Search in sources :

Example 26 with StorageDescriptor

use of org.apache.hadoop.hive.metastore.api.StorageDescriptor in project hive by apache.

the class TestHBaseAggregateStatsExtrapolation method MiddleOfPartitionsHaveBitVectorStatus.

@Test
public void MiddleOfPartitionsHaveBitVectorStatus() throws Exception {
    String dbName = "default";
    String tableName = "snp";
    long now = System.currentTimeMillis();
    List<FieldSchema> cols = new ArrayList<>();
    cols.add(new FieldSchema("col4", "long", "nocomment"));
    SerDeInfo serde = new SerDeInfo("serde", "seriallib", null);
    StorageDescriptor sd = new StorageDescriptor(cols, "file:/tmp", "input", "output", false, 0, serde, null, null, Collections.<String, String>emptyMap());
    List<FieldSchema> partCols = new ArrayList<>();
    partCols.add(new FieldSchema("ds", "string", ""));
    Table table = new Table(tableName, dbName, "me", (int) now, (int) now, 0, sd, partCols, Collections.<String, String>emptyMap(), null, null, null);
    store.createTable(table);
    List<List<String>> partVals = new ArrayList<>();
    for (int i = 0; i < 10; i++) {
        List<String> partVal = Arrays.asList("" + i);
        partVals.add(partVal);
        StorageDescriptor psd = new StorageDescriptor(sd);
        psd.setLocation("file:/tmp/default/hit/ds=" + partVal);
        Partition part = new Partition(partVal, dbName, tableName, (int) now, (int) now, psd, Collections.<String, String>emptyMap());
        store.addPartition(part);
        if (i > 2 && i < 7) {
            ColumnStatistics cs = new ColumnStatistics();
            ColumnStatisticsDesc desc = new ColumnStatisticsDesc(false, dbName, tableName);
            desc.setLastAnalyzed(now);
            desc.setPartName("ds=" + partVal);
            cs.setStatsDesc(desc);
            ColumnStatisticsObj obj = new ColumnStatisticsObj();
            obj.setColName("col4");
            obj.setColType("long");
            ColumnStatisticsData data = new ColumnStatisticsData();
            LongColumnStatsData dcsd = new LongColumnStatsData();
            dcsd.setHighValue(1000 + i);
            dcsd.setLowValue(-1000 - i);
            dcsd.setNumNulls(i);
            dcsd.setNumDVs(10 * i);
            dcsd.setBitVectors(bitVectors);
            data.setLongStats(dcsd);
            obj.setStatsData(data);
            cs.addToStatsObj(obj);
            store.updatePartitionColumnStatistics(cs, partVal);
        }
    }
    Checker statChecker = new Checker() {

        @Override
        public void checkStats(AggrStats aggrStats) throws Exception {
            Assert.assertEquals(4, aggrStats.getPartsFound());
            Assert.assertEquals(1, aggrStats.getColStatsSize());
            ColumnStatisticsObj cso = aggrStats.getColStats().get(0);
            Assert.assertEquals("col4", cso.getColName());
            Assert.assertEquals("long", cso.getColType());
            LongColumnStatsData lcsd = cso.getStatsData().getLongStats();
            Assert.assertEquals(1006, lcsd.getHighValue(), 0.01);
            Assert.assertEquals(-1006, lcsd.getLowValue(), 0.01);
            Assert.assertEquals(45, lcsd.getNumNulls());
            Assert.assertEquals(3, lcsd.getNumDVs());
        }
    };
    List<String> partNames = new ArrayList<>();
    for (int i = 0; i < 10; i++) {
        partNames.add("ds=" + i);
    }
    AggrStats aggrStats = store.get_aggr_stats_for(dbName, tableName, partNames, Arrays.asList("col4"));
    statChecker.checkStats(aggrStats);
}
Also used : ColumnStatistics(org.apache.hadoop.hive.metastore.api.ColumnStatistics) Partition(org.apache.hadoop.hive.metastore.api.Partition) Table(org.apache.hadoop.hive.metastore.api.Table) AggrStats(org.apache.hadoop.hive.metastore.api.AggrStats) FieldSchema(org.apache.hadoop.hive.metastore.api.FieldSchema) SerDeInfo(org.apache.hadoop.hive.metastore.api.SerDeInfo) ArrayList(java.util.ArrayList) StorageDescriptor(org.apache.hadoop.hive.metastore.api.StorageDescriptor) LongColumnStatsData(org.apache.hadoop.hive.metastore.api.LongColumnStatsData) ColumnStatisticsObj(org.apache.hadoop.hive.metastore.api.ColumnStatisticsObj) ColumnStatisticsDesc(org.apache.hadoop.hive.metastore.api.ColumnStatisticsDesc) ArrayList(java.util.ArrayList) List(java.util.List) ColumnStatisticsData(org.apache.hadoop.hive.metastore.api.ColumnStatisticsData) Test(org.junit.Test)

