Hive sql 使用group by 字段被限制使用 collect_set/collect_list处理

hive> SELECT userid ,Sequnce ,ActionTime FROM T_BZ_ClientActionLog GROUP BY Sequnce ,ActionTime limit 100;

FAILED: SemanticException [Error 10025]: Line 1:7 Expression not in GROUP BY key 'userid'

userid被要求也处在group by分组字段里面。

这个不同于mysql语句，mysql这样写是没有问题的。

以下是处理方法；

hive> SELECT sequnce,actiontime,collect_set(pagecode),collect_set(actioncode) FROM T_BZ GROUP BY Sequnce ,ActionTime limit 100;

Total jobs = 1

Launching Job 1 out of 1

Number of reduce tasks not specified. Estimated from input data size: 1

In order to change the average load for a reducer (in bytes):

set hive.exec.reducers.bytes.per.reducer=<number>

In order to limit the maximum number of reducers:

set hive.exec.reducers.max=<number>

In order to set a constant number of reducers:

set mapreduce.job.reduces=<number>

Starting Job = job_1407387657227_0043, Tracking URL = http://n1.hadoop:8089/proxy/application_1407387657227_0043/
Kill Command = /app/prog/hadoop/bin/hadoop job -kill job_1407387657227_0043

Hadoop job information for Stage-1: number of mappers: 1; number of reducers: 1

2014-08-07 20:07:12,881 Stage-1 map = 0%, reduce = 0%

2014-08-07 20:07:24,192 Stage-1 map = 100%, reduce = 0%, Cumulative CPU 18.84 sec

2014-08-07 20:07:29,347 Stage-1 map = 100%, reduce = 100%, Cumulative CPU 20.71 sec

MapReduce Total cumulative CPU time: 20 seconds 710 msec

Ended Job = job_1407387657227_0043

MapReduce Jobs Launched:

Job 0: Map: 1 Reduce: 1 Cumulative CPU: 20.71 sec HDFS Read: 96397668 HDFS Write: 6969 SUCCESS

Total MapReduce CPU Time Spent: 20 seconds 710 msec

OK

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:20:33 [] ["A0001"]

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:20:37 ["P001"] ["A0001"]

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:20:45 ["P003","P001"] ["A0002","A0001"]

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:21:07 ["P003"] ["A0011"]

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:21:11 ["P003","P001"] ["A0017","A0001"]

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:21:13 ["P001","P002"] ["A0003","A0001"]

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:21:22 ["P002"] ["A0006"]

可以看到结果的一个集合。

当然如果不想得到集合，可以这样写，获取集合的第一个元素：：

hive> SELECT sequnce,actiontime,collect_set(pagecode)[0],collect_set(actioncode)[0] FROM T_BZ GROUP BY Sequnce ,ActionTime limit 100;

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:20:33 A0001

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:20:37 P001 A0001

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:20:45 P003 A0002

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:21:07 P003 A0011

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:21:11 P003 A0017

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:21:13 P001 A0003

00015a21-ef6d-4f05-b04e-ffd98fab2922 2014-07-24 01:21:22 P002 A0006

这样的结果就和mysql一致了。

当然如果不想去重还可以使用collect_list处理，这两个函数是HIVE的UDF函数。