Search in sources :

Example 6 with ValidReaderWriteIdList

use of org.apache.hadoop.hive.common.ValidReaderWriteIdList in project hive by apache.

the class TestAcidUtils method testBaseWithDeleteDeltas.

@Test
public void testBaseWithDeleteDeltas() throws Exception {
    Configuration conf = new Configuration();
    conf.setInt(HiveConf.ConfVars.HIVE_TXN_OPERATIONAL_PROPERTIES.varname, AcidOperationalProperties.getDefault().toInt());
    MockFileSystem fs = new MockFileSystem(conf, new MockFile("mock:/tbl/part1/base_5/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/base_10/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/base_49/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delta_025_025/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delta_029_029/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delete_delta_029_029/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delta_025_030/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delete_delta_025_030/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delta_050_105/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delete_delta_050_105/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delete_delta_110_110/bucket_0", 0, new byte[0]));
    AcidUtils.Directory dir = AcidUtils.getAcidState(new MockPath(fs, "mock:/tbl/part1"), conf, new ValidReaderWriteIdList("tbl:100:" + Long.MAX_VALUE + ":"));
    assertEquals("mock:/tbl/part1/base_49", dir.getBaseDirectory().toString());
    List<FileStatus> obsolete = dir.getObsolete();
    assertEquals(7, obsolete.size());
    assertEquals("mock:/tbl/part1/base_10", obsolete.get(0).getPath().toString());
    assertEquals("mock:/tbl/part1/base_5", obsolete.get(1).getPath().toString());
    assertEquals("mock:/tbl/part1/delete_delta_025_030", obsolete.get(2).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_025_030", obsolete.get(3).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_025_025", obsolete.get(4).getPath().toString());
    assertEquals("mock:/tbl/part1/delete_delta_029_029", obsolete.get(5).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_029_029", obsolete.get(6).getPath().toString());
    assertEquals(0, dir.getOriginalFiles().size());
    List<AcidUtils.ParsedDelta> deltas = dir.getCurrentDirectories();
    assertEquals(2, deltas.size());
    assertEquals("mock:/tbl/part1/delete_delta_050_105", deltas.get(0).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_050_105", deltas.get(1).getPath().toString());
// The delete_delta_110_110 should not be read because it is greater than the high watermark.
}
Also used : FileStatus(org.apache.hadoop.fs.FileStatus) Configuration(org.apache.hadoop.conf.Configuration) MockFile(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFile) MockPath(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockPath) ValidReaderWriteIdList(org.apache.hadoop.hive.common.ValidReaderWriteIdList) MockFileSystem(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFileSystem) Test(org.junit.Test)

Example 7 with ValidReaderWriteIdList

use of org.apache.hadoop.hive.common.ValidReaderWriteIdList in project hive by apache.

the class TestAcidUtils method testOriginal.

@Test
public void testOriginal() throws Exception {
    Configuration conf = new Configuration();
    MockFileSystem fs = new MockFileSystem(conf, new MockFile("mock:/tbl/part1/000000_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/000000_0" + Utilities.COPY_KEYWORD + "1", 500, new byte[0]), new MockFile("mock:/tbl/part1/000000_0" + Utilities.COPY_KEYWORD + "2", 500, new byte[0]), new MockFile("mock:/tbl/part1/000001_1", 500, new byte[0]), new MockFile("mock:/tbl/part1/000002_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/random", 500, new byte[0]), new MockFile("mock:/tbl/part1/_done", 0, new byte[0]), new MockFile("mock:/tbl/part1/subdir/000000_0", 0, new byte[0]));
    AcidUtils.Directory dir = AcidUtils.getAcidState(new MockPath(fs, "/tbl/part1"), conf, new ValidReaderWriteIdList("tbl:100:" + Long.MAX_VALUE + ":"));
    assertEquals(null, dir.getBaseDirectory());
    assertEquals(0, dir.getCurrentDirectories().size());
    assertEquals(0, dir.getObsolete().size());
    List<HdfsFileStatusWithId> result = dir.getOriginalFiles();
    assertEquals(7, result.size());
    assertEquals("mock:/tbl/part1/000000_0", result.get(0).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/000000_0" + Utilities.COPY_KEYWORD + "1", result.get(1).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/000000_0" + Utilities.COPY_KEYWORD + "2", result.get(2).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/000001_1", result.get(3).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/000002_0", result.get(4).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/random", result.get(5).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/subdir/000000_0", result.get(6).getFileStatus().getPath().toString());
}
Also used : Configuration(org.apache.hadoop.conf.Configuration) MockFile(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFile) HdfsFileStatusWithId(org.apache.hadoop.hive.shims.HadoopShims.HdfsFileStatusWithId) MockPath(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockPath) ValidReaderWriteIdList(org.apache.hadoop.hive.common.ValidReaderWriteIdList) MockFileSystem(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFileSystem) Test(org.junit.Test)

Example 8 with ValidReaderWriteIdList

use of org.apache.hadoop.hive.common.ValidReaderWriteIdList in project hive by apache.

the class TestAcidUtils method testMinorCompactedDeltaMakesInBetweenDelteDeltaObsolete.

