You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@kylin.apache.org by skyyws <sk...@163.com> on 2017/12/07 08:22:38 UTC
A problem about retention rate analyze
Hi guys,
I found that kylin supported retention rate analyze function, so I made some test for this function. The following SQL executed successful:
-----------------------------------------------------------------------------------------------------------------------
select city, version,
intersect_count(uuid, dt, array['20161014']) as first_day,
intersect_count(uuid, dt, array['20161015']) as second_day,
intersect_count(uuid, dt, array['20161016']) as third_day,
intersect_count(uuid, dt, array['20161014', '20161015']) as retention_oneday,
intersect_count(uuid, dt, array['20161014', '20161015', '20161016']) as retention_twoday
from visit_log
where dt in ('2016104', '20161015', '20161016')
group by city, version
-----------------------------------------------------------------------------------------------------------------------
but, other SQLs executed failed like this:
-----------------------------------------------------------------------------------------------------------------------
select city, version, intersect_count(uuid, dt, array['20161014', '20161015']) as retention_oneday
from visit_log where dt in ('2016104', '20161015',) group by city, version
-----------------------------------------------------------------------------------------------------------------------
select city, version,intersect_count(uuid, dt, array['20161014', '20161015', '20161016']) as retention_twoday
from visit_log where dt in ('2016104', '20161015', '20161016') group by city, version
-----------------------------------------------------------------------------------------------------------------------
which means I cannot use just one intersect_count UDAF in a SQL, at lease two intersect_count, is this a bug or designed to do so?
2017-12-07
skyyws
Re: Re: Re: A problem about retention rate analyze
Posted by ShaoFeng Shi <sh...@apache.org>.
Got it; Thanks for the feedback!
2017-12-08 18:26 GMT+08:00 skyyws <sk...@163.com>:
> ok, I've already created a jira about this problem:
> https://issues.apache.org/jira/browse/KYLIN-3091
>
> 2017-12-08
> ------------------------------
> skyyws
> ------------------------------
>
> *发件人：*ShaoFeng Shi <sh...@apache.org>
> *发送时间：*2017-12-08 17:52
> *主题：*Re: Re: A problem about retention rate analyze
> *收件人：*"user"<us...@kylin.apache.org>
> *抄送：*
>
> Should be a bug, could you please report a JIRA to track this? Thanks!
>
> 2017-12-08 9:08 GMT+08:00 skyyws <sk...@163.com>:
>
>> ok, my kylin version is kylin 2.0.0-hbase 0.98.8. here is the error log:
>> ------------------------------------------------------------
>> -----------------------------------------------------------
>> Caused by: java.lang.IndexOutOfBoundsException: Index: 2, Size: 2
>> at java.util.ArrayList.rangeCheck(ArrayList.java:635)
>> at java.util.ArrayList.get(ArrayList.java:411)
>> at org.apache.kylin.query.relnode.ColumnRowType.getColumnByInde
>> x(ColumnRowType.java:49)
>> at org.apache.kylin.query.relnode.OLAPAggregateRel.fillbackOpti
>> mizedColumn(OLAPAggregateRel.java:396)
>> at org.apache.kylin.query.relnode.OLAPAggregateRel.buildRewrite
>> FieldsAndMetricsColumns(OLAPAggregateRel.java:347)
>> at org.apache.kylin.query.relnode.OLAPAggregateRel.implementRew
>> rite(OLAPAggregateRel.java:283)
>> at org.apache.kylin.query.relnode.OLAPRel$RewriteImplementor.
>> visitChild(OLAPRel.java:158)
>> at org.apache.kylin.query.relnode.OLAPLimitRel.implementRewrite
>> (OLAPLimitRel.java:107)
>> at org.apache.kylin.query.relnode.OLAPRel$RewriteImplementor.
