lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-2883] Refactor hive sync tool / config to use reflection and standardize configs (#4175)

Browse Source 
- Refactor hive sync tool / config to use reflection and standardize configs

Co-authored-by: sivabalan <n.siva.b@gmail.com>
Co-authored-by: Rajesh Mahindra <rmahindra@Rajeshs-MacBook-Pro.local>
Co-authored-by: Raymond Xu <2701446+xushiyan@users.noreply.github.com>
This commit is contained in:
Rajesh Mahindra
2022-03-21 19:56:31 -07:00
committed by GitHub
parent 9b6e138af2
commit 5f570ea151
43 changed files with 1521 additions and 1217 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
hudi-spark-datasource/hudi-spark/src/test/java/HoodieJavaApp.java
View File

@@ -24,6 +24,7 @@ import org.apache.hudi.common.model.HoodieTableType;
import org.apache.hudi.common.testutils.HoodieTestDataGenerator;
import org.apache.hudi.config.HoodieCompactionConfig;
import org.apache.hudi.config.HoodieWriteConfig;
import org.apache.hudi.hive.HiveSyncConfig;
import org.apache.hudi.hive.MultiPartKeysValueExtractor;
import org.apache.hudi.hive.NonPartitionedExtractor;
import org.apache.hudi.hive.SlashEncodedDayPartitionValueExtractor;
@@ -255,24 +256,24 @@ public class HoodieJavaApp {
private DataFrameWriter<Row> updateHiveSyncConfig(DataFrameWriter<Row> writer) {
if (enableHiveSync) {
LOG.info("Enabling Hive sync to " + hiveJdbcUrl);
writer = writer.option(DataSourceWriteOptions.HIVE_TABLE().key(), hiveTable)
.option(DataSourceWriteOptions.HIVE_DATABASE().key(), hiveDB)
.option(DataSourceWriteOptions.HIVE_URL().key(), hiveJdbcUrl)
.option(DataSourceWriteOptions.HIVE_USER().key(), hiveUser)
.option(DataSourceWriteOptions.HIVE_PASS().key(), hivePass)
.option(DataSourceWriteOptions.HIVE_SYNC_ENABLED().key(), "true");
writer = writer.option(HiveSyncConfig.META_SYNC_TABLE_NAME.key(), hiveTable)
.option(HiveSyncConfig.META_SYNC_DATABASE_NAME.key(), hiveDB)
.option(HiveSyncConfig.HIVE_URL.key(), hiveJdbcUrl)
.option(HiveSyncConfig.HIVE_USER.key(), hiveUser)
.option(HiveSyncConfig.HIVE_PASS.key(), hivePass)
.option(HiveSyncConfig.HIVE_SYNC_ENABLED.key(), "true");
if (nonPartitionedTable) {
writer = writer
.option(DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS().key(),
.option(HiveSyncConfig.META_SYNC_PARTITION_EXTRACTOR_CLASS.key(),
NonPartitionedExtractor.class.getCanonicalName())
.option(DataSourceWriteOptions.PARTITIONPATH_FIELD().key(), "");
} else if (useMultiPartitionKeys) {
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS().key(), "year,month,day").option(
DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS().key(),
writer = writer.option(HiveSyncConfig.META_SYNC_PARTITION_FIELDS.key(), "year,month,day").option(
HiveSyncConfig.META_SYNC_PARTITION_EXTRACTOR_CLASS.key(),
MultiPartKeysValueExtractor.class.getCanonicalName());
} else {
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS().key(), "dateStr").option(
DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS().key(),
writer = writer.option(HiveSyncConfig.META_SYNC_PARTITION_FIELDS.key(), "dateStr").option(
HiveSyncConfig.META_SYNC_PARTITION_EXTRACTOR_CLASS.key(),
SlashEncodedDayPartitionValueExtractor.class.getCanonicalName());
}
}

23
hudi-spark-datasource/hudi-spark/src/test/java/HoodieJavaGenerateApp.java
View File

