Pig的COUNT问题

grunt> A = LOAD 'c.txt' AS (col1:chararray, col2:int, col3:int, col4:int, col5:double, col6:double, col7:int);
grunt> B = GROUP A BY (col2, col3, col4);
grunt> C = FOREACH B {D = DISTINCT A.col7; GENERATE group, COUNT(D);};
grunt> DUMP C;
复制代码

我主要就是过滤相指定相同字段，然后统计，但是输出都是1，我最终要统计所以的字段到底有多少天
请问是要什么方法？

这里我是要的方法是 D = GROUP C ALL;
                            E = FOREACH D GENERATE COUNT(C)

提供一些表中数据,和你要达到的效果

testFile
1 2
3 4
1 2
3 4
5 6

test.pig

grunt> 
A = LOAD 'testFile.txt' AS (x:int,y:int);
B = GROUP A BY y;
C = FOREACH B {D = DISTINCT A.y; GENERATE  COUNT(D) as k;};
D = GROUP C ALL;
E = FOREACH D GENERATE COUNT(C.k);
DUMP E;
复制代码

如果y值为相同的就是1个，最终统计出3个值

使用下面代码试试：是对col2去重，然后输出

A = LOAD '1.txt' AS (col1: chararray, col2: chararray); 
B = GROUP A BY (col2); 
C = FOREACH B { 
        D = LIMIT A 1;
       GENERATE FLATTEN(D); 
}; 
DUMP C;
复制代码

grunt> A = LOAD 'c.txt' AS (col1:chararray, col2:int, col3:int, col4:int, col5:double, col6:double, col7:int);
grunt> B = GROUP A BY (col2, col3, col4);
grunt> C = FOREACH B {D = DISTINCT A.col7; GENERATE group, COUNT(D);};
grunt> DUMP C;

把上面内容，添加如下内容

B = GROUP A BY (col2, col3, col4,col7);
C = FOREACH B GENERATE (D = DISTINCT col7; group, COUNT(D));


修改为下面形式

grunt> A = LOAD 'c.txt' AS (col1:chararray, col2:int, col3:int, col4:int, col5:double, col6:double, col7:int);
grunt> B = GROUP A BY (col2, col3, col4,col7);
C = FOREACH B GENERATE (D = DISTINCT col7; group, COUNT(D));
grunt> DUMP C;

howtodown 发表于 2014-10-21 14:30
本帖最后由 howtodown 于 2014-10-21 14:31 编辑
使用下面代码试试：是对col2去重，然后输出

恩，之前用过这段代码，但是输出的全是1原因是，和早上想问的一样，我需要做最后的统计就是
不过多谢版主

sstutu 发表于 2014-10-21 15:07
把上面内容，添加如下内容

B = GROUP A BY (col2, col3, col4,col7);
...

多谢，学习了