List of usage examples for org.apache.lucene.queryparser.flexible.core.nodes TokenizedPhraseQueryNode add
@Override
public final void add(QueryNode child)
From source file:at.ac.univie.mminf.luceneSKOS.queryparser.flexible.standard.processors.SKOSQueryNodeProcessor.java
License:Apache License
@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(); TokenStream source;/*from ww w . java 2s .co m*/ try { source = this.analyzer.tokenStream(field, text); source.reset(); } catch (IOException e1) { throw new RuntimeException(e1); } CachingTokenFilter buffer = new CachingTokenFilter(source); PositionIncrementAttribute posIncrAtt = null; int numTokens = 0; int positionCount = 0; boolean severalTokensAtSamePosition = false; 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 } try { // rewind the buffer stream buffer.reset(); // close original stream - all tokens buffered source.close(); } catch (IOException e) { // ignore } 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)) { // no phrase query: LinkedList<QueryNode> children = new LinkedList<QueryNode>(); 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 { // phrase query: MultiPhraseQueryNode mpq = new MultiPhraseQueryNode(); List<FieldQueryNode> multiTerms = new ArrayList<FieldQueryNode>(); 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; } } return node; }
From source file:com.sindicetech.siren.qparser.keyword.processors.DatatypeAnalyzerProcessor.java
License:Open Source License
@Override protected QueryNode postProcessNode(final QueryNode node) throws QueryNodeException { if (node instanceof TextableQueryNode && !(node instanceof WildcardQueryNode) && !(node instanceof FuzzyQueryNode) && !(node instanceof RegexpQueryNode) && !(node.getParent() instanceof RangeQueryNode)) { final FieldQueryNode fieldNode = ((FieldQueryNode) node); final String field = fieldNode.getFieldAsString(); final String datatype = DatatypeProcessor.getDatatype(this.getQueryConfigHandler(), node); if (datatype == null) { return node; }//w ww . j a va2 s .co m final Analyzer analyzer = this.getAnalyzer(datatype); TokenBuffer buffer = new TokenBuffer(analyzer, fieldNode); if (!buffer.hasCharTermAttribute()) { return new NoTokenFoundQueryNode(); } switch (buffer.getNumTokens()) { case 0: return this.toEmptyQueryNode(); case 1: fieldNode.setText(buffer.getFirstTerm()); return fieldNode; default: final LinkedList<QueryNode> children = buffer.getFieldQueryNodes(field, datatype, this.isPositionIncrementsEnabled()); // Check for phrase query if (node.getParent() instanceof TokenizedPhraseQueryNode) { throw new QueryNodeException(new MessageImpl("Cannot build a MultiPhraseQuery")); } // If multiple terms at one single position, this must be a query // expansion. Perform a OR between the terms. if (buffer.hasSeveralTokensAtSamePosition() && buffer.getPositionCount() == 1) { OrQueryNode or = new OrQueryNode(children); GroupQueryNode group = new GroupQueryNode(or); // assign datatype or.setTag(DatatypeQueryNode.DATATYPE_TAGID, datatype); group.setTag(DatatypeQueryNode.DATATYPE_TAGID, datatype); return group; } // if several tokens at same position && position count > 1, then // results can be unexpected else { final TokenizedPhraseQueryNode pq = new TokenizedPhraseQueryNode(); for (int i = 0; i < children.size(); i++) { pq.add(children.get(i)); } // assign datatype pq.setTag(DatatypeQueryNode.DATATYPE_TAGID, datatype); return pq; } } } else if (node instanceof TwigQueryNode) { nbTwigs--; assert nbTwigs >= 0; } return node; }
From source file:com.sindicetech.siren.qparser.keyword.processors.PhraseQueryNodeProcessor.java
License:Open Source License
