[teiid-issues] [JBoss JIRA] (TEIID-4113) Impala translator - Incorrect aggregate replacement in query

[Steven Hawkins (JIRA)]

    [ https://issues.jboss.org/browse/TEIID-4113?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13185612#comment-13185612 ] 

Steven Hawkins commented on TEIID-4113:
---------------------------------------

> Should be able to produce the issue with the //Teiid query block that produces the output seen in the //Query sent to Impala block above.

Here's the abridged version of a BaseQueryTest that I'm running on master/8.13.3 with the simple query expression for the view definition, which confirms the appropriate source query:

{code}
   public void testQuery() throws Exception{
    	ImpalaExecutionFactory ief = new ImpalaExecutionFactory();
    	ief.setDatabaseVersion("2.0");
    	ief.start();
        DefaultCapabilitiesFinder finder = new DefaultCapabilitiesFinder(CapabilitiesConverter.convertCapabilities(ief));        
        HardcodedDataManager dataMgr = new HardcodedDataManager();
    
        List<?>[] expected = new List<?>[] {  };

        dataMgr.addData("SELECT g_0.the_attribute, convert((SUM(g_0.first_metric) / convert(COUNT(DISTINCT CASE WHEN g_0.second_metric >= 0 THEN g_0.second_metric END), long)), double), g_0.another_attribute FROM y.SourceTable AS g_0 WHERE g_0.a_filter_val = 111 GROUP BY g_0.the_attribute, g_0.another_attribute", //$NON-NLS-1$
                        expected);

        doProcess(RealMetadataFactory.fromDDL("create foreign table SourceTable (the_attribute string(255), first_metric long, second_metric bigdecimal, another_attribute String(255), a_filter_val integer); "
        		+ " create view VirtualTable (the_attribute string(255), first_metric long, second_metric bigdecimal, another_attribute String(255), a_filter_val integer) as (select * from SourceTable)", "x", "y"),  
                "SELECT the_attribute, cast( sum( first_metric ) / count( distinct case when second_metric >= 0 then second_metric end ) as double ) as some_alias, PARSETIMESTAMP(another_attribute, 'yyyy-MM-dd') as somedate  FROM VirtualTable WHERE a_filter_val=111 GROUP BY the_attribute, another_attribute", //$NON-NLS-1$
                finder, dataMgr, expected, DEBUG);
    }
 
}
{code}

So there must be a few more needed details.

> Impala translator - Incorrect aggregate replacement in query
> ------------------------------------------------------------
>
>                 Key: TEIID-4113
>                 URL: https://issues.jboss.org/browse/TEIID-4113
>             Project: Teiid
>          Issue Type: Bug
>          Components: JDBC Connector
>    Affects Versions: 8.13.3
>         Environment: Ubuntu Trusty
>            Reporter: Don Krapohl
>            Assignee: Steven Hawkins
>              Labels: Impala_Translator
>
> When PARSETIMESTAMP is included in a query aggregates in other columns may not be written correctly.  Example is  as below, sum(first_metric) is rewritten to anon_grp0.agg0 in the query sent to Impala.
> Given the schema:
> //source
> SourceTable
> the_attribute string(255),
> first_metric long,
> second_metric bigdecimal
> another_attribute String(255)
> //view
> VirtualTable
> the_attribute string(255),
> first_metric long,
> second_metric bigdecimal
> another_attribute String(255)
> //  Teiid query
> SELECT the_attribute,
> cast(
>   sum(
>     first_metric
>   ) 
>   / 
>   count(
>     distinct case when  second_metric >= 0 then second_metric end
>   ) as double
> )  as some_alias,
> PARSETIMESTAMP(another_attribute, 'yyyy-MM-dd') as somedate
> FROM VirtualTable 
> WHERE a_filter_val=111 
> GROUP BY the_attribute, another_attribute
> //Query sent to impala
> SELECT g_0.the_attribute AS c_0, 
> cast(
>   (
>     anon_grp0.agg0 
>     / 
>     cast(
>     anon_grp0.agg1 AS bigint
>     )
>   ) AS double
> ) AS c_1, 
> g_0.another_attribute AS c_2 
> FROM VirtualTable
> WHERE a_filter_val=111
> GROUP BY g_0.the_attribute, g_0.another_attribute



--
This message was sent by Atlassian JIRA
(v6.4.11#64026)