>> visitChild(OLAPRel.java:158)
>> at org.apache.kylin.query.relnode.OLAPToEnumerableConverter.imp
>> lement(OLAPToEnumerableConverter.java:100)
>> at org.apache.calcite.adapter.enumerable.EnumerableRelImplement
>> or.implementRoot(EnumerableRelImplementor.java:108)
>> at org.apache.calcite.adapter.enumerable.EnumerableInterpretabl
>> e.toBindable(EnumerableInterpretable.java:92)
>> at org.apache.calcite.prepare.CalcitePrepareImpl$CalcitePrepari
>> ngStmt.implement(CalcitePrepareImpl.java:1248)
>> at org.apache.calcite.prepare.Prepare.prepareSql(Prepare.java:
>> 306)
>> at org.apache.calcite.prepare.Prepare.prepareSql(Prepare.java:
>> 203)
>> at org.apache.calcite.prepare.CalcitePrepareImpl.prepare2_(Calc
>> itePrepareImpl.java:776)
>> at org.apache.calcite.prepare.CalcitePrepareImpl.prepare_(Calci
>> tePrepareImpl.java:632)
>> at org.apache.calcite.prepare.CalcitePrepareImpl.prepareSql(Cal
>> citePrepareImpl.java:602)
>> at org.apache.calcite.jdbc.CalciteConnectionImpl.parseQuery(Cal
>> citeConnectionImpl.java:214)
>> at org.apache.calcite.jdbc.CalciteMetaImpl.prepareAndExecute(Ca
>> lciteMetaImpl.java:595)
>> at org.apache.calcite.avatica.AvaticaConnection.prepareAndExecu
>> teInternal(AvaticaConnection.java:615)
>> at org.apache.calcite.avatica.AvaticaStatement.executeInternal(
>> AvaticaStatement.java:148)
>> ------------------------------------------------------------
>> -----------------------------------------------------------
>> 2017-12-08
>> ------------------------------
>> skyyws
>> ------------------------------
>>
>> *发件人：*ShaoFeng Shi <sh...@apache.org>
>> *发送时间：*2017-12-07 22:53
>> *主题：*Re: A problem about retention rate analyze
>> *收件人：*"user"<us...@kylin.apache.org>
>> *抄送：*
>>
>> Hi Sky,
>>
>> What's the error you got when querying just one intersect_count? Besides,
>> what's your Kylin version?
>>
>> 2017-12-07 16:22 GMT+08:00 skyyws <sk...@163.com>:
>>
>>> Hi guys,
>>> I found that kylin supported retention rate analyze function, so I made
>>> some test for this function. The following SQL executed successful:
>>> ------------------------------------------------------------
>>> -----------------------------------------------------------
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> *select city, version,intersect_count(uuid, dt, array['20161014']) as
>>> first_day,intersect_count(uuid, dt, array['20161015']) as
>>> second_day,intersect_count(uuid, dt, array['20161016']) as
>>> third_day,intersect_count(uuid, dt, array['20161014', '20161015']) as
>>> retention_oneday,intersect_count(uuid, dt, array['20161014', '20161015',
>>> '20161016']) as retention_twodayfrom visit_logwhere dt in ('2016104',
>>> '20161015', '20161016')group by city, version*
>>> ------------------------------------------------------------
>>> -----------------------------------------------------------
>>> but, other SQLs executed failed like this:
>>> ------------------------------------------------------------
>>> -----------------------------------------------------------
>>>
>>> *select city, version, intersect_count(uuid, dt, array['20161014',
>>> '20161015']) as retention_onedayfrom visit_log where dt in ('2016104',
>>> '20161015',) group by city, version*
>>> ------------------------------------------------------------
>>> -----------------------------------------------------------
>>>
>>> *select city, version,intersect_count(uuid, dt, array['20161014',
>>> '20161015', '20161016']) as retention_twodayfrom visit_log where dt in
>>> ('2016104', '20161015', '20161016') group by city, version*
>>> ------------------------------------------------------------
>>> -----------------------------------------------------------
>>> which means I cannot use just one intersect_count UDAF in a SQL, at
>>> lease two intersect_count, is this a bug or designed to do so?
