garyli1019 commented on a change in pull request #1848:
URL: https://github.com/apache/hudi/pull/1848#discussion_r466829856
##########
File path: hudi-spark/src/main/scala/org/apache/hudi/DefaultSource.scala
##########
@@ -132,11 +132,15 @@ class DefaultSource extends RelationProvider
log.info("Constructing hoodie (as parquet) data source with options :" +
optParams)
// simply return as a regular parquet relation
- DataSource.apply(
+ val relation = DataSource.apply(
sparkSession = sqlContext.sparkSession,
userSpecifiedSchema = Option(schema),
className = "parquet",
options = optParams)
.resolveRelation()
+
+
sqlContext.sparkContext.hadoopConfiguration.unset("mapreduce.input.pathFilter.class")
Review comment:
yes make sense. We unset this before the incremental query. When we only
have two data source query type, it's fine to unset before running incremental
query, but right now we have to unset this here. Still wondering why the local
build is able to pass though...
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
