Search in sources :

Example 1 with Sum

use of storm.trident.operation.builtin.Sum in project jstorm by alibaba.

the class TridentReach method buildTopology.

public static StormTopology buildTopology(LocalDRPC drpc) {
    TridentTopology topology = new TridentTopology();
    TridentState urlToTweeters = topology.newStaticState(new StaticSingleKeyMapState.Factory(TWEETERS_DB));
    TridentState tweetersToFollowers = topology.newStaticState(new StaticSingleKeyMapState.Factory(FOLLOWERS_DB));
    topology.newDRPCStream("reach", drpc).stateQuery(urlToTweeters, new Fields("args"), new MapGet(), new Fields("tweeters")).each(new Fields("tweeters"), new ExpandList(), new Fields("tweeter")).shuffle().stateQuery(tweetersToFollowers, new Fields("tweeter"), new MapGet(), new Fields("followers")).each(new Fields("followers"), new ExpandList(), new Fields("follower")).groupBy(new Fields("follower")).aggregate(new One(), new Fields("one")).aggregate(new Fields("one"), new Sum(), new Fields("reach"));
    return topology.build();
}
Also used : Fields(backtype.storm.tuple.Fields) TridentTopology(storm.trident.TridentTopology) TridentState(storm.trident.TridentState) MapGet(storm.trident.operation.builtin.MapGet) Sum(storm.trident.operation.builtin.Sum)

Example 2 with Sum

use of storm.trident.operation.builtin.Sum in project jstorm by alibaba.

the class TridentReachTest method testTridentReach.

@Test
public void testTridentReach() {
    TridentTopology tridentTopology = new TridentTopology();
    TridentState urlToTweeters = tridentTopology.newStaticState(new TridentReach.StaticSingleKeyMapState.Factory(TWEETERS));
    TridentState tweetersToFollowers = tridentTopology.newStaticState(new TridentReach.StaticSingleKeyMapState.Factory(FOLLOWERS));
    LocalDRPC localDRPC = new LocalDRPC();
    tridentTopology.newDRPCStream("reach", localDRPC).stateQuery(urlToTweeters, new Fields("args"), new MapGet(), new Fields("tweeters")).each(new Fields("tweeters"), new TridentReach.ExpandList(), new Fields("tweeter")).shuffle().stateQuery(tweetersToFollowers, new Fields("tweeter"), new MapGet(), new Fields("followers")).each(new Fields("followers"), new TridentReach.ExpandList(), new Fields("follower")).groupBy(new Fields("follower")).aggregate(new TridentReach.One(), new Fields("one")).aggregate(new Fields("one"), new Sum(), new Fields("reach"));
    Map config = new HashMap();
    config.put(Config.TOPOLOGY_NAME, "TridentReachTest");
    JStormUnitTestDRPCValidator validator = new JStormUnitTestDRPCValidator(localDRPC) {

        @Override
        public boolean validate(Map config) {
            String query = executeLocalDRPC("reach", "aaa");
            assertEquals("[[0]]", query);
            query = executeLocalDRPC("reach", "foo.com/blog/1");
            assertEquals("[[16]]", query);
            query = executeLocalDRPC("reach", "engineering.twitter.com/blog/5");
            assertEquals("[[14]]", query);
            return true;
        }
    };
    try {
        JStormUnitTestRunner.submitTopology(tridentTopology.build(), config, 120, validator);
    } finally {
        localDRPC.shutdown();
    }
}
Also used : TridentReach(org.apache.storm.starter.trident.TridentReach) TridentState(storm.trident.TridentState) HashMap(java.util.HashMap) MapGet(storm.trident.operation.builtin.MapGet) Sum(storm.trident.operation.builtin.Sum) JStormUnitTestDRPCValidator(com.jstorm.example.unittests.utils.JStormUnitTestDRPCValidator) Fields(backtype.storm.tuple.Fields) TridentTopology(storm.trident.TridentTopology) LocalDRPC(backtype.storm.LocalDRPC) HashMap(java.util.HashMap) Map(java.util.Map) Test(org.junit.Test)

Example 3 with Sum

use of storm.trident.operation.builtin.Sum in project jstorm by alibaba.

the class TridentWordCountTest method testTridentWordCount.

