[
https://issues.apache.org/jira/browse/PHOENIX-1239?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14134956#comment-14134956
]
James Taylor commented on PHOENIX-1239:
---------------------------------------
Sure, that makes sense. Think about whether it's better to derive the common
base class from TraverseNoParseNodeVisitor (which by default does not descend
into child nodes) or TraverseAllParseNodeVisitor (which by default descends
into all child nodes). Maybe existing derivation from
TraverseNoParseNodeVisitor is causing some pain?
> IN clause (and some others) cannot get moved from HAVING clause to WHERE
> clause when qualified
> ----------------------------------------------------------------------------------------------
>
> Key: PHOENIX-1239
> URL: https://issues.apache.org/jira/browse/PHOENIX-1239
> Project: Phoenix
> Issue Type: Bug
> Affects Versions: 3.0.0, 4.0.0, 5.0.0
> Reporter: Maryann Xue
> Priority: Minor
> Original Estimate: 24h
> Remaining Estimate: 24h
>
> {code}
> @Test
> public void testInListHavingToWhere() throws SQLException {
> String query = "select count(1) from atable group by a_string having
> a_string in ('foo', 'bar')";
> List<Object> binds = Collections.emptyList();
> Expressions expressions = compileStatement(query,binds);
> Expression w = TestUtil.in(A_STRING,"foo","bar");
> assertEquals(w, expressions.whereClause);
> assertNull(expressions.havingClause);
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
