Examples with HiveEndPoint org.apache.hive.hcatalog.streaming.HiveEndPoint used on opensource projects

Example 1 with HiveEndPoint

use of org.apache.hive.hcatalog.streaming.HiveEndPoint in project storm by apache.

the class TestHiveWriter method testWriteBasic.

@Test
public void testWriteBasic() throws Exception {
    DelimitedRecordHiveMapper mapper = new DelimitedRecordHiveMapper().withColumnFields(new Fields(colNames)).withPartitionFields(new Fields(partNames));
    HiveEndPoint endPoint = new HiveEndPoint(metaStoreURI, dbName, tblName, Arrays.asList(partitionVals));
    HiveWriter writer = new HiveWriter(endPoint, 10, true, timeout, callTimeoutPool, mapper, ugi);
    writeTuples(writer, mapper, 3);
    writer.flush(false);
    writer.close();
    checkRecordCountInTable(dbName, tblName, 3);
}

Example 2 with HiveEndPoint

use of org.apache.hive.hcatalog.streaming.HiveEndPoint in project storm by apache.

the class TestHiveWriter method testInstantiate.

@Test
public void testInstantiate() throws Exception {
    DelimitedRecordHiveMapper mapper = new DelimitedRecordHiveMapper().withColumnFields(new Fields(colNames)).withPartitionFields(new Fields(partNames));
    HiveEndPoint endPoint = new HiveEndPoint(metaStoreURI, dbName, tblName, Arrays.asList(partitionVals));
    HiveWriter writer = new HiveWriter(endPoint, 10, true, timeout, callTimeoutPool, mapper, ugi);
    writer.close();
}

Example 3 with HiveEndPoint

use of org.apache.hive.hcatalog.streaming.HiveEndPoint in project hive by apache.

the class TestCompactor method minorCompactWhileStreaming.

@Test
public void minorCompactWhileStreaming() throws Exception {
    String dbName = "default";
    String tblName = "cws";
    List<String> colNames = Arrays.asList("a", "b");
    String columnNamesProperty = "a,b";
    String columnTypesProperty = "int:string";
    executeStatementOnDriver("drop table if exists " + tblName, driver);
    executeStatementOnDriver("CREATE TABLE " + tblName + "(a INT, b STRING) " + //currently ACID requires table to be bucketed
    " CLUSTERED BY(a) INTO 1 BUCKETS" + " STORED AS ORC  TBLPROPERTIES ('transactional'='true')", driver);
    HiveEndPoint endPt = new HiveEndPoint(null, dbName, tblName, null);
    DelimitedInputWriter writer = new DelimitedInputWriter(new String[] { "a", "b" }, ",", endPt);
    StreamingConnection connection = endPt.newConnection(false, "UT_" + Thread.currentThread().getName());
    try {
        // Write a couple of batches
        for (int i = 0; i < 2; i++) {
            writeBatch(connection, writer, false);
        }
        // Start a third batch, but don't close it.
        writeBatch(connection, writer, true);
        // Now, compact
        TxnStore txnHandler = TxnUtils.getTxnStore(conf);
        txnHandler.compact(new CompactionRequest(dbName, tblName, CompactionType.MINOR));
        Worker t = new Worker();
        t.setThreadId((int) t.getId());
        t.setHiveConf(conf);
        AtomicBoolean stop = new AtomicBoolean(true);
        AtomicBoolean looped = new AtomicBoolean();
        t.init(stop, looped);
        t.run();
        // Find the location of the table
        IMetaStoreClient msClient = new HiveMetaStoreClient(conf);
        Table table = msClient.getTable(dbName, tblName);
        FileSystem fs = FileSystem.get(conf);
        FileStatus[] stat = fs.listStatus(new Path(table.getSd().getLocation()), AcidUtils.deltaFileFilter);
        String[] names = new String[stat.length];
        Path resultFile = null;
        for (int i = 0; i < names.length; i++) {
            names[i] = stat[i].getPath().getName();
            if (names[i].equals("delta_0000001_0000004")) {
                resultFile = stat[i].getPath();
            }
        }
        Arrays.sort(names);
        String[] expected = new String[] { "delta_0000001_0000002", "delta_0000001_0000004", "delta_0000003_0000004", "delta_0000005_0000006" };
        if (!Arrays.deepEquals(expected, names)) {
            Assert.fail("Expected: " + Arrays.toString(expected) + ", found: " + Arrays.toString(names));
        }
        checkExpectedTxnsPresent(null, new Path[] { resultFile }, columnNamesProperty, columnTypesProperty, 0, 1L, 4L);
    } finally {
        connection.close();
    }
}

Example 4 with HiveEndPoint

use of org.apache.hive.hcatalog.streaming.HiveEndPoint in project hive by apache.

the class TestCompactor method majorCompactWhileStreaming.

