You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by GitBox <gi...@apache.org> on 2021/08/12 03:14:45 UTC
[GitHub] [flink] lirui-apache commented on a change in pull request #16430: [FLINK-23032][hive] Refactor HiveSource to make it usable in data str…
lirui-apache commented on a change in pull request #16430:
URL: https://github.com/apache/flink/pull/16430#discussion_r687337673
##########
File path: flink-connectors/flink-connector-hive/src/main/java/org/apache/flink/connectors/hive/HiveTableSource.java
##########
@@ -133,66 +123,22 @@ public boolean isBounded() {
@VisibleForTesting
protected DataStream<RowData> getDataStream(StreamExecutionEnvironment execEnv) {
- validateScanConfigurations();
- checkAcidTable(catalogTable, tablePath);
+ HiveSourceBuilder sourceBuilder =
+ new HiveSourceBuilder(jobConf, flinkConf, tablePath, hiveVersion, catalogTable);
List<HiveTablePartition> allHivePartitions =
getAllPartitions(
jobConf,
hiveVersion,
tablePath,
- catalogTable,
- hiveShim,
+ catalogTable.getPartitionKeys(),
remainingPartitions);
- Configuration configuration = Configuration.fromMap(catalogTable.getOptions());
- HiveSource.HiveSourceBuilder sourceBuilder =
- new HiveSource.HiveSourceBuilder(
- jobConf,
- tablePath,
- catalogTable,
- allHivePartitions,
- limit,
- hiveVersion,
- flinkConf.get(HiveOptions.TABLE_EXEC_HIVE_FALLBACK_MAPRED_READER),
- (RowType) getProducedDataType().getLogicalType());
- if (isStreamingSource()) {
- if (catalogTable.getPartitionKeys().isEmpty()) {
- PartitionOrder partitionOrder = configuration.get(STREAMING_SOURCE_PARTITION_ORDER);
- if (partitionOrder != PartitionOrder.CREATE_TIME) {
- throw new UnsupportedOperationException(
- "Only '"
- + PartitionOrder.CREATE_TIME
- + "' is supported for non partition table.");
- }
- }
-
- Duration monitorInterval =
- configuration.get(STREAMING_SOURCE_MONITOR_INTERVAL) == null
- ? DEFAULT_SCAN_MONITOR_INTERVAL
- : configuration.get(STREAMING_SOURCE_MONITOR_INTERVAL);
- sourceBuilder.monitorContinuously(monitorInterval);
-
- if (!catalogTable.getPartitionKeys().isEmpty()) {
- sourceBuilder.setFetcher(new HiveContinuousPartitionFetcher());
- final String defaultPartitionName =
- jobConf.get(
- HiveConf.ConfVars.DEFAULTPARTITIONNAME.varname,
- HiveConf.ConfVars.DEFAULTPARTITIONNAME.defaultStrVal);
- HiveContinuousPartitionFetcherContext<?> fetcherContext =
- new HiveContinuousPartitionFetcherContext(
- tablePath,
- hiveShim,
- new JobConfWrapper(jobConf),
- catalogTable.getPartitionKeys(),
- getTableSchema().getFieldDataTypes(),
- getTableSchema().getFieldNames(),
- configuration,
- defaultPartitionName);
- sourceBuilder.setFetcherContext(fetcherContext);
- }
- }
+ sourceBuilder
+ .setPartitions(allHivePartitions)
Review comment:
You're right that we don't need to call `setPartitions` in streaming mode. But we still need to call it in batch mode. The `allHivePartitions` may not really be **all** partitions, because a partition pruning may have been applied.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@flink.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org