001/*
002 * Copyright (c) 2007-2016 Concurrent, Inc. All Rights Reserved.
003 *
004 * Project and contact information: http://www.cascading.org/
005 *
006 * This file is part of the Cascading project.
007 *
008 * Licensed under the Apache License, Version 2.0 (the "License");
009 * you may not use this file except in compliance with the License.
010 * You may obtain a copy of the License at
011 *
012 *     http://www.apache.org/licenses/LICENSE-2.0
013 *
014 * Unless required by applicable law or agreed to in writing, software
015 * distributed under the License is distributed on an "AS IS" BASIS,
016 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
017 * See the License for the specific language governing permissions and
018 * limitations under the License.
019 */
020
021package cascading.operation.regex;
022
023import java.beans.ConstructorProperties;
024import java.util.regex.Pattern;
025
026import cascading.flow.FlowProcess;
027import cascading.operation.Function;
028import cascading.operation.FunctionCall;
029import cascading.operation.OperationCall;
030import cascading.tuple.Fields;
031import cascading.tuple.Tuple;
032import cascading.util.Pair;
033
034/**
035 * Class RegexGenerator will emit a new Tuple for every split on the incoming argument value delimited by the given patternString.
036 * <p/>
037 * RegexGenerator only expects one field value. If more than one argument value is passed, only the
038 * first is handled, the remainder are ignored.
039 * <p/>
040 * This could be used to break a document into single word tuples for later processing for a word count.
041 * <p/>
042 * Note a {@code null} valued argument passed to the parser will be converted to an empty string ({@code ""}) before
043 * the regex is applied.
044 * <p/>
045 * Any Object value will be coerced to a String type if type information is provided. See the
046 * {@link cascading.tuple.type.CoercibleType} interface to control how custom Object types are converted to String
047 * values.
048 */
049public class RegexSplitGenerator extends RegexOperation<Pair<Pattern, Tuple>> implements Function<Pair<Pattern, Tuple>>
050  {
051  /**
052   * Constructor RegexGenerator creates a new RegexGenerator instance.
053   *
054   * @param patternString of type String
055   */
056  @ConstructorProperties({"patternString"})
057  public RegexSplitGenerator( String patternString )
058    {
059    super( 1, Fields.size( 1 ), patternString );
060    }
061
062  /**
063   * Constructor RegexGenerator creates a new RegexGenerator instance.
064   *
065   * @param fieldDeclaration of type Fields
066   * @param patternString    of type String
067   */
068  @ConstructorProperties({"fieldDeclaration", "patternString"})
069  public RegexSplitGenerator( Fields fieldDeclaration, String patternString )
070    {
071    super( 1, fieldDeclaration, patternString );
072
073    if( fieldDeclaration.size() != 1 )
074      throw new IllegalArgumentException( "fieldDeclaration may only declare one field, was " + fieldDeclaration.print() );
075    }
076
077  @Override
078  public void prepare( FlowProcess flowProcess, OperationCall<Pair<Pattern, Tuple>> operationCall )
079    {
080    operationCall.setContext( new Pair<Pattern, Tuple>( getPattern(), Tuple.size( 1 ) ) );
081    }
082
083  @Override
084  public void operate( FlowProcess flowProcess, FunctionCall<Pair<Pattern, Tuple>> functionCall )
085    {
086    String value = functionCall.getArguments().getString( 0 );
087
088    if( value == null )
089      value = "";
090
091    String[] split = functionCall.getContext().getLhs().split( value );
092
093    for( String string : split )
094      {
095      functionCall.getContext().getRhs().set( 0, string );
096      functionCall.getOutputCollector().add( functionCall.getContext().getRhs() );
097      }
098    }
099  }