@Override protected QueryNode postProcessNode(final QueryNode node) throws QueryNodeException { if (node instanceof TextableQueryNode && !(node instanceof WildcardQueryNode) && !(node instanceof FuzzyQueryNode) && !(node instanceof RegexpQueryNode) && !(node instanceof ProtectedQueryNode) && !(node.getParent() instanceof RangeQueryNode)) { final FieldQueryNode fieldNode = ((FieldQueryNode) node); final String text = fieldNode.getTextAsString(); final String field = fieldNode.getFieldAsString(); final TokenStream source; try {/*from w ww . j a v a 2 s . c o m*/ source = this.analyzer.tokenStream(field, new StringReader(text)); source.reset(); } catch (final IOException e1) { throw new RuntimeException(e1); } final CachingTokenFilter buffer = new CachingTokenFilter(source); int numTokens = 0; try { while (buffer.incrementToken()) { numTokens++; } } catch (final IOException e) { // ignore } try { // rewind the buffer stream buffer.reset(); // close original stream - all tokens buffered source.close(); } catch (final IOException e) { // ignore } if (!buffer.hasAttribute(CharTermAttribute.class)) { return new NoTokenFoundQueryNode(); } final CharTermAttribute termAtt = buffer.getAttribute(CharTermAttribute.class); if (numTokens == 0) { return new NoTokenFoundQueryNode(); } // phrase query else if (numTokens != 1) { String datatype = (String) DatatypeProcessor.getDatatype(this.getQueryConfigHandler(), node); final TokenizedPhraseQueryNode pq = new TokenizedPhraseQueryNode(); // assign datatype pq.setTag(DatatypeQueryNode.DATATYPE_TAGID, datatype); for (int i = 0; i < numTokens; i++) { String term = null; try { final boolean hasNext = buffer.incrementToken(); assert hasNext == true; term = termAtt.toString(); } catch (final IOException e) { // safe to ignore, because we know the number of tokens } final FieldQueryNode newFieldNode = new FieldQueryNode(field, term, -1, -1); // set position increment newFieldNode.setPositionIncrement(i); // assign datatype newFieldNode.setTag(DatatypeQueryNode.DATATYPE_TAGID, datatype); pq.add(newFieldNode); } return pq; } } return node; }
From source file:org.sindice.siren.qparser.keyword.processors.DatatypeAnalyzerProcessor.java
License:Apache License
@Override protected QueryNode postProcessNode(final QueryNode node) throws QueryNodeException { if (node instanceof TextableQueryNode && !(node instanceof WildcardQueryNode) && !(node instanceof FuzzyQueryNode) && !(node instanceof RegexpQueryNode) && !(node.getParent() instanceof RangeQueryNode)) { this.positionIncrementsEnabled = false; final Boolean positionIncrementsEnabled = this.getQueryConfigHandler() .get(ConfigurationKeys.ENABLE_POSITION_INCREMENTS); if (positionIncrementsEnabled != null) { this.positionIncrementsEnabled = positionIncrementsEnabled; }/*w w w . ja va 2s . c o m*/ final FieldQueryNode fieldNode = ((FieldQueryNode) node); final String text = fieldNode.getTextAsString(); final String field = fieldNode.getFieldAsString(); final String datatype = (String) fieldNode.getTag(DatatypeQueryNode.DATATYPE_TAGID); if (datatype == null) { return node; } final Analyzer analyzer = this.getQueryConfigHandler().get(KeywordConfigurationKeys.DATATYPES_ANALYZERS) .get(datatype); if (analyzer == null) { throw new QueryNodeException(new MessageImpl(QueryParserMessages.INVALID_SYNTAX, "No analyzer associated with " + datatype)); } PositionIncrementAttribute posIncrAtt = null; int numTokens = 0; int positionCount = 0; boolean severalTokensAtSamePosition = false; final TokenStream source; try { source = analyzer.tokenStream(field, new StringReader(text)); source.reset(); } catch (final IOException e1) { throw new RuntimeException(e1); } final CachingTokenFilter buffer = new CachingTokenFilter(source); if (buffer.hasAttribute(PositionIncrementAttribute.class)) { posIncrAtt = buffer.getAttribute(PositionIncrementAttribute.class); } try { while (buffer.incrementToken()) { numTokens++; final int positionIncrement = (posIncrAtt != null) ? posIncrAtt.getPositionIncrement() : 1; if (positionIncrement != 0) { positionCount += positionIncrement; } else { severalTokensAtSamePosition = true; } } } catch (final IOException e) { // ignore } try { // rewind the buffer stream buffer.reset(); // close original stream - all tokens buffered