HiveUDF

This Hive UDF sample code contains 2 functions: MyUpper() and MyContains()

They are tested in Hive 0.12, 0.13 and 1.0.

To make Hive UDF work in Drill, please follow this blog:

http://www.openkb.info/2015/08/how-to-run-hive-udf-in-drill.html

a. How to build the jar

mvn package

##b. Prepare a Hive table with sample data

In Hive CLI, create a test table:

CREATE TABLE testarray(col1 string, col2 string) ROW FORMAT DELIMITED FIELDS TERMINATED BY "|";
CREATE TABLE testarray2(col1 array<string>, col2 string) ;

Put below data for testarray:

echo "abc|xyz" > test.data

Insert into testarray2:

insert into table testarray2 select array(col1,col2),col2 from testarray;

hive> select * from testarray2 ;
OK
["abc","xyz"]	xyz
Time taken: 0.06 seconds, Fetched: 1 row(s)

##c. Test UDF

ADD JAR ~/target/MyUDF-1.0.0.jar;
CREATE TEMPORARY FUNCTION MyUpper AS 'openkb.hive.udf.MyUpper'; 
CREATE TEMPORARY FUNCTION MyContains AS 'openkb.hive.udf.MyContains'; 

SELECT MyUpper(col2) FROM testarray2;
XYZ

SELECT MyContains(col1,col2) FROM testarray2;
true