use of org.apache.hadoop.hive.metastore.api.ShowCompactResponse in project hive by apache.
the class TestCompactor method schemaEvolutionAddColDynamicPartitioningUpdate.
@Test
public void schemaEvolutionAddColDynamicPartitioningUpdate() throws Exception {
String tblName = "udpct";
List<String> colNames = Arrays.asList("a", "b");
executeStatementOnDriver("drop table if exists " + tblName, driver);
executeStatementOnDriver("CREATE TABLE " + tblName + "(a INT, b STRING) " + " PARTITIONED BY(ds string)" + // currently ACID requires table to be bucketed
" CLUSTERED BY(a) INTO 2 BUCKETS" + " STORED AS ORC TBLPROPERTIES ('transactional'='true')", driver);
executeStatementOnDriver("insert into " + tblName + " partition (ds) values (1, 'fred', " + "'today'), (2, 'wilma', 'yesterday')", driver);
executeStatementOnDriver("update " + tblName + " set b = 'barney'", driver);
// Validate the update.
executeStatementOnDriver("SELECT * FROM " + tblName + " ORDER BY a", driver);
ArrayList<String> valuesReadFromHiveDriver = new ArrayList<String>();
driver.getResults(valuesReadFromHiveDriver);
Assert.assertEquals(2, valuesReadFromHiveDriver.size());
Assert.assertEquals("1\tbarney\ttoday", valuesReadFromHiveDriver.get(0));
Assert.assertEquals("2\tbarney\tyesterday", valuesReadFromHiveDriver.get(1));
// ALTER TABLE ... ADD COLUMNS
executeStatementOnDriver("ALTER TABLE " + tblName + " ADD COLUMNS(c int)", driver);
// Validate there is an added NULL for column c.
executeStatementOnDriver("SELECT * FROM " + tblName + " ORDER BY a", driver);
valuesReadFromHiveDriver = new ArrayList<String>();
driver.getResults(valuesReadFromHiveDriver);
Assert.assertEquals(2, valuesReadFromHiveDriver.size());
Assert.assertEquals("1\tbarney\tNULL\ttoday", valuesReadFromHiveDriver.get(0));
Assert.assertEquals("2\tbarney\tNULL\tyesterday", valuesReadFromHiveDriver.get(1));
// Second INSERT round with new inserts into previously existing partition 'yesterday'.
executeStatementOnDriver("insert into " + tblName + " partition (ds) values " + "(3, 'mark', 1900, 'soon'), (4, 'douglas', 1901, 'last_century'), " + "(5, 'doc', 1902, 'yesterday')", driver);
// Validate there the new insertions for column c.
executeStatementOnDriver("SELECT * FROM " + tblName + " ORDER BY a", driver);
valuesReadFromHiveDriver = new ArrayList<String>();
driver.getResults(valuesReadFromHiveDriver);
Assert.assertEquals(5, valuesReadFromHiveDriver.size());
Assert.assertEquals("1\tbarney\tNULL\ttoday", valuesReadFromHiveDriver.get(0));
Assert.assertEquals("2\tbarney\tNULL\tyesterday", valuesReadFromHiveDriver.get(1));
Assert.assertEquals("3\tmark\t1900\tsoon", valuesReadFromHiveDriver.get(2));
Assert.assertEquals("4\tdouglas\t1901\tlast_century", valuesReadFromHiveDriver.get(3));
Assert.assertEquals("5\tdoc\t1902\tyesterday", valuesReadFromHiveDriver.get(4));
executeStatementOnDriver("update " + tblName + " set c = 2000", driver);
// Validate the update of new column c, even in old rows.
executeStatementOnDriver("SELECT * FROM " + tblName + " ORDER BY a", driver);
valuesReadFromHiveDriver = new ArrayList<String>();
driver.getResults(valuesReadFromHiveDriver);
Assert.assertEquals(5, valuesReadFromHiveDriver.size());
Assert.assertEquals("1\tbarney\t2000\ttoday", valuesReadFromHiveDriver.get(0));
Assert.assertEquals("2\tbarney\t2000\tyesterday", valuesReadFromHiveDriver.get(1));
Assert.assertEquals("3\tmark\t2000\tsoon", valuesReadFromHiveDriver.get(2));
Assert.assertEquals("4\tdouglas\t2000\tlast_century", valuesReadFromHiveDriver.get(3));
Assert.assertEquals("5\tdoc\t2000\tyesterday", valuesReadFromHiveDriver.get(4));
Initiator initiator = new Initiator();
initiator.setThreadId((int) initiator.getId());
// Set to 1 so insert doesn't set it off but update does
conf.setIntVar(HiveConf.ConfVars.HIVE_COMPACTOR_DELTA_NUM_THRESHOLD, 1);
initiator.setConf(conf);
AtomicBoolean stop = new AtomicBoolean();
stop.set(true);
initiator.init(stop, new AtomicBoolean());
initiator.run();
TxnStore txnHandler = TxnUtils.getTxnStore(conf);
ShowCompactResponse rsp = txnHandler.showCompact(new ShowCompactRequest());
List<ShowCompactResponseElement> compacts = rsp.getCompacts();
Assert.assertEquals(4, compacts.size());
SortedSet<String> partNames = new TreeSet<String>();
for (int i = 0; i < compacts.size(); i++) {
Assert.assertEquals("default", compacts.get(i).getDbname());
Assert.assertEquals(tblName, compacts.get(i).getTablename());
Assert.assertEquals("initiated", compacts.get(i).getState());
partNames.add(compacts.get(i).getPartitionname());
}
List<String> names = new ArrayList<String>(partNames);
Assert.assertEquals("ds=last_century", names.get(0));
Assert.assertEquals("ds=soon", names.get(1));
Assert.assertEquals("ds=today", names.get(2));
Assert.assertEquals("ds=yesterday", names.get(3));
// Validate after compaction.
