1
0

[HUDI-485] Corrected the check for incremental sql (#2768)

* [HUDI-485]: corrected the check for incremental sql

* [HUDI-485]: added tests

* code review comments addressed

* [HUDI-485]: added happy flow test case
This commit is contained in:
Pratyaksh Sharma
2022-01-12 08:22:07 +05:30
committed by GitHub
parent 6cdcd89afa
commit a392e9ba46
5 changed files with 177 additions and 30 deletions

View File

@@ -996,7 +996,7 @@ public class TestHiveSyncTool {
// create empty commit
final String emptyCommitTime = "200";
HiveTestUtil.createCommitFile(commitMetadata, emptyCommitTime);
HiveTestUtil.createCommitFile(commitMetadata, emptyCommitTime, hiveSyncConfig.basePath);
HoodieHiveClient hiveClient =
new HoodieHiveClient(HiveTestUtil.hiveSyncConfig, HiveTestUtil.getHiveConf(), HiveTestUtil.fileSystem);

View File

@@ -128,6 +128,14 @@ public class HiveTestUtil {
clear();
}
public static void clearIncrementalPullSetup(String path1, String path2) throws IOException, HiveException, MetaException {
fileSystem.delete(new Path(path1), true);
if (path2 != null) {
fileSystem.delete(new Path(path2), true);
}
clear();
}
public static void clear() throws IOException, HiveException, MetaException {
fileSystem.delete(new Path(hiveSyncConfig.basePath), true);
HoodieTableMetaClient.withPropertyBuilder()
@@ -159,23 +167,28 @@ public class HiveTestUtil {
}
}
public static void createCOWTable(String instantTime, int numberOfPartitions, boolean useSchemaFromCommitMetadata)
throws IOException, URISyntaxException {
Path path = new Path(hiveSyncConfig.basePath);
FileIOUtils.deleteDirectory(new File(hiveSyncConfig.basePath));
public static void createCOWTable(String instantTime, int numberOfPartitions, boolean useSchemaFromCommitMetadata,
String basePath, String databaseName, String tableName) throws IOException, URISyntaxException {
Path path = new Path(basePath);
FileIOUtils.deleteDirectory(new File(basePath));
HoodieTableMetaClient.withPropertyBuilder()
.setTableType(HoodieTableType.COPY_ON_WRITE)
.setTableName(hiveSyncConfig.tableName)
.setPayloadClass(HoodieAvroPayload.class)
.initTable(configuration, hiveSyncConfig.basePath);
.setTableType(HoodieTableType.COPY_ON_WRITE)
.setTableName(tableName)
.setPayloadClass(HoodieAvroPayload.class)
.initTable(configuration, basePath);
boolean result = fileSystem.mkdirs(path);
checkResult(result);
ZonedDateTime dateTime = ZonedDateTime.now();
HoodieCommitMetadata commitMetadata = createPartitions(numberOfPartitions, true,
useSchemaFromCommitMetadata, dateTime, instantTime);
createdTablesSet.add(hiveSyncConfig.databaseName + "." + hiveSyncConfig.tableName);
createCommitFile(commitMetadata, instantTime);
useSchemaFromCommitMetadata, dateTime, instantTime, basePath);
createdTablesSet.add(databaseName + "." + tableName);
createCommitFile(commitMetadata, instantTime, basePath);
}
public static void createCOWTable(String instantTime, int numberOfPartitions, boolean useSchemaFromCommitMetadata)
throws IOException, URISyntaxException {
createCOWTable(instantTime, numberOfPartitions, useSchemaFromCommitMetadata, hiveSyncConfig.basePath, hiveSyncConfig.databaseName, hiveSyncConfig.tableName);
}
public static void createReplaceCommit(String instantTime, String partitions, WriteOperationType type, boolean isParquetSchemaSimple, boolean useSchemaFromCommitMetadata)
@@ -220,7 +233,7 @@ public class HiveTestUtil {
writeStats.forEach(s -> commitMetadata.addWriteStat(partitionPath, s));
commitMetadata.addMetadata(HoodieCommitMetadata.SCHEMA_KEY, schema.toString());
createdTablesSet.add(hiveSyncConfig.databaseName + "." + hiveSyncConfig.tableName);
createCommitFile(commitMetadata, instantTime);
createCommitFile(commitMetadata, instantTime, hiveSyncConfig.basePath);
}
public static void createMORTable(String commitTime, String deltaCommitTime, int numberOfPartitions,
@@ -238,7 +251,7 @@ public class HiveTestUtil {
checkResult(result);
ZonedDateTime dateTime = ZonedDateTime.now();
HoodieCommitMetadata commitMetadata = createPartitions(numberOfPartitions, true,
useSchemaFromCommitMetadata, dateTime, commitTime);
useSchemaFromCommitMetadata, dateTime, commitTime, hiveSyncConfig.basePath);
createdTablesSet
.add(hiveSyncConfig.databaseName + "." + hiveSyncConfig.tableName + HiveSyncTool.SUFFIX_READ_OPTIMIZED_TABLE);
createdTablesSet
@@ -260,9 +273,9 @@ public class HiveTestUtil {
public static void addCOWPartitions(int numberOfPartitions, boolean isParquetSchemaSimple,