@Test
public void testMinorCompactedDeltaMakesInBetweenDelteDeltaObsolete() throws Exception {
    // This test checks that if we have a minor compacted delta for the txn range [40,60]
    // then it will make any delete delta in that range as obsolete.
    Configuration conf = new Configuration();
    conf.setInt(HiveConf.ConfVars.HIVE_TXN_OPERATIONAL_PROPERTIES.varname, AcidUtils.AcidOperationalProperties.getDefault().toInt());
    MockFileSystem fs = new MockFileSystem(conf, new MockFile("mock:/tbl/part1/delta_40_60/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delete_delta_50_50/bucket_0", 500, new byte[0]));
    Path part = new MockPath(fs, "mock:/tbl/part1");
    AcidUtils.Directory dir = AcidUtils.getAcidState(part, conf, new ValidReaderWriteIdList("tbl:100:" + Long.MAX_VALUE + ":"));
    List<FileStatus> obsolete = dir.getObsolete();
    assertEquals(1, obsolete.size());
    assertEquals("mock:/tbl/part1/delete_delta_50_50", obsolete.get(0).getPath().toString());
    List<AcidUtils.ParsedDelta> delts = dir.getCurrentDirectories();
    assertEquals(1, delts.size());
    assertEquals("mock:/tbl/part1/delta_40_60", delts.get(0).getPath().toString());
}
Also used : Path(org.apache.hadoop.fs.Path) MockPath(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockPath) FileStatus(org.apache.hadoop.fs.FileStatus) Configuration(org.apache.hadoop.conf.Configuration) MockFile(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFile) MockPath(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockPath) ValidReaderWriteIdList(org.apache.hadoop.hive.common.ValidReaderWriteIdList) MockFileSystem(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFileSystem) Test(org.junit.Test)

Example 9 with ValidReaderWriteIdList

use of org.apache.hadoop.hive.common.ValidReaderWriteIdList in project hive by apache.

the class TestAcidUtils method testOverlapingDeltaAndDeleteDelta.

@Test
public void testOverlapingDeltaAndDeleteDelta() throws Exception {
    Configuration conf = new Configuration();
    conf.setInt(HiveConf.ConfVars.HIVE_TXN_OPERATIONAL_PROPERTIES.varname, AcidOperationalProperties.getDefault().toInt());
    MockFileSystem fs = new MockFileSystem(conf, new MockFile("mock:/tbl/part1/delta_0000063_63/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delta_000062_62/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delta_00061_61/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delete_delta_00064_64/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delta_40_60/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delete_delta_40_60/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delta_0060_60/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delta_052_55/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/delete_delta_052_55/bucket_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/base_50/bucket_0", 500, new byte[0]));
    Path part = new MockPath(fs, "mock:/tbl/part1");
    AcidUtils.Directory dir = AcidUtils.getAcidState(part, conf, new ValidReaderWriteIdList("tbl:100:" + Long.MAX_VALUE + ":"));
    assertEquals("mock:/tbl/part1/base_50", dir.getBaseDirectory().toString());
    List<FileStatus> obsolete = dir.getObsolete();
    assertEquals(3, obsolete.size());
    assertEquals("mock:/tbl/part1/delete_delta_052_55", obsolete.get(0).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_052_55", obsolete.get(1).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_0060_60", obsolete.get(2).getPath().toString());
    List<AcidUtils.ParsedDelta> delts = dir.getCurrentDirectories();
    assertEquals(6, delts.size());
    assertEquals("mock:/tbl/part1/delete_delta_40_60", delts.get(0).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_40_60", delts.get(1).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_00061_61", delts.get(2).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_000062_62", delts.get(3).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_0000063_63", delts.get(4).getPath().toString());
    assertEquals("mock:/tbl/part1/delete_delta_00064_64", delts.get(5).getPath().toString());
}
Also used : Path(org.apache.hadoop.fs.Path) MockPath(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockPath) FileStatus(org.apache.hadoop.fs.FileStatus) Configuration(org.apache.hadoop.conf.Configuration) MockFile(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFile) MockPath(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockPath) ValidReaderWriteIdList(org.apache.hadoop.hive.common.ValidReaderWriteIdList) MockFileSystem(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFileSystem) Test(org.junit.Test)

Example 10 with ValidReaderWriteIdList

use of org.apache.hadoop.hive.common.ValidReaderWriteIdList in project hive by apache.

the class TestAcidUtils method testOriginalDeltas.