executeStatementOnDriver("SELECT * FROM " + tblName + " ORDER BY a", driver);
valuesReadFromHiveDriver = new ArrayList<String>();
driver.getResults(valuesReadFromHiveDriver);
Assert.assertEquals(5, valuesReadFromHiveDriver.size());
Assert.assertEquals("1\tbarney\t2000\ttoday", valuesReadFromHiveDriver.get(0));
Assert.assertEquals("2\tbarney\t2000\tyesterday", valuesReadFromHiveDriver.get(1));
Assert.assertEquals("3\tmark\t2000\tsoon", valuesReadFromHiveDriver.get(2));
Assert.assertEquals("4\tdouglas\t2000\tlast_century", valuesReadFromHiveDriver.get(3));
Assert.assertEquals("5\tdoc\t2000\tyesterday", valuesReadFromHiveDriver.get(4));
}
use of org.apache.hadoop.hive.metastore.api.ShowCompactResponse in project hive by apache.
the class TestCompactor method dynamicPartitioningInsert.
@Test
public void dynamicPartitioningInsert() throws Exception {
String tblName = "dpct";
List<String> colNames = Arrays.asList("a", "b");
executeStatementOnDriver("drop table if exists " + tblName, driver);
executeStatementOnDriver("CREATE TABLE " + tblName + "(a INT, b STRING) " + " PARTITIONED BY(ds string)" + // currently ACID requires table to be bucketed
" CLUSTERED BY(a) INTO 2 BUCKETS" + " STORED AS ORC TBLPROPERTIES ('transactional'='true')", driver);
executeStatementOnDriver("insert into " + tblName + " partition (ds) values (1, 'fred', " + "'today'), (2, 'wilma', 'yesterday')", driver);
Initiator initiator = new Initiator();
initiator.setThreadId((int) initiator.getId());
conf.setIntVar(HiveConf.ConfVars.HIVE_COMPACTOR_DELTA_NUM_THRESHOLD, 0);
initiator.setConf(conf);
AtomicBoolean stop = new AtomicBoolean();
stop.set(true);
initiator.init(stop, new AtomicBoolean());
initiator.run();
TxnStore txnHandler = TxnUtils.getTxnStore(conf);
ShowCompactResponse rsp = txnHandler.showCompact(new ShowCompactRequest());
List<ShowCompactResponseElement> compacts = rsp.getCompacts();
Assert.assertEquals(2, compacts.size());
SortedSet<String> partNames = new TreeSet<String>();
for (int i = 0; i < compacts.size(); i++) {
Assert.assertEquals("default", compacts.get(i).getDbname());
Assert.assertEquals(tblName, compacts.get(i).getTablename());
Assert.assertEquals("initiated", compacts.get(i).getState());
partNames.add(compacts.get(i).getPartitionname());
}
List<String> names = new ArrayList<String>(partNames);
Assert.assertEquals("ds=today", names.get(0));
Assert.assertEquals("ds=yesterday", names.get(1));
}
use of org.apache.hadoop.hive.metastore.api.ShowCompactResponse in project hive by apache.
the class TestTxnHandler method testCompactWhenAlreadyCompacting.
/**
* Once a Compaction for a given resource is scheduled/working, we should not
* schedule another one to prevent concurrent compactions for the same resource.
