lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-3863] Add UT for drop partition column in deltastreamer testsuite (#5727)

Browse Source
This commit is contained in:
luoyajun
2022-06-13 22:29:32 +08:00
committed by GitHub
parent e89f5627e4
commit 0d859fe58b
1 changed files with 22 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
hudi-utilities/src/test/java/org/apache/hudi/utilities/functional/TestHoodieDeltaStreamer.java
View File

@@ -2142,6 +2142,28 @@ public class TestHoodieDeltaStreamer extends HoodieDeltaStreamerTestBase {
.getCommitsTimeline()).get().getMetadata(CHECKPOINT_KEY), "def");
}
@Test
public void testDropPartitionColumns() throws Exception {
String tableBasePath = dfsBasePath + "/test_drop_partition_columns" + testNum++;
// ingest data with dropping partition columns enabled
HoodieDeltaStreamer.Config cfg = TestHelpers.makeConfig(tableBasePath, WriteOperationType.UPSERT);
cfg.configs.add(String.format("%s=%s", HoodieTableConfig.DROP_PARTITION_COLUMNS.key(), "true"));
HoodieDeltaStreamer ds = new HoodieDeltaStreamer(cfg, jsc);
ds.sync();
// assert ingest successful
TestHelpers.assertAtLeastNCommits(1, tableBasePath, dfs);
TableSchemaResolver tableSchemaResolver = new TableSchemaResolver(
HoodieTableMetaClient.builder().setBasePath(tableBasePath).setConf(dfs.getConf()).build());
// get schema from data file written in the latest commit
Schema tableSchema = tableSchemaResolver.getTableAvroSchemaFromDataFile();
assertNotNull(tableSchema);
List<String> tableFields = tableSchema.getFields().stream().map(Schema.Field::name).collect(Collectors.toList());
// now assert that the partition column is not in the target schema
assertFalse(tableFields.contains("partition_path"));
}
class TestDeltaSync extends DeltaSync {
public TestDeltaSync(HoodieDeltaStreamer.Config cfg, SparkSession sparkSession, SchemaProvider schemaProvider, TypedProperties props,