[
https://issues.apache.org/jira/browse/HUDI-5442?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-5442:
-----------------------------
Sprint: 0.13.0 Final Sprint, 0.13.0 Final Sprint 2, 0.13.0 Final Sprint 3,
Sprint 2023-01-31 (was: 0.13.0 Final Sprint, 0.13.0 Final Sprint 2, 0.13.0
Final Sprint 3)
> Fix HiveHoodieTableFileIndex to use lazy listing
> ------------------------------------------------
>
> Key: HUDI-5442
> URL: https://issues.apache.org/jira/browse/HUDI-5442
> Project: Apache Hudi
> Issue Type: Bug
> Components: reader-core, trino-presto
> Reporter: Ethan Guo
> Assignee: Ethan Guo
> Priority: Critical
> Fix For: 0.13.0
>
>
> Currently, HiveHoodieTableFileIndex hard-codes the shouldListLazily to false,
> using eager listing only. This leads to scanning all table partitions in the
> file index, regardless of the queryPaths provided (for Trino Hive connector,
> only one partition is passed in).
> {code:java}
> public HiveHoodieTableFileIndex(HoodieEngineContext engineContext,
> HoodieTableMetaClient metaClient,
> TypedProperties configProperties,
> HoodieTableQueryType queryType,
> List<Path> queryPaths,
> Option<String> specifiedQueryInstant,
> boolean shouldIncludePendingCommits
> ) {
> super(engineContext,
> metaClient,
> configProperties,
> queryType,
> queryPaths,
> specifiedQueryInstant,
> shouldIncludePendingCommits,
> true,
> new NoopCache(),
> false);
> } {code}
> After flipping it to true for testing, the following exception is thrown.
> {code:java}
> io.trino.spi.TrinoException: Failed to parse partition column values from the
> partition-path: likely non-encoded slashes being used in partition column's
> values. You can try to work this around by switching listing mode to eager
> at
> io.trino.plugin.hive.BackgroundHiveSplitLoader$HiveSplitLoaderTask.process(BackgroundHiveSplitLoader.java:284)
> at io.trino.plugin.hive.util.ResumableTasks$1.run(ResumableTasks.java:38)
> at io.trino.$gen.Trino_392____20221217_092723_2.run(Unknown Source)
> at
> io.airlift.concurrent.BoundedExecutor.drainQueue(BoundedExecutor.java:80)
> at
> java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
> at
> java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
> at java.base/java.lang.Thread.run(Thread.java:833)
> Caused by: org.apache.hudi.exception.HoodieException: Failed to parse
> partition column values from the partition-path: likely non-encoded slashes
> being used in partition column's values. You can try to work this around by
> switching listing mode to eager
> at
> org.apache.hudi.BaseHoodieTableFileIndex.parsePartitionColumnValues(BaseHoodieTableFileIndex.java:317)
> at
> org.apache.hudi.BaseHoodieTableFileIndex.lambda$listPartitionPaths$6(BaseHoodieTableFileIndex.java:288)
> at
> java.base/java.util.stream.ReferencePipeline$3$1.accept(ReferencePipeline.java:197)
> at
> java.base/java.util.ArrayList$ArrayListSpliterator.forEachRemaining(ArrayList.java:1625)
> at
> java.base/java.util.stream.AbstractPipeline.copyInto(AbstractPipeline.java:509)
> at
> java.base/java.util.stream.AbstractPipeline.wrapAndCopyInto(AbstractPipeline.java:499)
> at
> java.base/java.util.stream.ReduceOps$ReduceOp.evaluateSequential(ReduceOps.java:921)
> at
> java.base/java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:234)
> at
> java.base/java.util.stream.ReferencePipeline.collect(ReferencePipeline.java:682)
> at
> org.apache.hudi.BaseHoodieTableFileIndex.listPartitionPaths(BaseHoodieTableFileIndex.java:291)
> at
> org.apache.hudi.BaseHoodieTableFileIndex.getAllQueryPartitionPaths(BaseHoodieTableFileIndex.java:205)
> at
> org.apache.hudi.BaseHoodieTableFileIndex.getAllInputFileSlices(BaseHoodieTableFileIndex.java:216)
> at
> org.apache.hudi.hadoop.HiveHoodieTableFileIndex.listFileSlices(HiveHoodieTableFileIndex.java:71)
> at
> org.apache.hudi.hadoop.HoodieCopyOnWriteTableInputFormat.listStatusForSnapshotMode(HoodieCopyOnWriteTableInputFormat.java:263)
> at
> org.apache.hudi.hadoop.HoodieCopyOnWriteTableInputFormat.listStatus(HoodieCopyOnWriteTableInputFormat.java:158)
> at
> org.apache.hadoop.mapred.FileInputFormat.getSplits(FileInputFormat.java:325)
> at
> org.apache.hudi.hadoop.HoodieParquetInputFormatBase.getSplits(HoodieParquetInputFormatBase.java:68)
> at
> io.trino.plugin.hive.BackgroundHiveSplitLoader.lambda$loadPartition$2(BackgroundHiveSplitLoader.java:493)
> at
> io.trino.plugin.hive.authentication.NoHdfsAuthentication.doAs(NoHdfsAuthentication.java:25)
> at io.trino.plugin.hive.HdfsEnvironment.doAs(HdfsEnvironment.java:97)
> at
> io.trino.plugin.hive.BackgroundHiveSplitLoader.loadPartition(BackgroundHiveSplitLoader.java:493)
> at
> io.trino.plugin.hive.BackgroundHiveSplitLoader.loadSplits(BackgroundHiveSplitLoader.java:353)
> at
> io.trino.plugin.hive.BackgroundHiveSplitLoader$HiveSplitLoaderTask.process(BackgroundHiveSplitLoader.java:277)
> ... 6 more {code}
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