Example 27 with StorageDescriptor

use of org.apache.hadoop.hive.metastore.api.StorageDescriptor in project hive by apache.

the class TestHBaseAggregateStatsExtrapolation method allPartitionsHaveBitVectorStatusDouble.

@Test
public void allPartitionsHaveBitVectorStatusDouble() throws Exception {
    String dbName = "default";
    String tableName = "snp";
    long now = System.currentTimeMillis();
    List<FieldSchema> cols = new ArrayList<>();
    cols.add(new FieldSchema("col1_double", "double", "nocomment"));
    SerDeInfo serde = new SerDeInfo("serde", "seriallib", null);
    StorageDescriptor sd = new StorageDescriptor(cols, "file:/tmp", "input", "output", false, 0, serde, null, null, Collections.<String, String>emptyMap());
    List<FieldSchema> partCols = new ArrayList<>();
    partCols.add(new FieldSchema("ds", "string", ""));
    Table table = new Table(tableName, dbName, "me", (int) now, (int) now, 0, sd, partCols, Collections.<String, String>emptyMap(), null, null, null);
    store.createTable(table);
    List<List<String>> partVals = new ArrayList<>();
    for (int i = 0; i < 10; i++) {
        List<String> partVal = Arrays.asList("" + i);
        partVals.add(partVal);
        StorageDescriptor psd = new StorageDescriptor(sd);
        psd.setLocation("file:/tmp/default/hit/ds=" + partVal);
        Partition part = new Partition(partVal, dbName, tableName, (int) now, (int) now, psd, Collections.<String, String>emptyMap());
        store.addPartition(part);
        ColumnStatistics cs = new ColumnStatistics();
        ColumnStatisticsDesc desc = new ColumnStatisticsDesc(false, dbName, tableName);
        desc.setLastAnalyzed(now);
        desc.setPartName("ds=" + partVal);
        cs.setStatsDesc(desc);
        ColumnStatisticsObj obj = new ColumnStatisticsObj();
        obj.setColName("col1_double");
        obj.setColType("double");
        ColumnStatisticsData data = new ColumnStatisticsData();
        DoubleColumnStatsData dcsd = new DoubleColumnStatsData();
        dcsd.setHighValue(1000 + i);
        dcsd.setLowValue(-1000 - i);
        dcsd.setNumNulls(i);
        dcsd.setNumDVs(10 * i + 1);
        dcsd.setBitVectors(bitVectors);
        data.setDoubleStats(dcsd);
        obj.setStatsData(data);
        cs.addToStatsObj(obj);
        store.updatePartitionColumnStatistics(cs, partVal);
    }
    Checker statChecker = new Checker() {

