hadoop - 将多个变量存储到单个文件中

Question

我有几个日志，并对它们进行一些过滤。
结果我有几个变量，每个变量都是一些统计数据。我将它们保存在不同的文件中，如下所示:

STORE (FOREACH (GROUP A ALL) GENERATE COUNT(A)) INTO '$save_dir/A_result';
STORE (FOREACH (GROUP B ALL) GENERATE COUNT(B)) INTO '$save_dir/B_result';
STORE (FOREACH (GROUP C ALL) GENERATE COUNT(C)) INTO '$save_dir/C_result';

任何这些行的结果-仅一个数字。
如何将它们联接或合并/合并为一个变量以存储到单个文件中，而不是三个不同的文件中？
我尝试了这种方法:

B_count = FOREACH (GROUP B ALL) GENERATE COUNT(B) as b1;
result_record = FOREACH (GROUP A all) GENERATE
                'my_results' AS name,
                COUNT(A) as a1,
                B_count.b1 as b1;
STORE result_record INTO '$save_dir/total';

但这是行不通的。

Answer 1

创建关系以存储单个计数，将它们全部合并到另一个关系中并将其存储到单个文件中

A_count = (FOREACH (GROUP A ALL) GENERATE COUNT(A));
B_count = (FOREACH (GROUP B ALL) GENERATE COUNT(B));
C_count = (FOREACH (GROUP C ALL) GENERATE COUNT(C));

D = UNION A_count,B_count,C_count;

STORE D INTO '$save_dir/total'