source.close(); } catch (final IOException e) { // ignore } if (!buffer.hasAttribute(CharTermAttribute.class)) { return new NoTokenFoundQueryNode(); } final CharTermAttribute termAtt = buffer.getAttribute(CharTermAttribute.class); if (numTokens == 0) { if (nbTwigs != 0) { // Twig special case return new WildcardNodeQueryNode(); } return new NoTokenFoundQueryNode(); } else if (numTokens == 1) { String term = null; try { boolean hasNext; hasNext = buffer.incrementToken(); assert hasNext == true; term = termAtt.toString(); } catch (final IOException e) { // safe to ignore, because we know the number of tokens } fieldNode.setText(term); return fieldNode; } else { // no phrase query: final LinkedList<QueryNode> children = new LinkedList<QueryNode>(); int position = -1; for (int i = 0; i < numTokens; i++) { String term = null; final int positionIncrement = 1; try { final boolean hasNext = buffer.incrementToken(); assert hasNext == true; term = termAtt.toString(); } catch (final IOException e) { // safe to ignore, because we know the number of tokens } final FieldQueryNode newFieldNode = new FieldQueryNode(field, term, -1, -1); if (this.positionIncrementsEnabled) { position += positionIncrement; newFieldNode.setPositionIncrement(position); } else { newFieldNode.setPositionIncrement(i); } children.add(new FieldQueryNode(field, term, -1, -1)); } if (node.getParent() instanceof TokenizedPhraseQueryNode) { throw new QueryNodeException(new MessageImpl("Cannot build a MultiPhraseQuery")); } // If multiple terms at one single position, this must be a query // expansion. Perform a OR between the terms. if (severalTokensAtSamePosition && positionCount == 1) { return new GroupQueryNode(new OrQueryNode(children)); } // if several tokens at same position && position count > 1, then // results can be unexpected else { final TokenizedPhraseQueryNode pq = new TokenizedPhraseQueryNode(); for (int i = 0; i < children.size(); i++) { pq.add(children.get(i)); } return pq; } } } else if (node instanceof TwigQueryNode) { nbTwigs--; assert nbTwigs >= 0; } return node; }
From source file:org.sindice.siren.qparser.keyword.processors.PhraseQueryNodeProcessor.java
License:Apache License
@Override protected QueryNode postProcessNode(final QueryNode node) throws QueryNodeException { if (node instanceof TextableQueryNode && !(node instanceof WildcardQueryNode) && !(node instanceof FuzzyQueryNode) && !(node instanceof RegexpQueryNode) && !(node.getParent() instanceof RangeQueryNode)) { final FieldQueryNode fieldNode = ((FieldQueryNode) node); final String text = fieldNode.getTextAsString(); final String field = fieldNode.getFieldAsString(); final TokenStream source; try {//from w ww. j av a 2 s. c o m source = this.analyzer.tokenStream(field, new StringReader(text)); source.reset(); } catch (final IOException e1) { throw new RuntimeException(e1); } final CachingTokenFilter buffer = new CachingTokenFilter(source); int numTokens = 0; try { while (buffer.incrementToken()) { numTokens++; } } catch (final IOException e) { // ignore } try { // rewind the buffer stream buffer.reset(); // close original stream - all tokens buffered source.close(); } catch (final IOException e) { // ignore } if (!buffer.hasAttribute(CharTermAttribute.class)) { return new NoTokenFoundQueryNode(); } final CharTermAttribute termAtt = buffer.getAttribute(CharTermAttribute.class); if (numTokens == 0) { return new NoTokenFoundQueryNode(); } else if (numTokens != 1) { // phrase query final TokenizedPhraseQueryNode pq = new TokenizedPhraseQueryNode(); for (int i = 0; i < numTokens; i++) { String term = null; try { final boolean hasNext = buffer.incrementToken(); assert hasNext == true; term = termAtt.toString(); } catch (final IOException e) { // safe to ignore, because we know the number of tokens } final FieldQueryNode newFieldNode = new FieldQueryNode(field, term, -1, -1); newFieldNode.setPositionIncrement(i); pq.add(newFieldNode); } return pq; } } return node; }