Examples of cascading.scheme.hadoop.TextLine

cascading.scheme.hadoop.TextLine
A TextLine is a type of {@link cascading.scheme.Scheme} for plain text files. Files are broken intolines. Either line-feed or carriage-return are used to signal end of line.
By default, this scheme returns a {@link Tuple} with two fields, "offset" and "line".
Many of the constructors take both "sourceFields" and "sinkFields". sourceFields denote the field names to be used instead of the names "offset" and "line". sinkFields is a selector and is by default {@link Fields#ALL}. Any available field names can be given if only a subset of the incoming fields should be used.
If a {@link Fields} instance is passed on the constructor as sourceFields having only one field, the return tupleswill simply be the "line" value using the given field name.
Note that TextLine will concatenate all the Tuple values for the selected fields with a TAB delimiter before writing out the line.
Note sink compression is {@link Compress#DISABLE} by default. If {@code null} is passed to the constructorfor the compression value, it will remain disabled.
If any of the input files end with ".zip", an error will be thrown. *
By default, all text is encoded/decoded as UTF-8. This can be changed via the {@code charsetName} constructorargument.

    for( int i = 0; i < n; i++ )
      {
      String nameIn = "in" + i;
      String nameOut = "out" + i;
      Pipe pipe = new Pipe( nameIn );
      sources.put( nameIn, new Hfs( new TextLine( new Fields( "key" + i ) ), "foo/in" + i ) );
      sinks.put( nameOut, new Hfs( new TextLine(), "foo/out" + i ) );


      count += 2; // 2 taps


      if( i > 0 )
        {

View Full Code Here

   * @throws IOException
   */
  @Test
  public void testIdentity() throws Exception
    {
    Tap source = new Hfs( new TextLine(), "input/path" );
    Tap sink = new Hfs( new TextLine(), "output/path", true );


    Pipe pipe = new Pipe( "test" );


    Flow flow = getPlatform().getFlowConnector().connect( source, sink, pipe );

View Full Code Here

  public void testOneJob() throws IOException
    {
    Map sources = new HashMap();
    Map sinks = new HashMap();


    sources.put( "count", new Hfs( new TextLine( new Fields( "first", "second" ) ), "input/path" ) );
    sinks.put( "count", new Hfs( new TextLine( new Fields( 0, 1 ) ), "output/path" ) );


    Pipe pipe = new Pipe( "count" );
    pipe = new GroupBy( pipe, new Fields( 1 ) );
    pipe = new Every( pipe, new Fields( 1 ), new Count(), new Fields( 0, 1 ) );

View Full Code Here

  public void testOneJob2() throws IOException
    {
    Map sources = new HashMap();
    Map sinks = new HashMap();


    sources.put( "count", new Hfs( new TextLine( new Fields( "first", "second" ) ), "input/path" ) );
    sinks.put( "count", new Hfs( new TextLine( new Fields( 0, 1 ) ), "output/path" ) );


    Pipe pipe = new Pipe( "count" );
    pipe = new Each( pipe, new Fields( 1 ), new Identity(), new Fields( 2 ) ); // in:second out:all
    pipe = new Each( pipe, new Fields( 0 ), new Identity( new Fields( "_all" ) ), new Fields( 1 ) ); // in:all out:_all
    pipe = new GroupBy( pipe, new Fields( 0 ) ); // in:_all out:_all

View Full Code Here

  public void testOneJob3() throws IOException
    {
    Map sources = new HashMap();
    Map sinks = new HashMap();


    sources.put( "a", new Hfs( new TextLine( new Fields( "first", "second" ) ), "input/path/a" ) );
    sources.put( "b", new Hfs( new TextLine( new Fields( "third", "fourth" ) ), "input/path/b" ) );


    Pipe pipeA = new Pipe( "a" );
    Pipe pipeB = new Pipe( "b" );


    Pipe splice = new CoGroup( pipeA, new Fields( 1 ), pipeB, new Fields( 1 ) );


    sinks.put( splice.getName(), new Hfs( new TextLine( new Fields( 0, 1 ) ), "output/path" ) );


    List steps = getPlatform().getFlowConnector().connect( sources, sinks, splice ).getFlowSteps();


    assertEquals( "wrong size", 1, steps.size() );