>>>
>>> 2017-12-07
>>> ------------------------------
>>> skyyws
>>>
>>
>>
>>
>> --
>> Best regards,
>>
>> Shaofeng Shi 史少锋
>>
>>
>
>
> --
> Best regards,
>
> Shaofeng Shi 史少锋
>
>
--
Best regards,
Shaofeng Shi 史少锋
Re: Re: Re: A problem about retention rate analyze
Posted by skyyws <sk...@163.com>.
ok, I've already created a jira about this problem:
https://issues.apache.org/jira/browse/KYLIN-3091
2017-12-08
skyyws
发件人：ShaoFeng Shi <sh...@apache.org>
发送时间：2017-12-08 17:52
主题：Re: Re: A problem about retention rate analyze
收件人："user"<us...@kylin.apache.org>
抄送：
Should be a bug, could you please report a JIRA to track this? Thanks!
2017-12-08 9:08 GMT+08:00 skyyws <sk...@163.com>:
ok, my kylin version is kylin 2.0.0-hbase 0.98.8. here is the error log:
-----------------------------------------------------------------------------------------------------------------------
Caused by: java.lang.IndexOutOfBoundsException: Index: 2, Size: 2
at java.util.ArrayList.rangeCheck(ArrayList.java:635)
at java.util.ArrayList.get(ArrayList.java:411)
at org.apache.kylin.query.relnode.ColumnRowType.getColumnByIndex(ColumnRowType.java:49)
at org.apache.kylin.query.relnode.OLAPAggregateRel.fillbackOptimizedColumn(OLAPAggregateRel.java:396)
at org.apache.kylin.query.relnode.OLAPAggregateRel.buildRewriteFieldsAndMetricsColumns(OLAPAggregateRel.java:347)
at org.apache.kylin.query.relnode.OLAPAggregateRel.implementRewrite(OLAPAggregateRel.java:283)
at org.apache.kylin.query.relnode.OLAPRel$RewriteImplementor.visitChild(OLAPRel.java:158)
at org.apache.kylin.query.relnode.OLAPLimitRel.implementRewrite(OLAPLimitRel.java:107)
at org.apache.kylin.query.relnode.OLAPRel$RewriteImplementor.visitChild(OLAPRel.java:158)
at org.apache.kylin.query.relnode.OLAPToEnumerableConverter.implement(OLAPToEnumerableConverter.java:100)
at org.apache.calcite.adapter.enumerable.EnumerableRelImplementor.implementRoot(EnumerableRelImplementor.java:108)
at org.apache.calcite.adapter.enumerable.EnumerableInterpretable.toBindable(EnumerableInterpretable.java:92)
at org.apache.calcite.prepare.CalcitePrepareImpl$CalcitePreparingStmt.implement(CalcitePrepareImpl.java:1248)
at org.apache.calcite.prepare.Prepare.prepareSql(Prepare.java:306)
at org.apache.calcite.prepare.Prepare.prepareSql(Prepare.java:203)
at org.apache.calcite.prepare.CalcitePrepareImpl.prepare2_(CalcitePrepareImpl.java:776)
at org.apache.calcite.prepare.CalcitePrepareImpl.prepare_(CalcitePrepareImpl.java:632)
at org.apache.calcite.prepare.CalcitePrepareImpl.prepareSql(CalcitePrepareImpl.java:602)
at org.apache.calcite.jdbc.CalciteConnectionImpl.parseQuery(CalciteConnectionImpl.java:214)
at org.apache.calcite.jdbc.CalciteMetaImpl.prepareAndExecute(CalciteMetaImpl.java:595)
at org.apache.calcite.avatica.AvaticaConnection.prepareAndExecuteInternal(AvaticaConnection.java:615)
at org.apache.calcite.avatica.AvaticaStatement.executeInternal(AvaticaStatement.java:148)
-----------------------------------------------------------------------------------------------------------------------
2017-12-08
skyyws
发件人：ShaoFeng Shi <sh...@apache.org>
发送时间：2017-12-07 22:53
主题：Re: A problem about retention rate analyze
收件人："user"<us...@kylin.apache.org>
抄送：
Hi Sky,
What's the error you got when querying just one intersect_count? Besides, what's your Kylin version?