        @Override
        public void checkStats(AggrStats aggrStats) throws Exception {
            Assert.assertEquals(10, aggrStats.getPartsFound());
            Assert.assertEquals(1, aggrStats.getColStatsSize());
            ColumnStatisticsObj cso = aggrStats.getColStats().get(0);
            Assert.assertEquals("col1_double", cso.getColName());
            Assert.assertEquals("double", cso.getColType());
            DoubleColumnStatsData lcsd = cso.getStatsData().getDoubleStats();
            Assert.assertEquals(1009, lcsd.getHighValue(), 0.01);
            Assert.assertEquals(-1009, lcsd.getLowValue(), 0.01);
            Assert.assertEquals(45, lcsd.getNumNulls());
            Assert.assertEquals(3, lcsd.getNumDVs());
        }
    };
    List<String> partNames = new ArrayList<>();
    for (int i = 0; i < 10; i++) {
        partNames.add("ds=" + i);
    }
    AggrStats aggrStats = store.get_aggr_stats_for(dbName, tableName, partNames, Arrays.asList("col1_double"));
    statChecker.checkStats(aggrStats);
}
Also used : ColumnStatistics(org.apache.hadoop.hive.metastore.api.ColumnStatistics) Partition(org.apache.hadoop.hive.metastore.api.Partition) Table(org.apache.hadoop.hive.metastore.api.Table) AggrStats(org.apache.hadoop.hive.metastore.api.AggrStats) FieldSchema(org.apache.hadoop.hive.metastore.api.FieldSchema) SerDeInfo(org.apache.hadoop.hive.metastore.api.SerDeInfo) ArrayList(java.util.ArrayList) StorageDescriptor(org.apache.hadoop.hive.metastore.api.StorageDescriptor) ColumnStatisticsObj(org.apache.hadoop.hive.metastore.api.ColumnStatisticsObj) DoubleColumnStatsData(org.apache.hadoop.hive.metastore.api.DoubleColumnStatsData) ColumnStatisticsDesc(org.apache.hadoop.hive.metastore.api.ColumnStatisticsDesc) ArrayList(java.util.ArrayList) List(java.util.List) ColumnStatisticsData(org.apache.hadoop.hive.metastore.api.ColumnStatisticsData) Test(org.junit.Test)

Example 28 with StorageDescriptor

use of org.apache.hadoop.hive.metastore.api.StorageDescriptor in project hive by apache.

the class TestHBaseAggregateStatsExtrapolation method allPartitionsHaveBitVectorStatusString.

@Test
public void allPartitionsHaveBitVectorStatusString() throws Exception {
    String dbName = "default";
    String tableName = "snp";
    long now = System.currentTimeMillis();
    List<FieldSchema> cols = new ArrayList<>();
    cols.add(new FieldSchema("col1_string", "string", "nocomment"));
    SerDeInfo serde = new SerDeInfo("serde", "seriallib", null);
    StorageDescriptor sd = new StorageDescriptor(cols, "file:/tmp", "input", "output", false, 0, serde, null, null, Collections.<String, String>emptyMap());
    List<FieldSchema> partCols = new ArrayList<>();
    partCols.add(new FieldSchema("ds", "string", ""));
    Table table = new Table(tableName, dbName, "me", (int) now, (int) now, 0, sd, partCols, Collections.<String, String>emptyMap(), null, null, null);
    store.createTable(table);
    List<List<String>> partVals = new ArrayList<>();
    for (int i = 0; i < 10; i++) {
        List<String> partVal = Arrays.asList("" + i);
        partVals.add(partVal);
        StorageDescriptor psd = new StorageDescriptor(sd);
        psd.setLocation("file:/tmp/default/hit/ds=" + partVal);
        Partition part = new Partition(partVal, dbName, tableName, (int) now, (int) now, psd, Collections.<String, String>emptyMap());
        store.addPartition(part);
        ColumnStatistics cs = new ColumnStatistics();
        ColumnStatisticsDesc desc = new ColumnStatisticsDesc(false, dbName, tableName);
        desc.setLastAnalyzed(now);
        desc.setPartName("ds=" + partVal);
        cs.setStatsDesc(desc);
        ColumnStatisticsObj obj = new ColumnStatisticsObj();
        obj.setColName("col1_string");
        obj.setColType("string");
        ColumnStatisticsData data = new ColumnStatisticsData();
        StringColumnStatsData dcsd = new StringColumnStatsData();
        dcsd.setAvgColLen(i + 1);
        dcsd.setMaxColLen(i + 10);
        dcsd.setNumNulls(i);
        dcsd.setNumDVs(10 * i + 1);
        dcsd.setBitVectors(bitVectors);
        data.setStringStats(dcsd);
        obj.setStatsData(data);
        cs.addToStatsObj(obj);
        store.updatePartitionColumnStatistics(cs, partVal);
    }
    Checker statChecker = new Checker() {

