Salvatore R created TEIID-4312: ---------------------------------- Summary: INNER and LEFT joins of CTEs fail or return incorrect results Key: TEIID-4312 URL: https://issues.jboss.org/browse/TEIID-4312 Project: Teiid Issue Type: Bug Reporter: Salvatore R Assignee: Steven Hawkins I defined a table "test_a" with the same data and structure in PostgreSQL and MySQL: {code:sql} CREATE TABLE test_a(a integer, b integer); INSERT INTO test_a VALUES (1, 1); INSERT INTO test_a VALUES (1, 2); INSERT INTO test_a VALUES (2, 1); INSERT INTO test_a VALUES (2, 2); INSERT INTO test_a VALUES (3, 2); INSERT INTO test_a VALUES (3, 10); {code} The following query, based on the table in *PostgreSQL*, fails: {code:sql} with CTE1 as ( WITH CTE11 as (SELECT a from pg.test_a), CTE21 as (select t1.a from CTE11 t1 join CTE11 t2 on t1.a=t2.a), CTE31 as (select a from CTE21) SELECT CTE31.a FROM CTE21 join CTE31 on CTE31.a=CTE21.a ) select * from CTE1 {code} with this exception: {code:sql} 16:36:28,752 WARN [org.teiid.CONNECTOR] (Worker13_QueryProcessorQueue114) eYLiZgMIChSF Connector worker process failed for atomic-request=eYLiZgMIChSF.11.2.24: org.teiid.translator.jdbc.JDBCExecutionException: 0 TEIID11008:TEIID11004 Error executing statement(s): [Prepared Values: [] SQL: WITH CTE11 (a) AS (SELECT NULL FROM "public"."test_a" AS g_0), CTE21 (a) AS (SELECT g_0.a FROM CTE11 AS g_0, CTE11 AS g_1 WHERE g_0.a = g_1.a) SELECT g_1.a FROM CTE21 AS g_0, CTE21 AS g_1 WHERE g_1.a = g_0.a] at org.teiid.translator.jdbc.JDBCQueryExecution.execute(JDBCQueryExecution.java:131) at org.teiid.dqp.internal.datamgr.ConnectorWorkItem.execute(ConnectorWorkItem.java:365) at sun.reflect.GeneratedMethodAccessor94.invoke(Unknown Source) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:606) at org.teiid.dqp.internal.datamgr.ConnectorManager$1.invoke(ConnectorManager.java:211) at com.sun.proxy.$Proxy56.execute(Unknown Source) at org.teiid.dqp.internal.process.DataTierTupleSource.getResults(DataTierTupleSource.java:306) at org.teiid.dqp.internal.process.DataTierTupleSource$1.call(DataTierTupleSource.java:112) at org.teiid.dqp.internal.process.DataTierTupleSource$1.call(DataTierTupleSource.java:108) at java.util.concurrent.FutureTask.run(FutureTask.java:262) at org.teiid.dqp.internal.process.FutureWork.run(FutureWork.java:65) at org.teiid.dqp.internal.process.DQPWorkContext.runInContext(DQPWorkContext.java:276) at org.teiid.dqp.internal.process.ThreadReuseExecutor$RunnableWrapper.run(ThreadReuseExecutor.java:119) at org.teiid.dqp.internal.process.ThreadReuseExecutor$3.run(ThreadReuseExecutor.java:210) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) at java.lang.Thread.run(Thread.java:745) Caused by: org.postgresql.util.PSQLException: ERROR: failed to find conversion function from unknown to text at org.postgresql.core.v3.QueryExecutorImpl.receiveErrorResponse(QueryExecutorImpl.java:2157) at org.postgresql.core.v3.QueryExecutorImpl.processResults(QueryExecutorImpl.java:1886) at org.postgresql.core.v3.QueryExecutorImpl.execute(QueryExecutorImpl.java:255) at org.postgresql.jdbc2.AbstractJdbc2Statement.execute(AbstractJdbc2Statement.java:555) at org.postgresql.jdbc2.AbstractJdbc2Statement.executeWithFlags(AbstractJdbc2Statement.java:417) at org.postgresql.jdbc2.AbstractJdbc2Statement.executeQuery(AbstractJdbc2Statement.java:302) at org.jboss.jca.adapters.jdbc.WrappedPreparedStatement.executeQuery(WrappedPreparedStatement.java:504) at org.teiid.translator.jdbc.JDBCQueryExecution.execute(JDBCQueryExecution.java:123) ... 17 more {code} The same query based on the table in *MySQL* wrongly returns an empty result. The main differences is that the query is fully pushed down to PostgreSQL but it isn't in MySQL. Regarding wrong results, I am also experiencing a similar problem with the following query: {code:sql} with CTE1 as ( WITH alias as (SELECT a from pg.test_a), alias2 as (select t2.a as a1, t1.a from alias t1 join (SELECT 1 as a) t2 on t1.a=t2.a), CTE31 as (select t2.a as a1 from alias2 t2) SELECT CTE31.a1 FROM alias2 join CTE31 on CTE31.a1=alias2.a ), CTE2 as ( WITH alias as (SELECT 1 as a), alias2 as (select t2.a a1, t1.a from alias t1 join (SELECT 1 as a) t2 on t1.a=t2.a), CTE32 as (select t2.a from alias2 t2) SELECT CTE32.a FROM alias2 join CTE32 on CTE32.a=alias2.a ) select * from CTE1 as T1 join CTE2 as T2 on T1.a1=T2.a {code} It returns 4 rows (as expected) if based on MySQL table but it returns 16 rows if pushed down to PostgreSQL. I don't know if the two behaviors are related or not, but I can create a different ticket for the second issue, if needed. -- This message was sent by Atlassian JIRA (v6.4.11#64026)