[ https://issues.jboss.org/browse/TEIID-1616?page=com.atlassian.jira.plugin... ] Ramesh Reddy commented on TEIID-1616: ------------------------------------- When the source query is formed like this SELECT v_0.c_1 AS c_0 FROM (SELECT g_1.INTNUM AS c_0, cast(g_1.DOUBLENUM AS double precision) AS c_1, 'SmallA' AS c_2 FROM smalla AS g_1 UNION SELECT g_0.INTNUM AS c_0, cast(g_0.DOUBLENUM AS double precision) AS c_1, 'SmallB' AS c_2 FROM smallb AS g_0) AS v_0 ORDER BY c_0 Since the alias on outer query uses same aliases as the inner query (with different data types) Teradata gets confused and does not sort it correctly. One option mentioned by Steve is to turn off the "in-line" view support, however that is one of critical capability that is sought after by a user. Is there any way Teradata translator can re-write the aliases with unique names? -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/TEIID-1616?page=com.atlassian.jira.plugin... ] Steven Hawkins commented on TEIID-1616: --------------------------------------- Remind me again, why is it that setQueryOrderBy support is disabled? It seems like their issue here and in several other instances is that their view removal logic stinks, that is why I simply wanted to disable inline view support. Perhaps if I update the planner's aggregate decomposition, then could we just disable view support? -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/TEIID-1616?page=com.atlassian.jira.plugin... ] Steven Hawkins resolved TEIID-1616. ----------------------------------- Resolution: Done switched teradata to use positional ordering and corrected ColumnReference.getName -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira