[
https://issues.apache.org/jira/browse/KYLIN-3660?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16671200#comment-16671200
]
jiahua.hu commented on KYLIN-3660:
----------------------------------
However, there will be such a situation. T1 is my full data table, then I need
to filter the data in t1 according to the conditions. As mentioned above, this
is limited by a subset of the t2 table. In the current situation, I need to
build a model for each situation. But the establishment process of each model
is the same, the difference is only need to associate different sub-tables
extracted from t2, or view. Therefore, I have a suggestion, whether kylin can
support this feature in the follow-up.
但是，会有这样的情况。t1是我的全量数据表，然后我需要根据条件来过滤t1中的数据。如上面所述，此处是通过t2表的子集来限制。在当前情况下，我需要取针对每一个情况建立model。但是每一个model的建立过程都是一样的，区别只是需要关联不同的从t2中提取出来的子表，或者是view。因此，我有个建议，kylin在后续中能否支持这一特性。
some case like:
mode1 select * from t1 join t2 on t1.user_id = t2.user_id; t2: create
view t2 as select * from tt where exid='50';
mode2 select * from t1 join t3 on t1.user_id = t3.user_id; t3: create
view t3 as select * from tt where exid='51';
moden select * from t1 join tn on t1.user_id = tn.user_id; tn: create
view tn as select * from tt where exid='52';
> subquery, join wiht two cube return No Result
> ---------------------------------------------
>
> Key: KYLIN-3660
> URL: https://issues.apache.org/jira/browse/KYLIN-3660
> Project: Kylin
> Issue Type: Bug
> Components: Query Engine
> Affects Versions: v2.5.0
> Reporter: jiahua.hu
> Priority: Major
>
> I have two cubes corresponding to two tables t1 and t2. When I pass the
> user_Id in t2 to limit the query range of the t1 table, I return an empty
> result. Then there is data in my t2 table.
> 我有两个cube分别对应两张表t1和t2。当我通过t2中的user_Id去限制t1表的查询范围时，返回空结果。然后我的t2表里是有数据的。
> select * from t1 inner join (select user_id from t2 where EXID='50') t2 on
> t1.user_id = t2.user_id;
> select * from t1 where user_id in (select user_id from t2 where EXID='50');
> select * from t1 where user_id in (select '1002001');
> The above three statements execution results are returned No Result.
>
> But the following statement execution can return results
> select * from t1 where user_id in ('1002001');
>
> =======================================================
> Query Id: 4a12730c-5222-b4a9-43fc-282e6bea7b51
> SQL: select * from t1 inner join (select user_id from t2 where EXID='50') t2
> on t1.user_id = t2.user_id;
> User: ADMIN
> Success: true
> Duration: 0.001
> Project: test_project
> Realization Names: [CUBE[name=t1_CUBE], CUBE[name=t2_CUBE]]
> Cuboid Ids: [3, 63]
> Total scan count: 50123
> Total scan bytes: 5412669
> Result row count: 0
> Accept Partial: true
> Is Partial Result: false
> Hit Exception Cache: false
> Storage cache used: true
> Is Query Push-Down: false
> Is Prepare: false
> Trace URL: null
> Message: null
> ============================================================
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
