examples with QueryNode org.apache.lucene.queryparser.flexible.core.nodes.QueryNode used on opensource projects

Example 16 with QueryNode

use of org.apache.lucene.queryparser.flexible.core.nodes.QueryNode in project lucene-skos by behas.

the class SKOSQueryNodeProcessor method postProcessNode.

@Override
protected QueryNode postProcessNode(QueryNode node) throws QueryNodeException {
    if (node instanceof TextableQueryNode && !(node instanceof WildcardQueryNode) && !(node instanceof FuzzyQueryNode) && !(node instanceof RegexpQueryNode) && !(node.getParent() instanceof RangeQueryNode)) {
        FieldQueryNode fieldNode = ((FieldQueryNode) node);
        String text = fieldNode.getTextAsString();
        String field = fieldNode.getFieldAsString();
        CachingTokenFilter buffer = null;
        PositionIncrementAttribute posIncrAtt = null;
        int numTokens = 0;
        int positionCount = 0;
        boolean severalTokensAtSamePosition = false;
        try {
            try (TokenStream source = this.analyzer.tokenStream(field, text)) {
                buffer = new CachingTokenFilter(source);
                buffer.reset();
                if (buffer.hasAttribute(PositionIncrementAttribute.class)) {
                    posIncrAtt = buffer.getAttribute(PositionIncrementAttribute.class);
                }
                try {
                    while (buffer.incrementToken()) {
                        numTokens++;
                        int positionIncrement = (posIncrAtt != null) ? posIncrAtt.getPositionIncrement() : 1;
                        if (positionIncrement != 0) {
                            positionCount += positionIncrement;
                        } else {
                            severalTokensAtSamePosition = true;
                        }
                    }
                } catch (IOException e) {
                // ignore
                }
                // rewind the buffer stream
                //will never through on subsequent reset calls
                buffer.reset();
            } catch (IOException e) {
                throw new RuntimeException(e);
            }
            if (!buffer.hasAttribute(CharTermAttribute.class)) {
                return new NoTokenFoundQueryNode();
            }
            CharTermAttribute termAtt = buffer.getAttribute(CharTermAttribute.class);
            if (numTokens == 0) {
                return new NoTokenFoundQueryNode();
            } else if (numTokens == 1) {
                String term = null;
                try {
                    boolean hasNext;
                    hasNext = buffer.incrementToken();
                    assert hasNext == true;
                    term = termAtt.toString();
                } catch (IOException e) {
                // safe to ignore, because we know the number of tokens
                }
                fieldNode.setText(term);
                return fieldNode;
            } else if (severalTokensAtSamePosition || !(node instanceof QuotedFieldQueryNode)) {
                if (positionCount == 1 || !(node instanceof QuotedFieldQueryNode)) {
                    if (positionCount == 1) {
                        // simple case: only one position, with synonyms
                        LinkedList<QueryNode> children = new LinkedList<>();
                        for (int i = 0; i < numTokens; i++) {
                            String term = null;
                            try {
                                boolean hasNext = buffer.incrementToken();
                                assert hasNext == true;
                                term = termAtt.toString();
                            } catch (IOException e) {
                            // safe to ignore, because we know the number of tokens
                            }
                            if (buffer.hasAttribute(SKOSTypeAttribute.class) && boosts != null) {
                                SKOSTypeAttribute skosAttr = buffer.getAttribute(SKOSTypeAttribute.class);
                                children.add(new BoostQueryNode(new FieldQueryNode(field, term, -1, -1), getBoost(skosAttr.getSkosType())));
                            } else {
                                children.add(new FieldQueryNode(field, term, -1, -1));
                            }
                        }
                        return new GroupQueryNode(new StandardBooleanQueryNode(children, positionCount == 1));
                    } else {
                        // multiple positions
                        QueryNode q = new StandardBooleanQueryNode(Collections.<QueryNode>emptyList(), false);
                        QueryNode currentQuery = null;
                        for (int i = 0; i < numTokens; i++) {
                            String term = null;
                            try {
                                boolean hasNext = buffer.incrementToken();
                                assert hasNext == true;
                                term = termAtt.toString();
                            } catch (IOException e) {
                            // safe to ignore, because we know the number of tokens
                            }
                            if (posIncrAtt != null && posIncrAtt.getPositionIncrement() == 0) {
                                if (!(currentQuery instanceof BooleanQueryNode)) {
                                    QueryNode t = currentQuery;
                                    currentQuery = new StandardBooleanQueryNode(Collections.<QueryNode>emptyList(), true);
                                    ((BooleanQueryNode) currentQuery).add(t);
                                }
                                ((BooleanQueryNode) currentQuery).add(new FieldQueryNode(field, term, -1, -1));
                            } else {
                                if (currentQuery != null) {
                                    if (this.defaultOperator == Operator.OR) {
                                        q.add(currentQuery);
                                    } else {
                                        q.add(new ModifierQueryNode(currentQuery, Modifier.MOD_REQ));
                                    }
                                }
                                currentQuery = new FieldQueryNode(field, term, -1, -1);
                            }
                        }
                        if (this.defaultOperator == Operator.OR) {
                            q.add(currentQuery);
                        } else {
                            q.add(new ModifierQueryNode(currentQuery, Modifier.MOD_REQ));
                        }
                        if (q instanceof BooleanQueryNode) {
                            q = new GroupQueryNode(q);
                        }
                        return q;
                    }
                } else {
                    // phrase query:
                    MultiPhraseQueryNode mpq = new MultiPhraseQueryNode();
                    List<FieldQueryNode> multiTerms = new ArrayList<>();
                    int position = -1;
                    int i = 0;
                    int termGroupCount = 0;
                    for (; i < numTokens; i++) {
                        String term = null;
                        int positionIncrement = 1;
                        try {
                            boolean hasNext = buffer.incrementToken();
                            assert hasNext == true;
                            term = termAtt.toString();
                            if (posIncrAtt != null) {
                                positionIncrement = posIncrAtt.getPositionIncrement();
                            }
                        } catch (IOException e) {
                        // safe to ignore, because we know the number of tokens
                        }
                        if (positionIncrement > 0 && multiTerms.size() > 0) {
                            for (FieldQueryNode termNode : multiTerms) {
                                if (this.positionIncrementsEnabled) {
                                    termNode.setPositionIncrement(position);
                                } else {
                                    termNode.setPositionIncrement(termGroupCount);
                                }
                                mpq.add(termNode);
                            }
                            // Only increment once for each "group" of
                            // terms that were in the same position:
                            termGroupCount++;
                            multiTerms.clear();
                        }
                        position += positionIncrement;
                        multiTerms.add(new FieldQueryNode(field, term, -1, -1));
                    }
                    for (FieldQueryNode termNode : multiTerms) {
                        if (this.positionIncrementsEnabled) {
                            termNode.setPositionIncrement(position);
                        } else {
                            termNode.setPositionIncrement(termGroupCount);
                        }
                        mpq.add(termNode);
                    }
                    return mpq;
                }
            } else {
                TokenizedPhraseQueryNode pq = new TokenizedPhraseQueryNode();
                int position = -1;
                for (int i = 0; i < numTokens; i++) {
                    String term = null;
                    int positionIncrement = 1;
                    try {
                        boolean hasNext = buffer.incrementToken();
                        assert hasNext == true;
                        term = termAtt.toString();
                        if (posIncrAtt != null) {
                            positionIncrement = posIncrAtt.getPositionIncrement();
                        }
                    } catch (IOException e) {
                    // safe to ignore, because we know the number of tokens
                    }
                    FieldQueryNode newFieldNode = new FieldQueryNode(field, term, -1, -1);
                    if (this.positionIncrementsEnabled) {
                        position += positionIncrement;
                        newFieldNode.setPositionIncrement(position);
                    } else {
                        newFieldNode.setPositionIncrement(i);
                    }
                    pq.add(newFieldNode);
                }
                return pq;
            }
        } finally {
            if (buffer != null) {
                try {
                    buffer.close();
                } catch (IOException e) {
                // safe to ignore
                }
            }
        }
    }
    return node;
}