boolean useSchemaFromCommitMetadata, ZonedDateTime startFrom, String instantTime) throws IOException, URISyntaxException {
HoodieCommitMetadata commitMetadata =
createPartitions(numberOfPartitions, isParquetSchemaSimple, useSchemaFromCommitMetadata, startFrom, instantTime);
createPartitions(numberOfPartitions, isParquetSchemaSimple, useSchemaFromCommitMetadata, startFrom, instantTime, hiveSyncConfig.basePath);
createdTablesSet.add(hiveSyncConfig.databaseName + "." + hiveSyncConfig.tableName);
createCommitFile(commitMetadata, instantTime);
createCommitFile(commitMetadata, instantTime, hiveSyncConfig.basePath);
}
public static void addCOWPartition(String partitionPath, boolean isParquetSchemaSimple,
@@ -270,14 +283,14 @@ public class HiveTestUtil {
HoodieCommitMetadata commitMetadata =
createPartition(partitionPath, isParquetSchemaSimple, useSchemaFromCommitMetadata, instantTime);
createdTablesSet.add(hiveSyncConfig.databaseName + "." + hiveSyncConfig.tableName);
createCommitFile(commitMetadata, instantTime);
createCommitFile(commitMetadata, instantTime, hiveSyncConfig.basePath);
}
public static void addMORPartitions(int numberOfPartitions, boolean isParquetSchemaSimple, boolean isLogSchemaSimple,
boolean useSchemaFromCommitMetadata, ZonedDateTime startFrom, String instantTime, String deltaCommitTime)
throws IOException, URISyntaxException, InterruptedException {
HoodieCommitMetadata commitMetadata = createPartitions(numberOfPartitions, isParquetSchemaSimple,
useSchemaFromCommitMetadata, startFrom, instantTime);
useSchemaFromCommitMetadata, startFrom, instantTime, hiveSyncConfig.basePath);
createdTablesSet.add(hiveSyncConfig.databaseName + "." + hiveSyncConfig.tableName + HiveSyncTool.SUFFIX_READ_OPTIMIZED_TABLE);
createdTablesSet.add(hiveSyncConfig.databaseName + "." + hiveSyncConfig.tableName + HiveSyncTool.SUFFIX_SNAPSHOT_TABLE);
HoodieCommitMetadata compactionMetadata = new HoodieCommitMetadata();
@@ -312,13 +325,13 @@ public class HiveTestUtil {
}
private static HoodieCommitMetadata createPartitions(int numberOfPartitions, boolean isParquetSchemaSimple,
boolean useSchemaFromCommitMetadata, ZonedDateTime startFrom, String instantTime) throws IOException, URISyntaxException {
boolean useSchemaFromCommitMetadata, ZonedDateTime startFrom, String instantTime, String basePath) throws IOException, URISyntaxException {
startFrom = startFrom.truncatedTo(ChronoUnit.DAYS);
HoodieCommitMetadata commitMetadata = new HoodieCommitMetadata();
for (int i = 0; i < numberOfPartitions; i++) {
String partitionPath = startFrom.format(dtfOut);
Path partPath = new Path(hiveSyncConfig.basePath + "/" + partitionPath);
Path partPath = new Path(basePath + "/" + partitionPath);
fileSystem.makeQualified(partPath);
fileSystem.mkdirs(partPath);
List<HoodieWriteStat> writeStats = createTestData(partPath, isParquetSchemaSimple, instantTime);
@@ -382,7 +395,7 @@ public class HiveTestUtil {
}
private static void generateParquetDataWithSchema(Path filePath, Schema schema)
throws IOException, URISyntaxException {
throws IOException {
org.apache.parquet.schema.MessageType parquetSchema = new AvroSchemaConverter().convert(schema);
BloomFilter filter = BloomFilterFactory.createBloomFilter(1000, 0.0001, -1,
BloomFilterTypeCode.SIMPLE.name());
@@ -446,9 +459,9 @@ public class HiveTestUtil {
}
}
public static void createCommitFile(HoodieCommitMetadata commitMetadata, String instantTime) throws IOException {
public static void createCommitFile(HoodieCommitMetadata commitMetadata, String instantTime, String basePath) throws IOException {
byte[] bytes = commitMetadata.toJsonString().getBytes(StandardCharsets.UTF_8);
Path fullPath = new Path(hiveSyncConfig.basePath + "/" + HoodieTableMetaClient.METAFOLDER_NAME + "/"
Path fullPath = new Path(basePath + "/" + HoodieTableMetaClient.METAFOLDER_NAME + "/"
+ HoodieTimeline.makeCommitFileName(instantTime));
FSDataOutputStream fsout = fileSystem.create(fullPath, true);
fsout.write(bytes);
@@ -466,7 +479,7 @@ public class HiveTestUtil {
public static void createCommitFileWithSchema(HoodieCommitMetadata commitMetadata, String instantTime, boolean isSimpleSchema) throws IOException {
addSchemaToCommitMetadata(commitMetadata, isSimpleSchema, true);
createCommitFile(commitMetadata, instantTime);
createCommitFile(commitMetadata, instantTime, hiveSyncConfig.basePath);
}
private static void createCompactionCommitFile(HoodieCommitMetadata commitMetadata, String instantTime)