[ 
https://issues.apache.org/jira/browse/CASSANDRA-13547?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16036519#comment-16036519
 ] 

Krishna Dattu Koneru commented on CASSANDRA-13547:
--------------------------------------------------

Hi , I am new here and would like to work on this bug . Any help is appreciated.

I see to two problems here :

1. MV's row is not updated if the updated column in base table is not in the 
select list of the view (even if it is in where clause)
2. columns that are not in [updated columns + view PK columns] are not updated 
or do not have any data after updates.


For the first problem I see these two possible solutions :
1 - add a restriction to CQL syntax that all the columns that are used in where 
clause must be in the select list of MV 
2 - Look in where clause (of MV) too to check if update to the base table 
column should update view or not.

For the second problem, can anyone point the direction in which I should be 
looking ? 
It seems that the mutations are built correctly but some how they are not 
"applied" correctly.


> Filtered materialized views missing data
> ----------------------------------------
>
>                 Key: CASSANDRA-13547
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-13547
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Materialized Views
>         Environment: Official Cassandra 3.10 Docker image (ID 154b919bf8ce).
>            Reporter: Craig Nicholson
>            Priority: Blocker
>              Labels: materializedviews
>             Fix For: 3.11.x
>
>
> When creating a materialized view against a base table the materialized view 
> does not always reflect the correct data.
> Using the following test schema:
> {code:title=Schema|language=sql}
> DROP KEYSPACE IF EXISTS test;
> CREATE KEYSPACE test
>   WITH REPLICATION = { 
>    'class' : 'SimpleStrategy', 
>    'replication_factor' : 1 
>   };
> CREATE TABLE test.table1 (
>                 id int,
>                 name text,
>                 enabled boolean,
>                 foo text,
>                 PRIMARY KEY (id, name));
> CREATE MATERIALIZED VIEW test.table1_mv1 AS SELECT id, name, foo
>                 FROM test.table1
>                 WHERE id IS NOT NULL 
>                 AND name IS NOT NULL 
>                 AND enabled = TRUE
>                 PRIMARY KEY ((name), id);
> CREATE MATERIALIZED VIEW test.table1_mv2 AS SELECT id, name, foo, enabled
>                 FROM test.table1
>                 WHERE id IS NOT NULL 
>                 AND name IS NOT NULL 
>                 AND enabled = TRUE
>                 PRIMARY KEY ((name), id);
> {code}
> When I insert a row into the base table the materialized views are updated 
> appropriately. (+)
> {code:title=Insert row|language=sql}
> cqlsh> INSERT INTO test.table1 (id, name, enabled, foo) VALUES (1, 'One', 
> TRUE, 'Bar');
> cqlsh> SELECT * FROM test.table1;
>  id | name | enabled | foo
> ----+------+---------+-----
>   1 |  One |    True | Bar
> (1 rows)
> cqlsh> SELECT * FROM test.table1_mv1;
>  name | id | foo
> ------+----+-----
>   One |  1 | Bar
> (1 rows)
> cqlsh> SELECT * FROM test.table1_mv2;
>  name | id | enabled | foo
> ------+----+---------+-----
>   One |  1 |    True | Bar
> (1 rows)
> {code}
> Updating the record in the base table and setting enabled to FALSE will 
> filter the record from both materialized views. (+)
> {code:title=Disable the row|language=sql}
> cqlsh> UPDATE test.table1 SET enabled = FALSE WHERE id = 1 AND name = 'One';
> cqlsh> SELECT * FROM test.table1;
>  id | name | enabled | foo
> ----+------+---------+-----
>   1 |  One |   False | Bar
> (1 rows)
> cqlsh> SELECT * FROM test.table1_mv1;
>  name | id | foo
> ------+----+-----
> (0 rows)
> cqlsh> SELECT * FROM test.table1_mv2;
>  name | id | enabled | foo
> ------+----+---------+-----
> (0 rows)
> {code}
> However a further update to the base table setting enabled to TRUE should 
> include the record in both materialzed views, however only one view 
> (table1_mv2) gets updated. (-)
> It appears that only the view (table1_mv2) that returns the filtered column 
> (enabled) is updated. (-)
> Additionally columns that are not part of the partiion or clustering key are 
> not updated. You can see that the foo column has a null value in table1_mv2. 
> (-)
> {code:title=Enable the row|language=sql}
> cqlsh> UPDATE test.table1 SET enabled = TRUE WHERE id = 1 AND name = 'One';
> cqlsh> SELECT * FROM test.table1;
>  id | name | enabled | foo
> ----+------+---------+-----
>   1 |  One |    True | Bar
> (1 rows)
> cqlsh> SELECT * FROM test.table1_mv1;
>  name | id | foo
> ------+----+-----
> (0 rows)
> cqlsh> SELECT * FROM test.table1_mv2;
>  name | id | enabled | foo
> ------+----+---------+------
>   One |  1 |    True | null
> (1 rows)
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org

Reply via email to