Example 17 with QueryNode

use of org.apache.lucene.queryparser.flexible.core.nodes.QueryNode in project lucene-solr by apache.

the class TestSpanQueryParser method getSpanQuery.

public SpanQuery getSpanQuery(String uniqueField, CharSequence query) throws QueryNodeException {
    this.spanQueryConfigHandler.set(SpansQueryConfigHandler.UNIQUE_FIELD, uniqueField);
    QueryNode queryTree = this.queryParser.parse(query, "defaultField");
    queryTree = this.spanProcessorPipeline.process(queryTree);
    return this.spansQueryTreeBuilder.build(queryTree);
}

Example 18 with QueryNode

use of org.apache.lucene.queryparser.flexible.core.nodes.QueryNode in project lucene-solr by apache.

the class TestSpanQueryParserSimpleSample method testBasicDemo.

public void testBasicDemo() throws Exception {
    SyntaxParser queryParser = new StandardSyntaxParser();
    // convert the CharSequence into a QueryNode tree
    QueryNode queryTree = queryParser.parse("body:text", null);
    // create a config handler with a attribute used in
    // UniqueFieldQueryNodeProcessor
    QueryConfigHandler spanQueryConfigHandler = new SpansQueryConfigHandler();
    spanQueryConfigHandler.set(SpansQueryConfigHandler.UNIQUE_FIELD, "index");
    // set up the processor pipeline with the ConfigHandler
    // and create the pipeline for this simple demo
    QueryNodeProcessorPipeline spanProcessorPipeline = new QueryNodeProcessorPipeline(spanQueryConfigHandler);
    // @see SpansValidatorQueryNodeProcessor
    spanProcessorPipeline.add(new SpansValidatorQueryNodeProcessor());
    // @see UniqueFieldQueryNodeProcessor
    spanProcessorPipeline.add(new UniqueFieldQueryNodeProcessor());
    // print to show out the QueryNode tree before being processed
    if (VERBOSE)
        System.out.println(queryTree);
    // Process the QueryTree using our new Processors
    queryTree = spanProcessorPipeline.process(queryTree);
    // print to show out the QueryNode tree after being processed
    if (VERBOSE)
        System.out.println(queryTree);
    // create a instance off the Builder
    SpansQueryTreeBuilder spansQueryTreeBuilder = new SpansQueryTreeBuilder();
    // convert QueryNode tree to span query Objects
    SpanQuery spanquery = spansQueryTreeBuilder.build(queryTree);
    assertTrue(spanquery instanceof SpanTermQuery);
    assertEquals(spanquery.toString(), "index:text");
}

Example 19 with QueryNode

use of org.apache.lucene.queryparser.flexible.core.nodes.QueryNode in project lucene-solr by apache.

the class StandardSyntaxParser method Clause.

public final QueryNode Clause(CharSequence field) throws ParseException {
    QueryNode q;
    Token fieldToken = null, boost = null, operator = null, term = null;
    FieldQueryNode qLower, qUpper;
    boolean lowerInclusive, upperInclusive;
    boolean group = false;
    if (jj_2_2(3)) {
        fieldToken = jj_consume_token(TERM);
        switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
            case OP_COLON:
            case OP_EQUAL:
                switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
                    case OP_COLON:
                        jj_consume_token(OP_COLON);
                        break;
                    case OP_EQUAL:
                        jj_consume_token(OP_EQUAL);
                        break;
                    default:
                        jj_la1[5] = jj_gen;
                        jj_consume_token(-1);
                        throw new ParseException();
                }
                field = EscapeQuerySyntaxImpl.discardEscapeChar(fieldToken.image);
                q = Term(field);
                break;
            case OP_LESSTHAN:
            case OP_LESSTHANEQ:
            case OP_MORETHAN:
            case OP_MORETHANEQ:
                switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
                    case OP_LESSTHAN:
                        operator = jj_consume_token(OP_LESSTHAN);
                        break;
                    case OP_LESSTHANEQ:
                        operator = jj_consume_token(OP_LESSTHANEQ);
                        break;
                    case OP_MORETHAN:
                        operator = jj_consume_token(OP_MORETHAN);
                        break;
                    case OP_MORETHANEQ:
                        operator = jj_consume_token(OP_MORETHANEQ);
                        break;
                    default:
                        jj_la1[6] = jj_gen;
                        jj_consume_token(-1);
                        throw new ParseException();
                }
                field = EscapeQuerySyntaxImpl.discardEscapeChar(fieldToken.image);
                switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
                    case TERM:
                        term = jj_consume_token(TERM);
                        break;
                    case QUOTED:
                        term = jj_consume_token(QUOTED);
                        break;
                    case NUMBER:
                        term = jj_consume_token(NUMBER);
                        break;
                    default:
                        jj_la1[7] = jj_gen;
                        jj_consume_token(-1);
                        throw new ParseException();
                }
                if (term.kind == QUOTED) {
                    term.image = term.image.substring(1, term.image.length() - 1);
                }
                switch(operator.kind) {
                    case OP_LESSTHAN:
                        lowerInclusive = true;
                        upperInclusive = false;
                        qLower = new FieldQueryNode(field, "*", term.beginColumn, term.endColumn);
                        qUpper = new FieldQueryNode(field, EscapeQuerySyntaxImpl.discardEscapeChar(term.image), term.beginColumn, term.endColumn);
                        break;
                    case OP_LESSTHANEQ:
                        lowerInclusive = true;
                        upperInclusive = true;
                        qLower = new FieldQueryNode(field, "*", term.beginColumn, term.endColumn);
                        qUpper = new FieldQueryNode(field, EscapeQuerySyntaxImpl.discardEscapeChar(term.image), term.beginColumn, term.endColumn);
                        break;
                    case OP_MORETHAN:
                        lowerInclusive = false;
                        upperInclusive = true;
                        qLower = new FieldQueryNode(field, EscapeQuerySyntaxImpl.discardEscapeChar(term.image), term.beginColumn, term.endColumn);
                        qUpper = new FieldQueryNode(field, "*", term.beginColumn, term.endColumn);
                        break;
                    case OP_MORETHANEQ:
                        lowerInclusive = true;
                        upperInclusive = true;
                        qLower = new FieldQueryNode(field, EscapeQuerySyntaxImpl.discardEscapeChar(term.image), term.beginColumn, term.endColumn);
                        qUpper = new FieldQueryNode(field, "*", term.beginColumn, term.endColumn);
                        break;
                    default:
                        {
                            if (true)
                                throw new Error("Unhandled case: operator=" + operator.toString());
                        }
                }
                q = new TermRangeQueryNode(qLower, qUpper, lowerInclusive, upperInclusive);
                break;
            default:
                jj_la1[8] = jj_gen;
                jj_consume_token(-1);
                throw new ParseException();
        }
    } else {
        switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
            case LPAREN:
            case QUOTED:
            case TERM:
            case REGEXPTERM:
            case RANGEIN_START:
            case RANGEEX_START:
            case NUMBER:
                if (jj_2_1(2)) {
                    fieldToken = jj_consume_token(TERM);
                    switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
                        case OP_COLON:
                            jj_consume_token(OP_COLON);
                            break;
                        case OP_EQUAL:
                            jj_consume_token(OP_EQUAL);
                            break;
                        default:
                            jj_la1[9] = jj_gen;
                            jj_consume_token(-1);
                            throw new ParseException();
                    }
                    field = EscapeQuerySyntaxImpl.discardEscapeChar(fieldToken.image);
                } else {
                    ;
                }
                switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
                    case QUOTED:
                    case TERM:
                    case REGEXPTERM:
                    case RANGEIN_START:
                    case RANGEEX_START:
                    case NUMBER:
                        q = Term(field);
                        break;
                    case LPAREN:
                        jj_consume_token(LPAREN);
                        q = Query(field);
                        jj_consume_token(RPAREN);
                        switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
                            case CARAT:
                                jj_consume_token(CARAT);
                                boost = jj_consume_token(NUMBER);
                                break;
                            default:
                                jj_la1[10] = jj_gen;
                                ;
                        }
                        group = true;
                        break;
                    default:
                        jj_la1[11] = jj_gen;
                        jj_consume_token(-1);
                        throw new ParseException();
                }
                break;
            default:
                jj_la1[12] = jj_gen;
                jj_consume_token(-1);
                throw new ParseException();
        }
    }
    if (boost != null) {
        float f = (float) 1.0;
        try {
            f = Float.parseFloat(boost.image);
            // avoid boosting null queries, such as those caused by stop words
            if (q != null) {
                q = new BoostQueryNode(q, f);
            }
        } catch (Exception ignored) {
        /* Should this be handled somehow? (defaults to "no boost", if
             * boost number is invalid)
             */
        }
    }
    if (group) {
        q = new GroupQueryNode(q);
    }
    {
        if (true)
            return q;
    }
    throw new Error("Missing return statement in function");
}