        @Override
        public void checkStats(AggrStats aggrStats) throws Exception {
            Assert.assertEquals(10, aggrStats.getPartsFound());
            Assert.assertEquals(1, aggrStats.getColStatsSize());
            ColumnStatisticsObj cso = aggrStats.getColStats().get(0);
            Assert.assertEquals("col1_string", cso.getColName());
            Assert.assertEquals("string", cso.getColType());
            StringColumnStatsData lcsd = cso.getStatsData().getStringStats();
            Assert.assertEquals(10, lcsd.getAvgColLen(), 0.01);
            Assert.assertEquals(19, lcsd.getMaxColLen(), 0.01);
            Assert.assertEquals(45, lcsd.getNumNulls());
            Assert.assertEquals(3, lcsd.getNumDVs());
        }
    };
    List<String> partNames = new ArrayList<>();
    for (int i = 0; i < 10; i++) {
        partNames.add("ds=" + i);
    }
    AggrStats aggrStats = store.get_aggr_stats_for(dbName, tableName, partNames, Arrays.asList("col1_string"));
    statChecker.checkStats(aggrStats);
}
Also used : ColumnStatistics(org.apache.hadoop.hive.metastore.api.ColumnStatistics) Partition(org.apache.hadoop.hive.metastore.api.Partition) Table(org.apache.hadoop.hive.metastore.api.Table) AggrStats(org.apache.hadoop.hive.metastore.api.AggrStats) FieldSchema(org.apache.hadoop.hive.metastore.api.FieldSchema) SerDeInfo(org.apache.hadoop.hive.metastore.api.SerDeInfo) ArrayList(java.util.ArrayList) StorageDescriptor(org.apache.hadoop.hive.metastore.api.StorageDescriptor) StringColumnStatsData(org.apache.hadoop.hive.metastore.api.StringColumnStatsData) ColumnStatisticsObj(org.apache.hadoop.hive.metastore.api.ColumnStatisticsObj) ColumnStatisticsDesc(org.apache.hadoop.hive.metastore.api.ColumnStatisticsDesc) ArrayList(java.util.ArrayList) List(java.util.List) ColumnStatisticsData(org.apache.hadoop.hive.metastore.api.ColumnStatisticsData) Test(org.junit.Test)

Example 29 with StorageDescriptor

use of org.apache.hadoop.hive.metastore.api.StorageDescriptor in project hive by apache.

the class TestHBaseAggregateStatsExtrapolation method TwoEndsAndMiddleOfPartitionsHaveBitVectorStatusLong.

@Test
public void TwoEndsAndMiddleOfPartitionsHaveBitVectorStatusLong() throws Exception {
    String dbName = "default";
    String tableName = "snp";
    long now = System.currentTimeMillis();
    List<FieldSchema> cols = new ArrayList<>();
    cols.add(new FieldSchema("col5", "long", "nocomment"));
    SerDeInfo serde = new SerDeInfo("serde", "seriallib", null);
    StorageDescriptor sd = new StorageDescriptor(cols, "file:/tmp", "input", "output", false, 0, serde, null, null, Collections.<String, String>emptyMap());
    List<FieldSchema> partCols = new ArrayList<>();
    partCols.add(new FieldSchema("ds", "string", ""));
    Table table = new Table(tableName, dbName, "me", (int) now, (int) now, 0, sd, partCols, Collections.<String, String>emptyMap(), null, null, null);
    store.createTable(table);
    List<List<String>> partVals = new ArrayList<>();
    for (int i = 0; i < 10; i++) {
        List<String> partVal = Arrays.asList("" + i);
        partVals.add(partVal);
        StorageDescriptor psd = new StorageDescriptor(sd);
        psd.setLocation("file:/tmp/default/hit/ds=" + partVal);
        Partition part = new Partition(partVal, dbName, tableName, (int) now, (int) now, psd, Collections.<String, String>emptyMap());
        store.addPartition(part);
        if (i == 0 || i == 2 || i == 3 || i == 5 || i == 6 || i == 8) {
            ColumnStatistics cs = new ColumnStatistics();
            ColumnStatisticsDesc desc = new ColumnStatisticsDesc(false, dbName, tableName);
            desc.setLastAnalyzed(now);
            desc.setPartName("ds=" + partVal);
            cs.setStatsDesc(desc);
            ColumnStatisticsObj obj = new ColumnStatisticsObj();
            obj.setColName("col5");
            obj.setColType("long");
            ColumnStatisticsData data = new ColumnStatisticsData();
            LongColumnStatsData dcsd = new LongColumnStatsData();
            dcsd.setHighValue(1000 + i);
            dcsd.setLowValue(-1000 - i);
            dcsd.setNumNulls(i);
            dcsd.setNumDVs(10 * i);
            dcsd.setBitVectors(bitVectors);
            data.setLongStats(dcsd);
            obj.setStatsData(data);
            cs.addToStatsObj(obj);
            store.updatePartitionColumnStatistics(cs, partVal);
        }
    }
    Checker statChecker = new Checker() {