* @throws Exception
*/
@Test
public void testCompactWhenAlreadyCompacting() throws Exception {
CompactionRequest rqst = new CompactionRequest("foo", "bar", CompactionType.MAJOR);
rqst.setPartitionname("ds=today");
CompactionResponse resp = txnHandler.compact(rqst);
Assert.assertEquals(resp, new CompactionResponse(1, TxnStore.INITIATED_RESPONSE, true));
ShowCompactResponse rsp = txnHandler.showCompact(new ShowCompactRequest());
List<ShowCompactResponseElement> compacts = rsp.getCompacts();
assertEquals(1, compacts.size());
rqst.setType(CompactionType.MINOR);
resp = txnHandler.compact(rqst);
Assert.assertEquals(resp, new CompactionResponse(1, TxnStore.INITIATED_RESPONSE, false));
rsp = txnHandler.showCompact(new ShowCompactRequest());
compacts = rsp.getCompacts();
assertEquals(1, compacts.size());
ShowCompactResponseElement c = compacts.get(0);
assertEquals("foo", c.getDbname());
assertEquals("bar", c.getTablename());
assertEquals("ds=today", c.getPartitionname());
assertEquals(CompactionType.MAJOR, c.getType());
assertEquals("initiated", c.getState());
assertEquals(0L, c.getStart());
}
use of org.apache.hadoop.hive.metastore.api.ShowCompactResponse in project hive by apache.
the class TestTxnCommands2 method testOriginalFileReaderWhenNonAcidConvertedToAcid.
@Test
public void testOriginalFileReaderWhenNonAcidConvertedToAcid() throws Exception {
// 1. Insert five rows to Non-ACID table.
runStatementOnDriver("insert into " + Table.NONACIDORCTBL + "(a,b) values(1,2),(3,4),(5,6),(7,8),(9,10)");
// 2. Convert NONACIDORCTBL to ACID table. //todo: remove trans_prop after HIVE-17089
runStatementOnDriver("alter table " + Table.NONACIDORCTBL + " SET TBLPROPERTIES ('transactional'='true', 'transactional_properties'='default')");
runStatementOnDriver("update " + Table.NONACIDORCTBL + " set b = b*2 where b in (4,10)");
runStatementOnDriver("delete from " + Table.NONACIDORCTBL + " where a = 7");
List<String> rs = runStatementOnDriver("select a,b from " + Table.NONACIDORCTBL + " order by a,b");
int[][] resultData = new int[][] { { 1, 2 }, { 3, 8 }, { 5, 6 }, { 9, 20 } };
Assert.assertEquals(stringifyValues(resultData), rs);
// 3. Perform a major compaction.
runStatementOnDriver("alter table " + Table.NONACIDORCTBL + " compact 'MAJOR'");
runWorker(hiveConf);
TxnStore txnHandler = TxnUtils.getTxnStore(hiveConf);
ShowCompactResponse resp = txnHandler.showCompact(new ShowCompactRequest());
Assert.assertEquals("Unexpected number of compactions in history", 1, resp.getCompactsSize());
Assert.assertEquals("Unexpected 0 compaction state", TxnStore.CLEANING_RESPONSE, resp.getCompacts().get(0).getState());
Assert.assertTrue(resp.getCompacts().get(0).getHadoopJobId().startsWith("job_local"));
// 3. Perform a delete.
runStatementOnDriver("delete from " + Table.NONACIDORCTBL + " where a = 1");
rs = runStatementOnDriver("select a,b from " + Table.NONACIDORCTBL + " order by a,b");
resultData = new int[][] { { 3, 8 }, { 5, 6 }, { 9, 20 } };
Assert.assertEquals(stringifyValues(resultData), rs);
}
use of org.apache.hadoop.hive.metastore.api.ShowCompactResponse in project hive by apache.
the class TestTxnNoBuckets method testCompactStatsGather.