2017-12-07 16:22 GMT+08:00 skyyws <sk...@163.com>:
Hi guys,
I found that kylin supported retention rate analyze function, so I made some test for this function. The following SQL executed successful:
-----------------------------------------------------------------------------------------------------------------------
select city, version,
intersect_count(uuid, dt, array['20161014']) as first_day,
intersect_count(uuid, dt, array['20161015']) as second_day,
intersect_count(uuid, dt, array['20161016']) as third_day,
intersect_count(uuid, dt, array['20161014', '20161015']) as retention_oneday,
intersect_count(uuid, dt, array['20161014', '20161015', '20161016']) as retention_twoday
from visit_log
where dt in ('2016104', '20161015', '20161016')
group by city, version
-----------------------------------------------------------------------------------------------------------------------
but, other SQLs executed failed like this:
-----------------------------------------------------------------------------------------------------------------------
select city, version, intersect_count(uuid, dt, array['20161014', '20161015']) as retention_oneday
from visit_log where dt in ('2016104', '20161015',) group by city, version
-----------------------------------------------------------------------------------------------------------------------
select city, version,intersect_count(uuid, dt, array['20161014', '20161015', '20161016']) as retention_twoday
from visit_log where dt in ('2016104', '20161015', '20161016') group by city, version
-----------------------------------------------------------------------------------------------------------------------
which means I cannot use just one intersect_count UDAF in a SQL, at lease two intersect_count, is this a bug or designed to do so?
2017-12-07
skyyws
--
Best regards,
Shaofeng Shi 史少锋
--
Best regards,
Shaofeng Shi 史少锋
Re: Re: A problem about retention rate analyze
Posted by ShaoFeng Shi <sh...@apache.org>.
Should be a bug, could you please report a JIRA to track this? Thanks!
2017-12-08 9:08 GMT+08:00 skyyws <sk...@163.com>:
> ok, my kylin version is kylin 2.0.0-hbase 0.98.8. here is the error log:
> ------------------------------------------------------------
> -----------------------------------------------------------
> Caused by: java.lang.IndexOutOfBoundsException: Index: 2, Size: 2
> at java.util.ArrayList.rangeCheck(ArrayList.java:635)
> at java.util.ArrayList.get(ArrayList.java:411)
> at org.apache.kylin.query.relnode.ColumnRowType.getColumnByIndex(
> ColumnRowType.java:49)
> at org.apache.kylin.query.relnode.OLAPAggregateRel.
> fillbackOptimizedColumn(OLAPAggregateRel.java:396)
> at org.apache.kylin.query.relnode.OLAPAggregateRel.
> buildRewriteFieldsAndMetricsColumns(OLAPAggregateRel.java:347)
> at org.apache.kylin.query.relnode.OLAPAggregateRel.
> implementRewrite(OLAPAggregateRel.java:283)
> at org.apache.kylin.query.relnode.OLAPRel$
> RewriteImplementor.visitChild(OLAPRel.java:158)
> at org.apache.kylin.query.relnode.OLAPLimitRel.
> implementRewrite(OLAPLimitRel.java:107)
> at org.apache.kylin.query.relnode.OLAPRel$
> RewriteImplementor.visitChild(OLAPRel.java:158)
> at org.apache.kylin.query.relnode.OLAPToEnumerableConverter.
> implement(OLAPToEnumerableConverter.java:100)
> at org.apache.calcite.adapter.enumerable.EnumerableRelImplementor.
> implementRoot(EnumerableRelImplementor.java:108)
> at org.apache.calcite.adapter.enumerable.EnumerableInterpretable.