        @Override
        public void checkStats(AggrStats aggrStats) throws Exception {
            Assert.assertEquals(6, aggrStats.getPartsFound());
            Assert.assertEquals(1, aggrStats.getColStatsSize());
            ColumnStatisticsObj cso = aggrStats.getColStats().get(0);
            Assert.assertEquals("col5", cso.getColName());
            Assert.assertEquals("long", cso.getColType());
            LongColumnStatsData lcsd = cso.getStatsData().getLongStats();
            Assert.assertEquals(1010, lcsd.getHighValue(), 0.01);
            Assert.assertEquals(-1010, lcsd.getLowValue(), 0.01);
            Assert.assertEquals(40, lcsd.getNumNulls());
            Assert.assertEquals(3, lcsd.getNumDVs());
        }
    };
    List<String> partNames = new ArrayList<>();
    for (int i = 0; i < 10; i++) {
        partNames.add("ds=" + i);
    }
    AggrStats aggrStats = store.get_aggr_stats_for(dbName, tableName, partNames, Arrays.asList("col5"));
    statChecker.checkStats(aggrStats);
}
Also used : ColumnStatistics(org.apache.hadoop.hive.metastore.api.ColumnStatistics) Partition(org.apache.hadoop.hive.metastore.api.Partition) Table(org.apache.hadoop.hive.metastore.api.Table) AggrStats(org.apache.hadoop.hive.metastore.api.AggrStats) FieldSchema(org.apache.hadoop.hive.metastore.api.FieldSchema) SerDeInfo(org.apache.hadoop.hive.metastore.api.SerDeInfo) ArrayList(java.util.ArrayList) StorageDescriptor(org.apache.hadoop.hive.metastore.api.StorageDescriptor) LongColumnStatsData(org.apache.hadoop.hive.metastore.api.LongColumnStatsData) ColumnStatisticsObj(org.apache.hadoop.hive.metastore.api.ColumnStatisticsObj) ColumnStatisticsDesc(org.apache.hadoop.hive.metastore.api.ColumnStatisticsDesc) ArrayList(java.util.ArrayList) List(java.util.List) ColumnStatisticsData(org.apache.hadoop.hive.metastore.api.ColumnStatisticsData) Test(org.junit.Test)

Example 30 with StorageDescriptor

use of org.apache.hadoop.hive.metastore.api.StorageDescriptor in project hive by apache.

the class TestObjectStore method testPartitionOps.