@@ -23,6 +23,7 @@ import org.apache.hudi.common.model.HoodieTableType;
import org.apache.hudi.common.table.timeline.HoodieActiveTimeline;
import org.apache.hudi.common.testutils.HoodieTestDataGenerator;
import org.apache.hudi.config.HoodieWriteConfig;
import org.apache.hudi.hive.HiveSyncConfig;
import org.apache.hudi.hive.MultiPartKeysValueExtractor;
import org.apache.hudi.hive.NonPartitionedExtractor;
import org.apache.hudi.hive.SlashEncodedDayPartitionValueExtractor;
@@ -125,24 +126,24 @@ public class HoodieJavaGenerateApp {
private DataFrameWriter<Row> updateHiveSyncConfig(DataFrameWriter<Row> writer) {
if (enableHiveSync) {
LOG.info("Enabling Hive sync to " + hiveJdbcUrl);
writer = writer.option(DataSourceWriteOptions.HIVE_TABLE().key(), hiveTable)
.option(DataSourceWriteOptions.HIVE_DATABASE().key(), hiveDB)
.option(DataSourceWriteOptions.HIVE_URL().key(), hiveJdbcUrl)
.option(DataSourceWriteOptions.HIVE_USER().key(), hiveUser)
.option(DataSourceWriteOptions.HIVE_PASS().key(), hivePass)
.option(DataSourceWriteOptions.HIVE_SYNC_ENABLED().key(), "true");
writer = writer.option(HiveSyncConfig.META_SYNC_TABLE_NAME.key(), hiveTable)
.option(HiveSyncConfig.META_SYNC_DATABASE_NAME.key(), hiveDB)
.option(HiveSyncConfig.HIVE_URL.key(), hiveJdbcUrl)
.option(HiveSyncConfig.HIVE_USER.key(), hiveUser)
.option(HiveSyncConfig.HIVE_PASS.key(), hivePass)
.option(HiveSyncConfig.HIVE_SYNC_ENABLED.key(), "true");
if (nonPartitionedTable) {
writer = writer
.option(DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS().key(),
.option(HiveSyncConfig.META_SYNC_PARTITION_EXTRACTOR_CLASS.key(),
NonPartitionedExtractor.class.getCanonicalName())
.option(DataSourceWriteOptions.PARTITIONPATH_FIELD().key(), "");
} else if (useMultiPartitionKeys) {
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS().key(), "year,month,day").option(
DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS().key(),
writer = writer.option(HiveSyncConfig.META_SYNC_PARTITION_FIELDS.key(), "year,month,day").option(
HiveSyncConfig.META_SYNC_PARTITION_EXTRACTOR_CLASS.key(),
MultiPartKeysValueExtractor.class.getCanonicalName());
} else {
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS().key(), "dateStr").option(
DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS().key(),
writer = writer.option(HiveSyncConfig.META_SYNC_PARTITION_FIELDS.key(), "dateStr").option(
HiveSyncConfig.META_SYNC_PARTITION_EXTRACTOR_CLASS.key(),
SlashEncodedDayPartitionValueExtractor.class.getCanonicalName());
}
}

21
hudi-spark-datasource/hudi-spark/src/test/java/HoodieJavaStreamingApp.java
View File

@@ -27,6 +27,7 @@ import org.apache.hudi.common.util.ValidationUtils;
import org.apache.hudi.config.HoodieCompactionConfig;
import org.apache.hudi.config.HoodieWriteConfig;
import org.apache.hudi.exception.TableNotFoundException;
import org.apache.hudi.hive.HiveSyncConfig;
import org.apache.hudi.hive.MultiPartKeysValueExtractor;
import org.apache.hudi.hive.SlashEncodedDayPartitionValueExtractor;
@@ -382,19 +383,19 @@ public class HoodieJavaStreamingApp {
private DataStreamWriter<Row> updateHiveSyncConfig(DataStreamWriter<Row> writer) {
if (enableHiveSync) {
LOG.info("Enabling Hive sync to " + hiveJdbcUrl);
writer = writer.option(DataSourceWriteOptions.HIVE_TABLE().key(), hiveTable)
.option(DataSourceWriteOptions.HIVE_DATABASE().key(), hiveDB)
.option(DataSourceWriteOptions.HIVE_URL().key(), hiveJdbcUrl)
.option(DataSourceWriteOptions.HIVE_USER().key(), hiveUser)
.option(DataSourceWriteOptions.HIVE_PASS().key(), hivePass)
.option(DataSourceWriteOptions.HIVE_SYNC_ENABLED().key(), "true");
writer = writer.option(HiveSyncConfig.META_SYNC_TABLE_NAME.key(), hiveTable)
.option(HiveSyncConfig.META_SYNC_DATABASE_NAME.key(), hiveDB)
.option(HiveSyncConfig.HIVE_URL.key(), hiveJdbcUrl)
.option(HiveSyncConfig.HIVE_USER.key(), hiveUser)
.option(HiveSyncConfig.HIVE_PASS.key(), hivePass)
.option(HiveSyncConfig.HIVE_SYNC_ENABLED.key(), "true");
if (useMultiPartitionKeys) {
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS().key(), "year,month,day").option(
DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS().key(),
writer = writer.option(HiveSyncConfig.META_SYNC_PARTITION_FIELDS.key(), "year,month,day").option(
HiveSyncConfig.META_SYNC_PARTITION_EXTRACTOR_CLASS.key(),
MultiPartKeysValueExtractor.class.getCanonicalName());
} else {
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS().key(), "dateStr").option(
DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS().key(),
writer = writer.option(HiveSyncConfig.META_SYNC_PARTITION_FIELDS.key(), "dateStr").option(
HiveSyncConfig.META_SYNC_PARTITION_EXTRACTOR_CLASS.key(),
SlashEncodedDayPartitionValueExtractor.class.getCanonicalName());
}
}