> toBindable(EnumerableInterpretable.java:92)
> at org.apache.calcite.prepare.CalcitePrepareImpl$
> CalcitePreparingStmt.implement(CalcitePrepareImpl.java:1248)
> at org.apache.calcite.prepare.Prepare.prepareSql(Prepare.java:306)
> at org.apache.calcite.prepare.Prepare.prepareSql(Prepare.java:203)
> at org.apache.calcite.prepare.CalcitePrepareImpl.prepare2_(
> CalcitePrepareImpl.java:776)
> at org.apache.calcite.prepare.CalcitePrepareImpl.prepare_(
> CalcitePrepareImpl.java:632)
> at org.apache.calcite.prepare.CalcitePrepareImpl.prepareSql(
> CalcitePrepareImpl.java:602)
> at org.apache.calcite.jdbc.CalciteConnectionImpl.parseQuery(
> CalciteConnectionImpl.java:214)
> at org.apache.calcite.jdbc.CalciteMetaImpl.prepareAndExecute(
> CalciteMetaImpl.java:595)
> at org.apache.calcite.avatica.AvaticaConnection.
> prepareAndExecuteInternal(AvaticaConnection.java:615)
> at org.apache.calcite.avatica.AvaticaStatement.executeInternal(
> AvaticaStatement.java:148)
> ------------------------------------------------------------
> -----------------------------------------------------------
> 2017-12-08
> ------------------------------
> skyyws
> ------------------------------
>
> *发件人：*ShaoFeng Shi <sh...@apache.org>
> *发送时间：*2017-12-07 22:53
> *主题：*Re: A problem about retention rate analyze
> *收件人：*"user"<us...@kylin.apache.org>
> *抄送：*
>
> Hi Sky,
>
> What's the error you got when querying just one intersect_count? Besides,
> what's your Kylin version?
>
> 2017-12-07 16:22 GMT+08:00 skyyws <sk...@163.com>:
>
>> Hi guys,
>> I found that kylin supported retention rate analyze function, so I made
>> some test for this function. The following SQL executed successful:
>> ------------------------------------------------------------
>> -----------------------------------------------------------
>>
>>
>>
>>
>>
>>
>>
>>
>> *select city, version,intersect_count(uuid, dt, array['20161014']) as
>> first_day,intersect_count(uuid, dt, array['20161015']) as
>> second_day,intersect_count(uuid, dt, array['20161016']) as
>> third_day,intersect_count(uuid, dt, array['20161014', '20161015']) as
>> retention_oneday,intersect_count(uuid, dt, array['20161014', '20161015',
>> '20161016']) as retention_twodayfrom visit_logwhere dt in ('2016104',
>> '20161015', '20161016')group by city, version*
>> ------------------------------------------------------------
>> -----------------------------------------------------------
>> but, other SQLs executed failed like this:
>> ------------------------------------------------------------
>> -----------------------------------------------------------
>>
>> *select city, version, intersect_count(uuid, dt, array['20161014',
>> '20161015']) as retention_onedayfrom visit_log where dt in ('2016104',
>> '20161015',) group by city, version*
>> ------------------------------------------------------------
>> -----------------------------------------------------------
>>
>> *select city, version,intersect_count(uuid, dt, array['20161014',
>> '20161015', '20161016']) as retention_twodayfrom visit_log where dt in
>> ('2016104', '20161015', '20161016') group by city, version*
>> ------------------------------------------------------------
>> -----------------------------------------------------------
>> which means I cannot use just one intersect_count UDAF in a SQL, at lease
>> two intersect_count, is this a bug or designed to do so?
>>
>> 2017-12-07
>> ------------------------------
>> skyyws
>>
>
>
>
> --
> Best regards,
>
> Shaofeng Shi 史少锋
>
>
--
Best regards,
Shaofeng Shi 史少锋
Re: Re: A problem about retention rate analyze
Posted by skyyws <sk...@163.com>.