Example 20 with QueryNode

use of org.apache.lucene.queryparser.flexible.core.nodes.QueryNode in project lucene-solr by apache.

the class StandardSyntaxParser method Query.

// These changes were made to introduce operator precedence:
// - Clause() now returns a QueryNode. 
// - The modifiers are consumed by Clause() and returned as part of the QueryNode Object
// - Query does not consume conjunctions (AND, OR) anymore. 
// - This is now done by two new non-terminals: ConjClause and DisjClause
// The parse tree looks similar to this:
//       Query ::= DisjQuery ( DisjQuery )*
//   DisjQuery ::= ConjQuery ( OR ConjQuery )* 
//   ConjQuery ::= Clause ( AND Clause )*
//      Clause ::= [ Modifier ] ... 
public final QueryNode Query(CharSequence field) throws ParseException {
    Vector<QueryNode> clauses = null;
    QueryNode c, first = null;
    first = DisjQuery(field);
    label_1: while (true) {
        switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
            case NOT:
            case PLUS:
            case MINUS:
            case LPAREN:
            case QUOTED:
            case TERM:
            case REGEXPTERM:
            case RANGEIN_START:
            case RANGEEX_START:
            case NUMBER:
                ;
                break;
            default:
                jj_la1[2] = jj_gen;
                break label_1;
        }
        c = DisjQuery(field);
        if (clauses == null) {
            clauses = new Vector<QueryNode>();
            clauses.addElement(first);
        }
        clauses.addElement(c);
    }
    if (clauses != null) {
        {
            if (true)
                return new BooleanQueryNode(clauses);
        }
    } else {
        // the returned result drops the negation.
        if (first instanceof ModifierQueryNode) {
            ModifierQueryNode m = (ModifierQueryNode) first;
            if (m.getModifier() == ModifierQueryNode.Modifier.MOD_NOT) {
                {
                    if (true)
                        return new BooleanQueryNode(Arrays.<QueryNode>asList(m));
                }
            }
        }
        {
            if (true)
                return first;
        }
    }
    throw new Error("Missing return statement in function");
}