// to make sure the validator is right
@Test
public void testTridentWordCount() {
    LocalDRPC localDRPC = new LocalDRPC();
    FixedLimitBatchSpout spout = new FixedLimitBatchSpout(SPOUT_LIMIT, new Fields("sentence"), SPOUT_BATCH_SIZE, new Values("the cow jumped over the moon"), new Values("the man went to the store and bought some candy"), new Values("four score and seven years ago"), new Values("how many apples can you eat"), new Values("to be or not to be the person"));
    TridentTopology tridentTopology = new TridentTopology();
    TridentState wordCount = tridentTopology.newStream("spout", spout).parallelismHint(1).each(new Fields("sentence"), new Split(), new Fields("word")).groupBy(new Fields("word")).persistentAggregate(new MemoryMapState.Factory(), new Count(), new Fields("count")).parallelismHint(16);
    tridentTopology.newDRPCStream("words", localDRPC).each(new Fields("args"), new Split(), new Fields("keyword")).groupBy(new Fields("keyword")).stateQuery(wordCount, new Fields("keyword"), new MapGet(), new Fields("result")).each(new Fields("result"), new FilterNull()).aggregate(new Fields("result"), new Sum(), new Fields("sum"));
    Map config = new HashMap();
    config.put(Config.TOPOLOGY_NAME, "TridentWordCountTest");
    JStormUnitTestValidator validator = new JStormUnitTestDRPCValidator(localDRPC) {

        Logger LOG = LoggerFactory.getLogger(JStormUnitTestValidator.class);

        @Override
        public boolean validate(Map config) {
            String queryResult = executeLocalDRPC("words", "the");
            //the result is like [[8080]], so remove the [[]]
            queryResult = queryResult.substring(2, queryResult.length() - 2);
            //how many times of emit can finish a loop
            int oneLoopNeedEmits = (int) Math.ceil(SPOUT_CONTENT_TYPES / (float) SPOUT_BATCH_SIZE);
            // of all the spout content
            //the loop time of the LimitFixBatchSpout content
            int loopTime = SPOUT_LIMIT / oneLoopNeedEmits;
            int receiveCountOfThe = Integer.valueOf(queryResult);
            LOG.info("Final receive total " + receiveCountOfThe + " \"the\" when expected " + (loopTime * 5));
            //5 "the" are in one loop
            boolean isCountOfTheRight = (receiveCountOfThe == loopTime * 5);
            // query the word count of these 3 words total
            queryResult = executeLocalDRPC("words", "be store kujou");
            queryResult = queryResult.substring(2, queryResult.length() - 2);
            int receiveCountOfBeAndStore = Integer.valueOf(queryResult);
            LOG.info("Final receive total " + receiveCountOfBeAndStore + " \"be\" and \"store\" and \"kujou\" " + "when expected " + (loopTime * 3));
            //2 "be" 1 "store" 0 "kujou" are in one loop
            boolean isCountOfBeAndStoreRight = (receiveCountOfBeAndStore == loopTime * 3);
            return isCountOfTheRight && isCountOfBeAndStoreRight;
        }
    };
    try {
        boolean result = JStormUnitTestRunner.submitTopology(tridentTopology.build(), config, 90, validator);
        assertTrue("Topology should pass the validator", result);
    } finally {
        localDRPC.shutdown();
    }
}
Also used : FilterNull(storm.trident.operation.builtin.FilterNull) JStormUnitTestValidator(com.jstorm.example.unittests.utils.JStormUnitTestValidator) TridentState(storm.trident.TridentState) HashMap(java.util.HashMap) Values(backtype.storm.tuple.Values) LoggerFactory(org.slf4j.LoggerFactory) MapGet(storm.trident.operation.builtin.MapGet) Sum(storm.trident.operation.builtin.Sum) JStormUnitTestDRPCValidator(com.jstorm.example.unittests.utils.JStormUnitTestDRPCValidator) Count(storm.trident.operation.builtin.Count) Logger(org.slf4j.Logger) Fields(backtype.storm.tuple.Fields) TridentTopology(storm.trident.TridentTopology) LocalDRPC(backtype.storm.LocalDRPC) HashMap(java.util.HashMap) Map(java.util.Map) Test(org.junit.Test)

Example 4 with Sum

use of storm.trident.operation.builtin.Sum in project jstorm by alibaba.

the class TridentMapExample method buildTopology.