View Full Code Here

  public void testOneJob4() throws IOException
    {
    Map sources = new HashMap();
    Map sinks = new HashMap();


    sources.put( "a", new Hfs( new TextLine( new Fields( "first", "second" ) ), "input/path/a" ) );
    sources.put( "b", new Hfs( new TextLine( new Fields( "third", "fourth" ) ), "input/path/b" ) );


    Pipe pipeA = new Pipe( "a" );
    Pipe pipeB = new Pipe( "b" );


    Pipe cogroup = new CoGroup( pipeA, new Fields( 1 ), pipeB, new Fields( 1 ) );


    cogroup = new Each( cogroup, new Identity() );


    sinks.put( cogroup.getName(), new Hfs( new TextLine( new Fields( 0, 1 ) ), "output/path" ) );


    List steps = getPlatform().getFlowConnector().connect( sources, sinks, cogroup ).getFlowSteps();


    assertEquals( "wrong size", 1, steps.size() );

View Full Code Here

  public void testOneJob5() throws IOException
    {
    Map sources = new HashMap();
    Map sinks = new HashMap();


    sources.put( "a", new Hfs( new TextLine( new Fields( "first", "second" ) ), "input/path/a" ) );
    sources.put( "b", new Hfs( new TextLine( new Fields( "third", "fourth" ) ), "input/path/b" ) );


    Pipe pipeA = new Pipe( "a" );
    Pipe pipeB = new Pipe( "b" );


    Pipe splice = new CoGroup( pipeA, pipeB );


    splice = new Each( splice, new Identity() );


    sinks.put( splice.getName(), new Hfs( new TextLine(), "output/path" ) );


    List steps = getPlatform().getFlowConnector().connect( sources, sinks, splice ).getFlowSteps();


    assertEquals( "wrong size", 1, steps.size() );

View Full Code Here

  public void testNoGroup() throws IOException
    {
    Map sources = new HashMap();
    Map sinks = new HashMap();


    sources.put( "count", new Hfs( new TextLine( new Fields( "first", "second" ) ), "input/path" ) );
    sinks.put( "count", new Hfs( new TextLine( new Fields( 0, 1 ) ), "output/path" ) );


    Pipe pipe = new Pipe( "count" );
    pipe = new Each( pipe, new Identity() );
    pipe = new Every( pipe, new Fields( 1 ), new Count(), new Fields( 0, 1 ) );

View Full Code Here


  /** This should result in only two steps, one for each side */
  @Test
  public void testSplit()
    {
    Tap source = new Hfs( new TextLine( new Fields( "offset", "line" ) ), "foo" );
    Tap sink1 = new Hfs( new TextLine(), "foo/split1", true );
    Tap sink2 = new Hfs( new TextLine(), "foo/split2", true );


    Pipe pipe = new Pipe( "split" );


    pipe = new Each( pipe, new Fields( "line" ), new RegexFilter( "^68.*" ) );

View Full Code Here


  /** this test verifies that the planner recognizes there are fewer tails than sinks. */
  @Test
  public void testSplitHangingTails()
    {
    Tap source = new Hfs( new TextLine( new Fields( "offset", "line" ) ), "foo" );
    Tap sink1 = new Hfs( new TextLine(), "foo/split1", true );
    Tap sink2 = new Hfs( new TextLine(), "foo/split2", true );


    Pipe pipe = new Pipe( "split" );


    pipe = new Each( pipe, new Fields( "line" ), new RegexFilter( "^68.*" ) );

View Full Code Here

0 1 2 3 4 5 6 7 8

TOP

Related Classes of cascading.scheme.hadoop.TextLine

cascading.flow.hadoop.BuildJobsHadoopPlatformTest

cascading.flow.hadoop.FlowPlatformTest

cascading.flow.hadoop.MapReduceFlowPlatformTest

cascading.flow.hadoop.util.HadoopMRUtil

cascading.flow.hadoop.util.HadoopUtil

cascading.lingual.platform.hadoop.HadoopPlatformBroker

cascading.lingual.platform.hadoop2.Hadoop2MR1PlatformBroker

cascading.platform.hadoop.BaseHadoopPlatform

cascading.tap.hadoop.HadoopMR1TapPlatformTest

cascading.tap.hadoop.HadoopTapPlatformTest

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.