环境： 0.10.0
     COGROUP : 2+个关系中做分组

最近做一个PIG类的项目，重新看了下PIG，顺便做总结。看到COGROUP时感觉有些问题

官方：http://pig.apache.org/docs/r0.8.1/piglatin_ref2.html

A = load 'student' as (name:chararray, age:int, gpa:float);
B = load 'student' as (name:chararray, age:int, gpa:float);
dump B;
(joe,18,2.5)
(sam,,3.0)
(bob,,3.5)X = cogroup A by age, B by age;
dump X;
(18,{(joe,18,2.5)},{(joe,18,2.5)})
(,{(sam,,3.0),(bob,,3.5)},{})
(,{},{(sam,,3.0),(bob,,3.5)})

很难理解的是为什么age是空的会有2条数据？参考其他博客看下COGROUP的用法，发现不是这样。

编码无悔博客：http://www.codelast.com/?p=3621

后来自己造了一批数据来尝试没有空值和有空值的COGROUP.

CA:(jack,32,run)(liza,22,eat)(tom,20,mouse)CB:(jack,32,run)(jerry,10,steal)(tom,20,mouse)COGROUP:(tom,{(tom,20,mouse)},{(tom,20,mouse)})(jack,{(jack,32,run)},{(jack,32,run)})(liza,{(liza,22,eat)},{})(jerry,{},{(jerry,10,steal)})

加一个有空值的字段。

 CA:(jack,32,run)(liza,22,eat)(,20,mouse)CB：(jack,32,run)(jerry,10,steal)(,20,mouse)COGROUP：(jack,{(jack,32,run)},{(jack,32,run)})(liza,{(liza,22,eat)},{})(jerry,{},{(jerry,10,steal)})(,{(,20,mouse)},{})(,{},{(,20,mouse)})

分别加M和N个空值字段，数据量不会出现M*N个空值吧？不是2个就是个大坑，会放大数据量。

CA:(jack,32,run)(liza,22,eat)(,200,mouse)(,201,mouse)(,202,mouse)(,203,mouse)(,204,mouse)CB：(jack,32,run)(jerry,10,steal)(,301,mouse)(,302,mouse)(,303,mouse)(,304,mouse)(,305,mouse)(,306,mouse)(,307,mouse)COGROUP：(jack,{(jack,32,run)},{(jack,32,run)})(liza,{(liza,22,eat)},{})(jerry,{},{(jerry,10,steal)})(,{(,200,mouse),(,201,mouse),(,202,mouse),(,203,mouse),(,204,mouse)},{})(,{},{(,301,mouse),(,302,mouse),(,303,mouse),(,304,mouse),(,305,mouse),(,306,mouse),(,307,mouse)})

还好，是个坑，但是没有放大数据量。官方已经有人提交了patch：https://issues.apache.org/jira/browse/PIG-361