/**
* HIVE-17900
*/
@Test
public void testCompactStatsGather() throws Exception {
hiveConf.setVar(HiveConf.ConfVars.DYNAMICPARTITIONINGMODE, "nonstrict");
runStatementOnDriver("drop table if exists T");
runStatementOnDriver("create table T(a int, b int) partitioned by (p int, q int) " + "stored as orc TBLPROPERTIES ('transactional'='true')");
int[][] targetVals = { { 4, 1, 1 }, { 4, 2, 2 }, { 4, 3, 1 }, { 4, 4, 2 } };
// we only recompute stats after major compact if they existed before
runStatementOnDriver("insert into T partition(p=1,q) " + makeValuesClause(targetVals));
runStatementOnDriver("analyze table T partition(p=1) compute statistics for columns");
IMetaStoreClient hms = Hive.get().getMSC();
List<String> partNames = new ArrayList<>();
partNames.add("p=1/q=2");
List<String> colNames = new ArrayList<>();
colNames.add("a");
Map<String, List<ColumnStatisticsObj>> map = hms.getPartitionColumnStatistics("default", "T", partNames, colNames);
Assert.assertEquals(4, map.get(partNames.get(0)).get(0).getStatsData().getLongStats().getHighValue());
int[][] targetVals2 = { { 5, 1, 1 }, { 5, 2, 2 }, { 5, 3, 1 }, { 5, 4, 2 } };
runStatementOnDriver("insert into T partition(p=1,q) " + makeValuesClause(targetVals2));
String query = "select ROW__ID, p, q, a, b, INPUT__FILE__NAME from T order by p, q, a, b";
List<String> rs = runStatementOnDriver(query);
String[][] expected = { { "{\"writeid\":1,\"bucketid\":536870912,\"rowid\":0}\t1\t1\t4\t1", "t/p=1/q=1/delta_0000001_0000001_0000/bucket_00000" }, { "{\"writeid\":1,\"bucketid\":536870912,\"rowid\":1}\t1\t1\t4\t3", "t/p=1/q=1/delta_0000001_0000001_0000/bucket_00000" }, { "{\"writeid\":2,\"bucketid\":536870912,\"rowid\":0}\t1\t1\t5\t1", "t/p=1/q=1/delta_0000002_0000002_0000/bucket_00000" }, { "{\"writeid\":2,\"bucketid\":536870912,\"rowid\":1}\t1\t1\t5\t3", "t/p=1/q=1/delta_0000002_0000002_0000/bucket_00000" }, { "{\"writeid\":1,\"bucketid\":536870912,\"rowid\":0}\t1\t2\t4\t2", "t/p=1/q=2/delta_0000001_0000001_0000/bucket_00000" }, { "{\"writeid\":1,\"bucketid\":536870912,\"rowid\":1}\t1\t2\t4\t4", "t/p=1/q=2/delta_0000001_0000001_0000/bucket_00000" }, { "{\"writeid\":2,\"bucketid\":536870912,\"rowid\":0}\t1\t2\t5\t2", "t/p=1/q=2/delta_0000002_0000002_0000/bucket_00000" }, { "{\"writeid\":2,\"bucketid\":536870912,\"rowid\":1}\t1\t2\t5\t4", "t/p=1/q=2/delta_0000002_0000002_0000/bucket_00000" } };
checkExpected(rs, expected, "insert data");
// run major compaction
runStatementOnDriver("alter table T partition(p=1,q=2) compact 'major'");
TestTxnCommands2.runWorker(hiveConf);
query = "select ROW__ID, p, q, a, b, INPUT__FILE__NAME from T order by p, q, a, b";
rs = runStatementOnDriver(query);
String[][] expected2 = { { "{\"writeid\":1,\"bucketid\":536870912,\"rowid\":0}\t1\t1\t4\t1", "t/p=1/q=1/delta_0000001_0000001_0000/bucket_00000" }, { "{\"writeid\":1,\"bucketid\":536870912,\"rowid\":1}\t1\t1\t4\t3", "t/p=1/q=1/delta_0000001_0000001_0000/bucket_00000" }, { "{\"writeid\":2,\"bucketid\":536870912,\"rowid\":0}\t1\t1\t5\t1", "t/p=1/q=1/delta_0000002_0000002_0000/bucket_00000" }, { "{\"writeid\":2,\"bucketid\":536870912,\"rowid\":1}\t1\t1\t5\t3", "t/p=1/q=1/delta_0000002_0000002_0000/bucket_00000" }, { "{\"writeid\":1,\"bucketid\":536870912,\"rowid\":0}\t1\t2\t4\t2", "t/p=1/q=2/base_0000002/bucket_00000" }, { "{\"writeid\":1,\"bucketid\":536870912,\"rowid\":1}\t1\t2\t4\t4", "t/p=1/q=2/base_0000002/bucket_00000" }, { "{\"writeid\":2,\"bucketid\":536870912,\"rowid\":0}\t1\t2\t5\t2", "t/p=1/q=2/base_0000002/bucket_00000" }, { "{\"writeid\":2,\"bucketid\":536870912,\"rowid\":1}\t1\t2\t5\t4", "t/p=1/q=2/base_0000002/bucket_00000" } };
checkExpected(rs, expected2, "after major compaction");
// check status of compaction job
TxnStore txnHandler = TxnUtils.getTxnStore(hiveConf);
ShowCompactResponse resp = txnHandler.showCompact(new ShowCompactRequest());
Assert.assertEquals("Unexpected number of compactions in history", 1, resp.getCompactsSize());
Assert.assertEquals("Unexpected 0 compaction state", TxnStore.CLEANING_RESPONSE, resp.getCompacts().get(0).getState());
Assert.assertTrue(resp.getCompacts().get(0).getHadoopJobId().startsWith("job_local"));
// now check that stats were updated
map = hms.getPartitionColumnStatistics("default", "T", partNames, colNames);
Assert.assertEquals("", 5, map.get(partNames.get(0)).get(0).getStatsData().getLongStats().getHighValue());
}
Aggregations