-
- Downloads
[SPARK-4203][SQL] Partition directories in random order when inserting into hive table
When doing an insert into hive table with partitions the folders written to the file system are in a random order instead of the order defined in table creation. Seems that the loadPartition method in Hive.java has a Map<String,String> parameter but expects to be called with a map that has a defined ordering such as LinkedHashMap. Working on a test but having intillij problems Author: Matthew Taylor <matthew.t@tbfe.net> Closes #3076 from tbfenet/partition_dir_order_problem and squashes the following commits: f1b9a52 [Matthew Taylor] Comment format fix bca709f [Matthew Taylor] review changes 0e50f6b [Matthew Taylor] test fix 99f1a31 [Matthew Taylor] partition ordering fix 369e618 [Matthew Taylor] partition ordering fix
Showing
- sql/hive/src/main/scala/org/apache/spark/sql/hive/execution/InsertIntoHiveTable.scala 11 additions, 2 deletions...apache/spark/sql/hive/execution/InsertIntoHiveTable.scala
- sql/hive/src/test/scala/org/apache/spark/sql/hive/InsertIntoHiveTableSuite.scala 32 additions, 2 deletions.../org/apache/spark/sql/hive/InsertIntoHiveTableSuite.scala
Please register or sign in to comment