We haven't yet added support for Hive's Map type.  Can we work together on
adding this?  Drill doesn't distinguish between maps and structs given its
support for schemaless data.  If you could post a small example piece of
data, maybe we could figure out the best way to work together to add this
functionality.  As I said, it is mostly just a metadata mapping exercise
since Drill already has complex type support in the execution engine.  You
can see how it works by looking at the JSONReader complex Parquet reader.

On Wed, Apr 1, 2015 at 7:54 AM, Sudhir Rama Rao <[email protected]> wrote:

> Drill Version : 0.8.0
>
> My Hive table is a Single column table with Map data type
>
> CREATE EXTERNAL TABLE `serverside`(
>   `record` map<string,string>)
> PARTITIONED BY (
>   `year` int,
>   `month` int,
>   `day` int,
>   `hour` int)
> ROW FORMAT DELIMITED
>   FIELDS TERMINATED BY '\t'
>   COLLECTION ITEMS TERMINATED BY '&'
>   MAP KEYS TERMINATED BY '='
> STORED AS INPUTFORMAT
>   'org.apache.hadoop.mapred.SequenceFileInputFormat'
> OUTPUTFORMAT
>   'org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat'
> LOCATION hdfs://'.....
>
>
> When i query the table via Drill I am getting the following error
>
> Query failed: RemoteRpcException: Failure while running fragment.,
> Unsupported Hive data type MAP.
>
> Following Hive data types are supported in Drill for querying: BOOLEAN,
> BYTE, SHORT, INT, LONG, FLOAT, DOUBLE, DATE, TIMESTAMP, BINARY, DECIMAL,
> STRING, and VARCHAR [ 9e8f3d2a-c737-4e63-be50-068a0e56df19 on
> dtee1edwdataservice003-6353.lvs01.dev.ebayc3.com:31010 ]
>
> [ 9e8f3d2a-c737-4e63-be50-068a0e56df19 on
> dtee1edwdataservice003-6353.lvs01.dev.ebayc3.com:31010 ]
>

Reply via email to