@Test
public void majorCompactWhileStreaming() throws Exception {
    String dbName = "default";
    String tblName = "cws";
    List<String> colNames = Arrays.asList("a", "b");
    String columnNamesProperty = "a,b";
    String columnTypesProperty = "int:string";
    executeStatementOnDriver("drop table if exists " + tblName, driver);
    executeStatementOnDriver("CREATE TABLE " + tblName + "(a INT, b STRING) " + //currently ACID requires table to be bucketed
    " CLUSTERED BY(a) INTO 1 BUCKETS" + " STORED AS ORC  TBLPROPERTIES ('transactional'='true') ", driver);
    HiveEndPoint endPt = new HiveEndPoint(null, dbName, tblName, null);
    DelimitedInputWriter writer = new DelimitedInputWriter(new String[] { "a", "b" }, ",", endPt);
    StreamingConnection connection = endPt.newConnection(false, "UT_" + Thread.currentThread().getName());
    try {
        // Write a couple of batches
        for (int i = 0; i < 2; i++) {
            writeBatch(connection, writer, false);
        }
        // Start a third batch, but don't close it.
        writeBatch(connection, writer, true);
        // Now, compact
        TxnStore txnHandler = TxnUtils.getTxnStore(conf);
        txnHandler.compact(new CompactionRequest(dbName, tblName, CompactionType.MAJOR));
        Worker t = new Worker();
        t.setThreadId((int) t.getId());
        t.setHiveConf(conf);
        AtomicBoolean stop = new AtomicBoolean(true);
        AtomicBoolean looped = new AtomicBoolean();
        t.init(stop, looped);
        t.run();
        // Find the location of the table
        IMetaStoreClient msClient = new HiveMetaStoreClient(conf);
        Table table = msClient.getTable(dbName, tblName);
        FileSystem fs = FileSystem.get(conf);
        FileStatus[] stat = fs.listStatus(new Path(table.getSd().getLocation()), AcidUtils.baseFileFilter);
        if (1 != stat.length) {
            Assert.fail("Expecting 1 file \"base_0000004\" and found " + stat.length + " files " + Arrays.toString(stat));
        }
        String name = stat[0].getPath().getName();
        Assert.assertEquals(name, "base_0000004");
        checkExpectedTxnsPresent(stat[0].getPath(), null, columnNamesProperty, columnTypesProperty, 0, 1L, 4L);
    } finally {
        connection.close();
    }
}

Example 5 with HiveEndPoint

use of org.apache.hive.hcatalog.streaming.HiveEndPoint in project hive by apache.

the class TestCompactor method minorCompactWhileStreamingWithSplitUpdate.

@Test
public void minorCompactWhileStreamingWithSplitUpdate() throws Exception {
    String dbName = "default";
    String tblName = "cws";
    List<String> colNames = Arrays.asList("a", "b");
    String columnNamesProperty = "a,b";
    String columnTypesProperty = "int:string";
    executeStatementOnDriver("drop table if exists " + tblName, driver);
    executeStatementOnDriver("CREATE TABLE " + tblName + "(a INT, b STRING) " + //currently ACID requires table to be bucketed
    " CLUSTERED BY(a) INTO 1 BUCKETS" + " STORED AS ORC  TBLPROPERTIES ('transactional'='true'," + "'transactional_properties'='default')", driver);
    HiveEndPoint endPt = new HiveEndPoint(null, dbName, tblName, null);
    DelimitedInputWriter writer = new DelimitedInputWriter(new String[] { "a", "b" }, ",", endPt);
    StreamingConnection connection = endPt.newConnection(false, "UT_" + Thread.currentThread().getName());
    try {
        // Write a couple of batches
        for (int i = 0; i < 2; i++) {
            writeBatch(connection, writer, false);
        }
        // Start a third batch, but don't close it.
        writeBatch(connection, writer, true);
        // Now, compact
        TxnStore txnHandler = TxnUtils.getTxnStore(conf);
        txnHandler.compact(new CompactionRequest(dbName, tblName, CompactionType.MINOR));
        Worker t = new Worker();
        t.setThreadId((int) t.getId());
        t.setHiveConf(conf);
        AtomicBoolean stop = new AtomicBoolean(true);
        AtomicBoolean looped = new AtomicBoolean();
        t.init(stop, looped);
        t.run();
        // Find the location of the table
        IMetaStoreClient msClient = new HiveMetaStoreClient(conf);
        Table table = msClient.getTable(dbName, tblName);
        FileSystem fs = FileSystem.get(conf);
        FileStatus[] stat = fs.listStatus(new Path(table.getSd().getLocation()), AcidUtils.deltaFileFilter);
        String[] names = new String[stat.length];
        Path resultFile = null;
        for (int i = 0; i < names.length; i++) {
            names[i] = stat[i].getPath().getName();
            if (names[i].equals("delta_0000001_0000004")) {
                resultFile = stat[i].getPath();
            }
        }
        Arrays.sort(names);
        String[] expected = new String[] { "delta_0000001_0000002", "delta_0000001_0000004", "delta_0000003_0000004", "delta_0000005_0000006" };
        if (!Arrays.deepEquals(expected, names)) {
            Assert.fail("Expected: " + Arrays.toString(expected) + ", found: " + Arrays.toString(names));
        }
        checkExpectedTxnsPresent(null, new Path[] { resultFile }, columnNamesProperty, columnTypesProperty, 0, 1L, 4L);
        // Verify that we have got correct set of delete_deltas also
        FileStatus[] deleteDeltaStat = fs.listStatus(new Path(table.getSd().getLocation()), AcidUtils.deleteEventDeltaDirFilter);
        String[] deleteDeltas = new String[deleteDeltaStat.length];
        Path minorCompactedDeleteDelta = null;
        for (int i = 0; i < deleteDeltas.length; i++) {
            deleteDeltas[i] = deleteDeltaStat[i].getPath().getName();
            if (deleteDeltas[i].equals("delete_delta_0000001_0000004")) {
                minorCompactedDeleteDelta = deleteDeltaStat[i].getPath();
            }
        }
        Arrays.sort(deleteDeltas);
        String[] expectedDeleteDeltas = new String[] { "delete_delta_0000001_0000004" };
        if (!Arrays.deepEquals(expectedDeleteDeltas, deleteDeltas)) {
            Assert.fail("Expected: " + Arrays.toString(expectedDeleteDeltas) + ", found: " + Arrays.toString(deleteDeltas));
        }
        // There should be no rows in the delete_delta because there have been no delete events.
        checkExpectedTxnsPresent(null, new Path[] { minorCompactedDeleteDelta }, columnNamesProperty, columnTypesProperty, 0, 0L, 0L);
    } finally {
        connection.close();
    }
}