@Test
public void testOriginalDeltas() throws Exception {
    Configuration conf = new Configuration();
    MockFileSystem fs = new MockFileSystem(conf, new MockFile("mock:/tbl/part1/000000_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/000001_1", 500, new byte[0]), new MockFile("mock:/tbl/part1/000002_0", 500, new byte[0]), new MockFile("mock:/tbl/part1/random", 500, new byte[0]), new MockFile("mock:/tbl/part1/_done", 0, new byte[0]), new MockFile("mock:/tbl/part1/subdir/000000_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delta_025_025/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delta_029_029/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delta_025_030/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delta_050_100/bucket_0", 0, new byte[0]), new MockFile("mock:/tbl/part1/delta_101_101/bucket_0", 0, new byte[0]));
    AcidUtils.Directory dir = AcidUtils.getAcidState(new MockPath(fs, "mock:/tbl/part1"), conf, new ValidReaderWriteIdList("tbl:100:" + Long.MAX_VALUE + ":"));
    assertEquals(null, dir.getBaseDirectory());
    List<FileStatus> obsolete = dir.getObsolete();
    assertEquals(2, obsolete.size());
    assertEquals("mock:/tbl/part1/delta_025_025", obsolete.get(0).getPath().toString());
    assertEquals("mock:/tbl/part1/delta_029_029", obsolete.get(1).getPath().toString());
    List<HdfsFileStatusWithId> result = dir.getOriginalFiles();
    assertEquals(5, result.size());
    assertEquals("mock:/tbl/part1/000000_0", result.get(0).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/000001_1", result.get(1).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/000002_0", result.get(2).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/random", result.get(3).getFileStatus().getPath().toString());
    assertEquals("mock:/tbl/part1/subdir/000000_0", result.get(4).getFileStatus().getPath().toString());
    List<AcidUtils.ParsedDelta> deltas = dir.getCurrentDirectories();
    assertEquals(2, deltas.size());
    AcidUtils.ParsedDelta delt = deltas.get(0);
    assertEquals("mock:/tbl/part1/delta_025_030", delt.getPath().toString());
    assertEquals(25, delt.getMinWriteId());
    assertEquals(30, delt.getMaxWriteId());
    delt = deltas.get(1);
    assertEquals("mock:/tbl/part1/delta_050_100", delt.getPath().toString());
    assertEquals(50, delt.getMinWriteId());
    assertEquals(100, delt.getMaxWriteId());
}
Also used : FileStatus(org.apache.hadoop.fs.FileStatus) Configuration(org.apache.hadoop.conf.Configuration) MockFile(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFile) HdfsFileStatusWithId(org.apache.hadoop.hive.shims.HadoopShims.HdfsFileStatusWithId) MockPath(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockPath) ValidReaderWriteIdList(org.apache.hadoop.hive.common.ValidReaderWriteIdList) MockFileSystem(org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFileSystem) Test(org.junit.Test)

Aggregations

ValidReaderWriteIdList (org.apache.hadoop.hive.common.ValidReaderWriteIdList)20 Configuration (org.apache.hadoop.conf.Configuration)17 Test (org.junit.Test)15 Path (org.apache.hadoop.fs.Path)13 MockFile (org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFile)12 MockFileSystem (org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockFileSystem)12 MockPath (org.apache.hadoop.hive.ql.io.orc.TestInputOutputFormat.MockPath)12 FileStatus (org.apache.hadoop.fs.FileStatus)8 FileSystem (org.apache.hadoop.fs.FileSystem)4 ValidWriteIdList (org.apache.hadoop.hive.common.ValidWriteIdList)4 RecordIdentifier (org.apache.hadoop.hive.ql.io.RecordIdentifier)4 AcidOutputFormat (org.apache.hadoop.hive.ql.io.AcidOutputFormat)3 AcidUtils (org.apache.hadoop.hive.ql.io.AcidUtils)3 ReaderKey (org.apache.hadoop.hive.ql.io.orc.OrcRawRecordMerger.ReaderKey)3 BitSet (java.util.BitSet)2 ReaderPair (org.apache.hadoop.hive.ql.io.orc.OrcRawRecordMerger.ReaderPair)2 ObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector)2 StructObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector)2 HdfsFileStatusWithId (org.apache.hadoop.hive.shims.HadoopShims.HdfsFileStatusWithId)2 OrcAcidUtils (org.apache.orc.impl.OrcAcidUtils)2