There is currently read support for parquet files through JNI bindings on the dataset API [1]. I am not 100% sure but I believe this supports hive partitioned data but will Hive Metastore.
[1] https://github.com/apache/arrow/blob/master/java/dataset/src/main/java/org/apache/arrow/dataset/file/FileSystemDatasetFactory.java On Fri, Aug 13, 2021 at 2:33 AM jasonbsun(孙褒) <[email protected]> wrote: > Hi, > > > Is there a way for arrow Java sdk to read or write native arrow schema > data from parquet on hive ? > > > > Jason >
