org.apache.nutch.crawl
Class GeneratorJob

java.lang.Object
  org.apache.hadoop.conf.Configured
      org.apache.nutch.util.NutchTool
          org.apache.nutch.crawl.GeneratorJob

All Implemented Interfaces:

Configurable, Tool

public class GeneratorJob
extends NutchTool
implements Tool

Nested Class Summary

static class	GeneratorJob.SelectorEntry
static class	GeneratorJob.SelectorEntryComparator

Field Summary

static String	BATCH_ID
static String	GENERATE_UPDATE_CRAWLDB
static String	GENERATOR_COUNT_MODE
static String	GENERATOR_COUNT_VALUE_DOMAIN
static String	GENERATOR_COUNT_VALUE_HOST
static String	GENERATOR_COUNT_VALUE_IP
static String	GENERATOR_CUR_TIME
static String	GENERATOR_DELAY
static String	GENERATOR_FILTER
static String	GENERATOR_MAX_COUNT
static String	GENERATOR_MIN_SCORE
static String	GENERATOR_NORMALISE
static String	GENERATOR_RANDOM_SEED
static String	GENERATOR_TOP_N
static org.slf4j.Logger	LOG

Fields inherited from class org.apache.nutch.util.NutchTool

currentJob, currentJobNum, numJobs, results, status

Constructor Summary

GeneratorJob()

GeneratorJob(Configuration conf)

Method Summary

String	generate(long topN, long curTime, boolean filter, boolean norm) Mark URLs ready for fetching.
static void	main(String[] args)
Map<String,Object>	run(Map<String,Object> args) Runs the tool, using a map of arguments.
int	run(String[] args)

Methods inherited from class org.apache.nutch.util.NutchTool

getProgress, getStatus, killJob, stopJob

Methods inherited from class org.apache.hadoop.conf.Configured

getConf, setConf

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.apache.hadoop.conf.Configurable

getConf, setConf

Field Detail

GENERATE_UPDATE_CRAWLDB

public static final String GENERATE_UPDATE_CRAWLDB

See Also:

Constant Field Values

GENERATOR_MIN_SCORE

public static final String GENERATOR_MIN_SCORE

See Also:

Constant Field Values

GENERATOR_FILTER

public static final String GENERATOR_FILTER

See Also:

Constant Field Values

GENERATOR_NORMALISE

public static final String GENERATOR_NORMALISE

See Also:

Constant Field Values

GENERATOR_MAX_COUNT

public static final String GENERATOR_MAX_COUNT

See Also:

Constant Field Values

GENERATOR_COUNT_MODE

public static final String GENERATOR_COUNT_MODE

See Also:

Constant Field Values

GENERATOR_COUNT_VALUE_DOMAIN

public static final String GENERATOR_COUNT_VALUE_DOMAIN

See Also:

Constant Field Values

GENERATOR_COUNT_VALUE_HOST

public static final String GENERATOR_COUNT_VALUE_HOST

See Also:

Constant Field Values

GENERATOR_COUNT_VALUE_IP

public static final String GENERATOR_COUNT_VALUE_IP

See Also:

Constant Field Values

GENERATOR_TOP_N

public static final String GENERATOR_TOP_N

See Also:

Constant Field Values

GENERATOR_CUR_TIME

public static final String GENERATOR_CUR_TIME

See Also:

Constant Field Values

GENERATOR_DELAY

public static final String GENERATOR_DELAY

See Also:

Constant Field Values

GENERATOR_RANDOM_SEED

public static final String GENERATOR_RANDOM_SEED

See Also:

Constant Field Values

BATCH_ID

public static final String BATCH_ID

See Also:

Constant Field Values

LOG

public static final org.slf4j.Logger LOG

Constructor Detail

GeneratorJob

public GeneratorJob()

GeneratorJob

public GeneratorJob(Configuration conf)

Method Detail

run

public Map<String,Object> run(Map<String,Object> args)
                       throws Exception

Description copied from class: NutchTool

Runs the tool, using a map of arguments. May return results, or null.

Specified by:

run in class NutchTool

Throws:

Exception

generate

public String generate(long topN,
                       long curTime,
                       boolean filter,
                       boolean norm)
                throws Exception

Mark URLs ready for fetching.

Throws:

ClassNotFoundException

InterruptedException

Exception

run

public int run(String[] args)
        throws Exception

Specified by:

run in interface Tool

Throws:

Exception

main

public static void main(String[] args)
                 throws Exception

Throws:

Exception