[jira] [Commented] (ASTERIXDB-2960) LOJ on subquery with UNNEST for array indexes throwing type error

["ASF subversion and git services (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/ASTERIXDB-2960?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17416335#comment-17416335 ] 

ASF subversion and git services commented on ASTERIXDB-2960:
------------------------------------------------------------

Commit d74d74196ccdcf9c26219b9440f7ea0c32b0472a in asterixdb's branch refs/heads/master from ggalvizo
[ https://gitbox.apache.org/repos/asf?p=asterixdb.git;h=d74d741 ]

[ASTERIXDB-2960][IDX] LOJ array-index INLJ probe PK fix

- user model changes: no
- storage format changes: no
- interface changes: no

Type errors were being thrown for LOJ on subqueries involving UNNESTed
values for array index INLJ. This was due to a bug where we would try to
perform a DISTINCT on probe PK vars that weren't present. If we cannot
infer any probe entries from the context, then insert a UUID on top of
the probe to allow the DISTINCT to work as intended.

Change-Id: I1eccce55b8d4d2fdb47f678d89fcf957f0ab2ebe
Reviewed-on: https://asterix-gerrit.ics.uci.edu/c/asterixdb/+/13183
Tested-by: Jenkins <je...@fulliautomatix.ics.uci.edu>
Integration-Tests: Jenkins <je...@fulliautomatix.ics.uci.edu>
Reviewed-by: Dmitry Lychagin <dm...@couchbase.com>


> LOJ on subquery with UNNEST for array indexes throwing type error
> -----------------------------------------------------------------
>
>                 Key: ASTERIXDB-2960
>                 URL: https://issues.apache.org/jira/browse/ASTERIXDB-2960
>             Project: Apache AsterixDB
>          Issue Type: Bug
>          Components: IDX - Indexes
>            Reporter: Glenn Justo Galvizo
>            Assignee: Glenn Justo Galvizo
>            Priority: Major
>
> A query like this currently throws an error when trying to optimize for an LOJ INLJ:
>  
> {code:java}
> SET               `compiler.arrayindex` "true";
> DROP DATAVERSE    TestDataverse IF EXISTS;
> CREATE DATAVERSE  TestDataverse;
> USE               TestDataverse;
> CREATE TYPE       GenericType AS { _id: uuid  };
> CREATE DATASET    IndexedDataset (GenericType)
> PRIMARY KEY       _id AUTOGENERATED;
> CREATE DATASET    ProbeDataset (GenericType)
> PRIMARY KEY       _id AUTOGENERATED;
> CREATE INDEX      TestIdx
> ON                IndexedDataset (
>     UNNEST items    
>     SELECT val : bigint
> );
> FROM             ProbeDataset P
> LEFT OUTER JOIN  ( 
>     FROM IndexedDataset I   
>     UNNEST I.items II    
>     SELECT VALUE II.val
> ) AS IIV
> ON                IIV /* +indexnl */ = TO_BIGINT(P.val)
> SELECT            DISTINCT P;
> {code}
>  
>  



--
This message was sent by Atlassian Jira
(v8.3.4#803005)