public static StormTopology buildTopology(LocalDRPC drpc) {
    FixedBatchSpout spout = new FixedBatchSpout(new Fields("word"), 3, new Values("the cow jumped over the moon"), new Values("the man went to the store and bought some candy"), new Values("four score and seven years ago"), new Values("how many apples can you eat"), new Values("to be or not to be the person"));
    spout.setCycle(true);
    TridentTopology topology = new TridentTopology();
    TridentState wordCounts = topology.newStream("spout1", spout).parallelismHint(16).flatMap(split).map(toUpper).filter(theFilter).peek(new Consumer() {

        @Override
        public void accept(TridentTuple input) {
            System.out.println(input.getString(0));
        }
    }).groupBy(new Fields("word")).persistentAggregate(new MemoryMapState.Factory(), new Count(), new Fields("count")).parallelismHint(16);
    topology.newDRPCStream("words", drpc).flatMap(split).groupBy(new Fields("args")).stateQuery(wordCounts, new Fields("args"), new MapGet(), new Fields("count")).filter(new FilterNull()).aggregate(new Fields("count"), new Sum(), new Fields("sum"));
    return topology.build();
}
Also used : FixedBatchSpout(storm.trident.testing.FixedBatchSpout) FilterNull(storm.trident.operation.builtin.FilterNull) Fields(backtype.storm.tuple.Fields) Consumer(storm.trident.operation.Consumer) TridentTopology(storm.trident.TridentTopology) TridentState(storm.trident.TridentState) Values(backtype.storm.tuple.Values) MapGet(storm.trident.operation.builtin.MapGet) Sum(storm.trident.operation.builtin.Sum) Count(storm.trident.operation.builtin.Count) TridentTuple(storm.trident.tuple.TridentTuple)

Example 5 with Sum

use of storm.trident.operation.builtin.Sum in project jstorm by alibaba.

the class TridentWordCount method buildTopology.

public static StormTopology buildTopology(LocalDRPC drpc) {
    FixedBatchSpout spout = new FixedBatchSpout(new Fields("sentence"), 3, new Values("the cow jumped over the moon"), new Values("the man went to the store and bought some candy"), new Values("four score and seven years ago"), new Values("how many apples can you eat"), new Values("to be or not to be the person"));
    spout.setCycle(true);
    TridentTopology topology = new TridentTopology();
    TridentState wordCounts = topology.newStream("spout1", spout).parallelismHint(16).each(new Fields("sentence"), new Split(), new Fields("word")).groupBy(new Fields("word")).persistentAggregate(new MemoryMapState.Factory(), new Count(), new Fields("count")).parallelismHint(16);
    topology.newDRPCStream("words", drpc).each(new Fields("args"), new Split(), new Fields("word")).groupBy(new Fields("word")).stateQuery(wordCounts, new Fields("word"), new MapGet(), new Fields("count")).each(new Fields("count"), new FilterNull()).aggregate(new Fields("count"), new Sum(), new Fields("sum"));
    return topology.build();
}
Also used : FixedBatchSpout(storm.trident.testing.FixedBatchSpout) FilterNull(storm.trident.operation.builtin.FilterNull) Fields(backtype.storm.tuple.Fields) TridentTopology(storm.trident.TridentTopology) TridentState(storm.trident.TridentState) Values(backtype.storm.tuple.Values) MapGet(storm.trident.operation.builtin.MapGet) Sum(storm.trident.operation.builtin.Sum) Count(storm.trident.operation.builtin.Count)

Aggregations

Fields (backtype.storm.tuple.Fields)6 TridentState (storm.trident.TridentState)6 TridentTopology (storm.trident.TridentTopology)6 MapGet (storm.trident.operation.builtin.MapGet)6 Sum (storm.trident.operation.builtin.Sum)6 Values (backtype.storm.tuple.Values)4 Count (storm.trident.operation.builtin.Count)4 FilterNull (storm.trident.operation.builtin.FilterNull)4 LocalDRPC (backtype.storm.LocalDRPC)3 JStormUnitTestDRPCValidator (com.jstorm.example.unittests.utils.JStormUnitTestDRPCValidator)3 Test (org.junit.Test)3 JStormUnitTestValidator (com.jstorm.example.unittests.utils.JStormUnitTestValidator)2 HashMap (java.util.HashMap)2 Map (java.util.Map)2 Logger (org.slf4j.Logger)2 LoggerFactory (org.slf4j.LoggerFactory)2 FixedBatchSpout (storm.trident.testing.FixedBatchSpout)2 TridentReach (org.apache.storm.starter.trident.TridentReach)1 Consumer (storm.trident.operation.Consumer)1 TridentTuple (storm.trident.tuple.TridentTuple)1