On Thu, Dec 26, 2013 at 7:03 PM, Satish Mittal <satish.mit...@inmobi.com>wrote:
> You could use HCatalog for reading/writing data into Hive metastore > tables. HCatalog provides HCatLoader/HCatStorer interface for Pig and > HCatInputFormat/HCatFormat interface for MR jobs to perform table > read/write, including support for static/dynamic partitions. Thanks for your reply. I've checked HCatalog. HCatalog is built on top of the Hive metastore and incorporates Hive's DDL. So the partitioning mechanism of Hive and HCatalog should be the same. I should have described my question more clearly and more shortly: I want to know that, whether Hive supports using a custom MapReduce script to load data into a table with static/hybrid/dynamic partitions? The partition column doesn't come from the original table, it's a completely new column generated automatically by the MapReduce script. Thanks.
