Hadoop 如何在Pig中将多行合并到一行中

我需要使用Pig脚本将多个元组组合成一个元组。你能提供一些指导吗

dump requestFile;

电流输出

(Logging Transaction ID:21214,/var/log/tibco/,NESS-A-1-LPNameRequesttoNESS.log,tibcoTest log)

(Default Data:LP Name Request Message Executed Successfully)

(LoanPath Request ID:  88128640)

(RequestGroupID#:   )

(SplitCount#:  2 )

(SplitIndex:  1)

(Correlation ID : 88128640-1 )

(Logging Transaction ID:21214,/var/log/tibco/,NESS-A-1-LPNameRequesttoNESS.log,tibcoTest log,Default Data:LP Name Request Message Executed Successfully,LoanPath Request ID:  88128640,RequestGroupID#:   ,SplitCount#:  2,SplitIndex:  1)
(Correlation ID : 88128640-1 )

所需输出

(Logging Transaction ID:21214,/var/log/tibco/,NESS-A-1-LPNameRequesttoNESS.log,tibcoTest log)

(Default Data:LP Name Request Message Executed Successfully)

(LoanPath Request ID:  88128640)

(RequestGroupID#:   )

(SplitCount#:  2 )

(SplitIndex:  1)

(Correlation ID : 88128640-1 )

(Logging Transaction ID:21214,/var/log/tibco/,NESS-A-1-LPNameRequesttoNESS.log,tibcoTest log,Default Data:LP Name Request Message Executed Successfully,LoanPath Request ID:  88128640,RequestGroupID#:   ,SplitCount#:  2,SplitIndex:  1)
(Correlation ID : 88128640-1 )

那么：

requestFile = foreach requestFile generate flatten(tuple);
G = GROUP requestFile ALL;
F = FOREACH G generate requestFile;         

---

将一组行分组的标准是什么？该数据是pig中一组操作的结果吗？如果是这样的话，也许你应该在你的pig脚本中对它们进行分组。你可以尝试一个适合你需求的pig UDF（用Java编写）。