Using Sqoop Netezza to hive import what if I want to changed some data type ?

[#1]

I am using sqoop netezza database to Hive import.Successfully, hive-import when I check data type so some of the data type changed as Netezza data type:

Netezza----> HIVE

BIGINT---BIGINT

DATE---STRING

INT------INT

CHARACTER---STRING

NUMERIC----DOUBLE

TIMESTAMP----STRING

BUT I want (1) Timestamp in hive instead of String (2) numeric--double...this is right??

I might transfer more tables so what Can i do to keep my data type in hive instead of sqoop automatically transfer..

what is best solution for this???

[#2 · Accepted solution]

You can pas the following property --map-column-hive and override the default type https://sqoop.apache.org/docs/1.4.6/SqoopUserGuide.html#_controlling_type_mapping

[#3]

Thanks Artem, How can I specify in sqoop query in one command

sqoop import --connect jdbc:netezza://HOSTNAME/dbNAME --username username --password password --query 'select * from db.schema.tablename WHERE $CONDITIONS' --hive-import --target-dir /user/hive/test1 --hive-table dbname.tablename --split-by columnname

if I am using this query I can get data type numeric to double and timestamp--string I want to changed this data type in hive as well.. So can you suggest me what my sqoop query looks like?

[#4]

@mike pal please read the instructions I provided, there's an example for --map-column-java but same should apply to --map-column-hive

sqoop import ... --map-column-java id=String,value=Integer

[#5]

Thanks Artem ..really apreciated

[#6]

@mike pal no problem, let me know if that works for you and accept the answer if you're satisfied.