ok, my kylin version is kylin 2.0.0-hbase 0.98.8. here is the error log:
-----------------------------------------------------------------------------------------------------------------------
Caused by: java.lang.IndexOutOfBoundsException: Index: 2, Size: 2
at java.util.ArrayList.rangeCheck(ArrayList.java:635)
at java.util.ArrayList.get(ArrayList.java:411)
at org.apache.kylin.query.relnode.ColumnRowType.getColumnByIndex(ColumnRowType.java:49)
at org.apache.kylin.query.relnode.OLAPAggregateRel.fillbackOptimizedColumn(OLAPAggregateRel.java:396)
at org.apache.kylin.query.relnode.OLAPAggregateRel.buildRewriteFieldsAndMetricsColumns(OLAPAggregateRel.java:347)
at org.apache.kylin.query.relnode.OLAPAggregateRel.implementRewrite(OLAPAggregateRel.java:283)
at org.apache.kylin.query.relnode.OLAPRel$RewriteImplementor.visitChild(OLAPRel.java:158)
at org.apache.kylin.query.relnode.OLAPLimitRel.implementRewrite(OLAPLimitRel.java:107)
at org.apache.kylin.query.relnode.OLAPRel$RewriteImplementor.visitChild(OLAPRel.java:158)
at org.apache.kylin.query.relnode.OLAPToEnumerableConverter.implement(OLAPToEnumerableConverter.java:100)
at org.apache.calcite.adapter.enumerable.EnumerableRelImplementor.implementRoot(EnumerableRelImplementor.java:108)
at org.apache.calcite.adapter.enumerable.EnumerableInterpretable.toBindable(EnumerableInterpretable.java:92)
at org.apache.calcite.prepare.CalcitePrepareImpl$CalcitePreparingStmt.implement(CalcitePrepareImpl.java:1248)
at org.apache.calcite.prepare.Prepare.prepareSql(Prepare.java:306)
at org.apache.calcite.prepare.Prepare.prepareSql(Prepare.java:203)
at org.apache.calcite.prepare.CalcitePrepareImpl.prepare2_(CalcitePrepareImpl.java:776)
at org.apache.calcite.prepare.CalcitePrepareImpl.prepare_(CalcitePrepareImpl.java:632)
at org.apache.calcite.prepare.CalcitePrepareImpl.prepareSql(CalcitePrepareImpl.java:602)
at org.apache.calcite.jdbc.CalciteConnectionImpl.parseQuery(CalciteConnectionImpl.java:214)
at org.apache.calcite.jdbc.CalciteMetaImpl.prepareAndExecute(CalciteMetaImpl.java:595)
at org.apache.calcite.avatica.AvaticaConnection.prepareAndExecuteInternal(AvaticaConnection.java:615)
at org.apache.calcite.avatica.AvaticaStatement.executeInternal(AvaticaStatement.java:148)
-----------------------------------------------------------------------------------------------------------------------
2017-12-08
skyyws
发件人：ShaoFeng Shi <sh...@apache.org>
发送时间：2017-12-07 22:53
主题：Re: A problem about retention rate analyze
收件人："user"<us...@kylin.apache.org>
抄送：
Hi Sky,
What's the error you got when querying just one intersect_count? Besides, what's your Kylin version?
2017-12-07 16:22 GMT+08:00 skyyws <sk...@163.com>:
Hi guys,
I found that kylin supported retention rate analyze function, so I made some test for this function. The following SQL executed successful:
-----------------------------------------------------------------------------------------------------------------------
select city, version,
intersect_count(uuid, dt, array['20161014']) as first_day,
intersect_count(uuid, dt, array['20161015']) as second_day,
intersect_count(uuid, dt, array['20161016']) as third_day,
intersect_count(uuid, dt, array['20161014', '20161015']) as retention_oneday,
intersect_count(uuid, dt, array['20161014', '20161015', '20161016']) as retention_twoday
from visit_log
where dt in ('2016104', '20161015', '20161016')
group by city, version
-----------------------------------------------------------------------------------------------------------------------
but, other SQLs executed failed like this:
-----------------------------------------------------------------------------------------------------------------------
select city, version, intersect_count(uuid, dt, array['20161014', '20161015']) as retention_oneday
from visit_log where dt in ('2016104', '20161015',) group by city, version
-----------------------------------------------------------------------------------------------------------------------
select city, version,intersect_count(uuid, dt, array['20161014', '20161015', '20161016']) as retention_twoday
from visit_log where dt in ('2016104', '20161015', '20161016') group by city, version
-----------------------------------------------------------------------------------------------------------------------
which means I cannot use just one intersect_count UDAF in a SQL, at lease two intersect_count, is this a bug or designed to do so?
