lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-89] Add configOption & refactor all configs based on that (#2833)

Browse Source 
Co-authored-by: Wenning Ding <wenningd@amazon.com>
This commit is contained in:
wenningd
2021-06-30 14:26:30 -07:00
committed by GitHub
parent 07e93de8b4
commit d412fb2fe6
173 changed files with 4277 additions and 3309 deletions
Download Patch File Download Diff File Expand all files Collapse all files

38
hudi-spark-datasource/hudi-spark/src/test/java/HoodieJavaStreamingApp.java
View File

@@ -332,9 +332,9 @@ public class HoodieJavaStreamingApp {
* Consume incrementally, only changes in commit 2 above. Currently only supported for COPY_ON_WRITE TABLE
*/
Dataset<Row> hoodieIncViewDF = spark.read().format("hudi")
.option(DataSourceReadOptions.QUERY_TYPE_OPT_KEY(), DataSourceReadOptions.QUERY_TYPE_INCREMENTAL_OPT_VAL())
.option(DataSourceReadOptions.QUERY_TYPE_OPT_KEY().key(), DataSourceReadOptions.QUERY_TYPE_INCREMENTAL_OPT_VAL())
// Only changes in write 2 above
.option(DataSourceReadOptions.BEGIN_INSTANTTIME_OPT_KEY(), commitInstantTime1)
.option(DataSourceReadOptions.BEGIN_INSTANTTIME_OPT_KEY().key(), commitInstantTime1)
// For incremental view, pass in the root/base path of dataset
.load(tablePath);
@@ -355,14 +355,14 @@ public class HoodieJavaStreamingApp {
DataStreamWriter<Row> writer = streamingInput.writeStream().format("org.apache.hudi")
.option("hoodie.insert.shuffle.parallelism", "2").option("hoodie.upsert.shuffle.parallelism", "2")
.option("hoodie.delete.shuffle.parallelism", "2")
.option(DataSourceWriteOptions.OPERATION_OPT_KEY(), operationType)
.option(DataSourceWriteOptions.TABLE_TYPE_OPT_KEY(), tableType)
.option(DataSourceWriteOptions.RECORDKEY_FIELD_OPT_KEY(), "_row_key")
.option(DataSourceWriteOptions.PARTITIONPATH_FIELD_OPT_KEY(), "partition")
.option(DataSourceWriteOptions.PRECOMBINE_FIELD_OPT_KEY(), "timestamp")
.option(HoodieCompactionConfig.INLINE_COMPACT_NUM_DELTA_COMMITS_PROP, "1")
.option(DataSourceWriteOptions.ASYNC_COMPACT_ENABLE_OPT_KEY(), "true")
.option(HoodieWriteConfig.TABLE_NAME, tableName).option("checkpointLocation", checkpointLocation)
.option(DataSourceWriteOptions.OPERATION_OPT_KEY().key(), operationType)
.option(DataSourceWriteOptions.TABLE_TYPE_OPT_KEY().key(), tableType)
.option(DataSourceWriteOptions.RECORDKEY_FIELD_OPT_KEY().key(), "_row_key")
.option(DataSourceWriteOptions.PARTITIONPATH_FIELD_OPT_KEY().key(), "partition")
.option(DataSourceWriteOptions.PRECOMBINE_FIELD_OPT_KEY().key(), "timestamp")
.option(HoodieCompactionConfig.INLINE_COMPACT_NUM_DELTA_COMMITS_PROP.key(), "1")
.option(DataSourceWriteOptions.ASYNC_COMPACT_ENABLE_OPT_KEY().key(), "true")
.option(HoodieWriteConfig.TABLE_NAME.key(), tableName).option("checkpointLocation", checkpointLocation)
.outputMode(OutputMode.Append());
updateHiveSyncConfig(writer);
@@ -379,18 +379,18 @@ 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_OPT_KEY(), hiveTable)
.option(DataSourceWriteOptions.HIVE_DATABASE_OPT_KEY(), hiveDB)
.option(DataSourceWriteOptions.HIVE_URL_OPT_KEY(), hiveJdbcUrl)
.option(DataSourceWriteOptions.HIVE_USER_OPT_KEY(), hiveUser)
.option(DataSourceWriteOptions.HIVE_PASS_OPT_KEY(), hivePass)
.option(DataSourceWriteOptions.HIVE_SYNC_ENABLED_OPT_KEY(), "true");
writer = writer.option(DataSourceWriteOptions.HIVE_TABLE_OPT_KEY().key(), hiveTable)
.option(DataSourceWriteOptions.HIVE_DATABASE_OPT_KEY().key(), hiveDB)
.option(DataSourceWriteOptions.HIVE_URL_OPT_KEY().key(), hiveJdbcUrl)
.option(DataSourceWriteOptions.HIVE_USER_OPT_KEY().key(), hiveUser)
.option(DataSourceWriteOptions.HIVE_PASS_OPT_KEY().key(), hivePass)
.option(DataSourceWriteOptions.HIVE_SYNC_ENABLED_OPT_KEY().key(), "true");
if (useMultiPartitionKeys) {
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS_OPT_KEY(), "year,month,day").option(
DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS_OPT_KEY(),
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS_OPT_KEY().key(), "year,month,day").option(
DataSourceWriteOptions.HIVE_PARTITION_EXTRACTOR_CLASS_OPT_KEY().key(),
MultiPartKeysValueExtractor.class.getCanonicalName());
} else {
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS_OPT_KEY(), "dateStr");
writer = writer.option(DataSourceWriteOptions.HIVE_PARTITION_FIELDS_OPT_KEY().key(), "dateStr");
}
}
return writer;