/**
   * Tests partition operations
   */
@Test
public void testPartitionOps() throws MetaException, InvalidObjectException, NoSuchObjectException, InvalidInputException {
    Database db1 = new Database(DB1, "description", "locationurl", null);
    objectStore.createDatabase(db1);
    StorageDescriptor sd = new StorageDescriptor(null, "location", null, null, false, 0, new SerDeInfo("SerDeName", "serializationLib", null), null, null, null);
    HashMap<String, String> tableParams = new HashMap<String, String>();
    tableParams.put("EXTERNAL", "false");
    FieldSchema partitionKey1 = new FieldSchema("Country", serdeConstants.STRING_TYPE_NAME, "");
    FieldSchema partitionKey2 = new FieldSchema("State", serdeConstants.STRING_TYPE_NAME, "");
    Table tbl1 = new Table(TABLE1, DB1, "owner", 1, 2, 3, sd, Arrays.asList(partitionKey1, partitionKey2), tableParams, null, null, "MANAGED_TABLE");
    objectStore.createTable(tbl1);
    HashMap<String, String> partitionParams = new HashMap<String, String>();
    partitionParams.put("PARTITION_LEVEL_PRIVILEGE", "true");
    List<String> value1 = Arrays.asList("US", "CA");
    Partition part1 = new Partition(value1, DB1, TABLE1, 111, 111, sd, partitionParams);
    objectStore.addPartition(part1);
    List<String> value2 = Arrays.asList("US", "MA");
    Partition part2 = new Partition(value2, DB1, TABLE1, 222, 222, sd, partitionParams);
    objectStore.addPartition(part2);
    Deadline.startTimer("getPartition");
    List<Partition> partitions = objectStore.getPartitions(DB1, TABLE1, 10);
    Assert.assertEquals(2, partitions.size());
    Assert.assertEquals(111, partitions.get(0).getCreateTime());
    Assert.assertEquals(222, partitions.get(1).getCreateTime());
    int numPartitions = objectStore.getNumPartitionsByFilter(DB1, TABLE1, "");
    Assert.assertEquals(partitions.size(), numPartitions);
    numPartitions = objectStore.getNumPartitionsByFilter(DB1, TABLE1, "country = \"US\"");
    Assert.assertEquals(2, numPartitions);
    objectStore.dropPartition(DB1, TABLE1, value1);
    partitions = objectStore.getPartitions(DB1, TABLE1, 10);
    Assert.assertEquals(1, partitions.size());
    Assert.assertEquals(222, partitions.get(0).getCreateTime());
    objectStore.dropPartition(DB1, TABLE1, value2);
    objectStore.dropTable(DB1, TABLE1);
    objectStore.dropDatabase(DB1);
}
Also used : Partition(org.apache.hadoop.hive.metastore.api.Partition) Table(org.apache.hadoop.hive.metastore.api.Table) HashMap(java.util.HashMap) SerDeInfo(org.apache.hadoop.hive.metastore.api.SerDeInfo) FieldSchema(org.apache.hadoop.hive.metastore.api.FieldSchema) Database(org.apache.hadoop.hive.metastore.api.Database) StorageDescriptor(org.apache.hadoop.hive.metastore.api.StorageDescriptor) Test(org.junit.Test)

Aggregations

StorageDescriptor (org.apache.hadoop.hive.metastore.api.StorageDescriptor)191 SerDeInfo (org.apache.hadoop.hive.metastore.api.SerDeInfo)139 Table (org.apache.hadoop.hive.metastore.api.Table)139 FieldSchema (org.apache.hadoop.hive.metastore.api.FieldSchema)138 ArrayList (java.util.ArrayList)126 Test (org.junit.Test)103 Partition (org.apache.hadoop.hive.metastore.api.Partition)78 HashMap (java.util.HashMap)49 Order (org.apache.hadoop.hive.metastore.api.Order)32 Database (org.apache.hadoop.hive.metastore.api.Database)31 List (java.util.List)28 ColumnStatistics (org.apache.hadoop.hive.metastore.api.ColumnStatistics)28 ColumnStatisticsData (org.apache.hadoop.hive.metastore.api.ColumnStatisticsData)28 ColumnStatisticsDesc (org.apache.hadoop.hive.metastore.api.ColumnStatisticsDesc)28 ColumnStatisticsObj (org.apache.hadoop.hive.metastore.api.ColumnStatisticsObj)28 AggrStats (org.apache.hadoop.hive.metastore.api.AggrStats)26 MetaException (org.apache.hadoop.hive.metastore.api.MetaException)26 LazySimpleSerDe (org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe)21 HiveInputFormat (org.apache.hadoop.hive.ql.io.HiveInputFormat)18 HiveOutputFormat (org.apache.hadoop.hive.ql.io.HiveOutputFormat)18