2017-12-07
skyyws
--
Best regards,
Shaofeng Shi 史少锋
Re: A problem about retention rate analyze
Posted by ShaoFeng Shi <sh...@apache.org>.
Hi Sky,
What's the error you got when querying just one intersect_count? Besides,
what's your Kylin version?
2017-12-07 16:22 GMT+08:00 skyyws <sk...@163.com>:
> Hi guys,
> I found that kylin supported retention rate analyze function, so I made
> some test for this function. The following SQL executed successful:
> ------------------------------------------------------------
> -----------------------------------------------------------
>
>
>
>
>
>
>
>
> *select city, version,intersect_count(uuid, dt, array['20161014']) as
> first_day,intersect_count(uuid, dt, array['20161015']) as
> second_day,intersect_count(uuid, dt, array['20161016']) as
> third_day,intersect_count(uuid, dt, array['20161014', '20161015']) as
> retention_oneday,intersect_count(uuid, dt, array['20161014', '20161015',
> '20161016']) as retention_twodayfrom visit_logwhere dt in ('2016104',
> '20161015', '20161016')group by city, version*
> ------------------------------------------------------------
> -----------------------------------------------------------
> but, other SQLs executed failed like this:
> ------------------------------------------------------------
> -----------------------------------------------------------
>
> *select city, version, intersect_count(uuid, dt, array['20161014',
> '20161015']) as retention_onedayfrom visit_log where dt in ('2016104',
> '20161015',) group by city, version*
> ------------------------------------------------------------
> -----------------------------------------------------------
>
> *select city, version,intersect_count(uuid, dt, array['20161014',
> '20161015', '20161016']) as retention_twodayfrom visit_log where dt in
> ('2016104', '20161015', '20161016') group by city, version*
> ------------------------------------------------------------
> -----------------------------------------------------------
> which means I cannot use just one intersect_count UDAF in a SQL, at lease
> two intersect_count, is this a bug or designed to do so?
>
> 2017-12-07
> ------------------------------
> skyyws
>
--
Best regards,
Shaofeng Shi 史少锋
Re: A problem about retention rate analyze
Posted by Billy Liu <bi...@apache.org>.
This is not by design, could you show more exception logs?
2017-12-07 16:22 GMT+08:00 skyyws <sk...@163.com>:
> Hi guys,
> I found that kylin supported retention rate analyze function, so I made
> some test for this function. The following SQL executed successful:
> ------------------------------------------------------------
> -----------------------------------------------------------
>
>
>
>
>
>
>
>
> *select city, version,intersect_count(uuid, dt, array['20161014']) as
> first_day,intersect_count(uuid, dt, array['20161015']) as
> second_day,intersect_count(uuid, dt, array['20161016']) as
> third_day,intersect_count(uuid, dt, array['20161014', '20161015']) as
> retention_oneday,intersect_count(uuid, dt, array['20161014', '20161015',
> '20161016']) as retention_twodayfrom visit_logwhere dt in ('2016104',
> '20161015', '20161016')group by city, version*
> ------------------------------------------------------------
> -----------------------------------------------------------
> but, other SQLs executed failed like this:
> ------------------------------------------------------------
> -----------------------------------------------------------
>
> *select city, version, intersect_count(uuid, dt, array['20161014',
> '20161015']) as retention_onedayfrom visit_log where dt in ('2016104',
> '20161015',) group by city, version*
> ------------------------------------------------------------
> -----------------------------------------------------------
>
> *select city, version,intersect_count(uuid, dt, array['20161014',
> '20161015', '20161016']) as retention_twodayfrom visit_log where dt in
> ('2016104', '20161015', '20161016') group by city, version*
> ------------------------------------------------------------
> -----------------------------------------------------------
> which means I cannot use just one intersect_count UDAF in a SQL, at lease
> two intersect_count, is this a bug or designed to do so?
>
> 2017-12-07
> ------------------------------
> skyyws
>