Shrimp is a library of user-defined functions for simplifying the programming in Hadoop and Pig.
### NVL: NVL lets you replace null (returned as a blank) with a string in the results of a query. If expr1 is null, then NVL returns expr2. If expr1 is not null, then NVL returns expr1.DEFINE NVL com.ttech.shrimp.NVL();
-- input:
-- (1,,2)
A = LOAD '/data/smp' USING PigStorage(',') as (f1:int, f2:chararray, f3:chararray);
-- NVL(exp1, exp2)
-- NVL2(string1, value_if_null )
B = FOREACH A GENERATE NVL(f1, '-1') as f1, NVL(f2, '-1') as f2, NVL(f3, '-1') as f3;
-- output:
-- (1, -1, 2)
dump B;
Source: http://docs.oracle.com/cd/B28359_01/server.111/b28286/functions110.htm
NVL2 lets you determine the value returned by a query based on whether a specified expression is null or not null. If expr1 is not null, then NVL2 returns expr2. If expr1 is null, then NVL2 returns expr3.
DEFINE NVL2 com.ttech.shrimp.NVL2();
-- input:
-- (1,,2)
-- NVL2(string1, value_if_NOT_null, value_if_null )
A = LOAD '/data/smp' USING PigStorage(',') as (f1:int, f2:chararray, f3:chararray);
-- NVL2(exp1, exp2, exp3);
-- NVL2(string1, value_if_NOT_null, value_if_null )
B = FOREACH A GENERATE NVL2(f1, f1, '-1');
-- output:
-- (1)
dump B;
Source: http://docs.oracle.com/cd/E11882_01/server.112/e26088/functions120.htm
To be able to load Oracle, date format must be "DD-MM-YYYY HH24:MI:SS". SQOOP_DATE_FORMAT lets you to convert the format to be able to load to the Oracle. Input format must be "YYYYMMDDHH24MISS"
DEFINE SQOOP_DATE_FORMAT com.ttech.shrimp.SQOOP_DATE_FORMAT();
-- input:
-- (20130905225900)
A = LOAD '/data/smp' USING PigStorage(',') as (f1:int, f2:chararray, f3:chararray);
B = FOREACH A GENERATE SQOOP_DATE_FORMAT(f1);
-- output:
-- (2013-09-05 22:59:00)
dump B;
Here are some common tasks when working with the source code.
mvn package