[
https://issues.apache.org/jira/browse/DERBY-6378?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dag H. Wanvik updated DERBY-6378:
---------------------------------
Attachment: derby-6378-2.diff
derby-6378-2.status
Uploading a second patch for this issue, which introduced a stack mechanism for
order by and offset/fetch clauses, factored out into
ResultSetNode#QueryExpressionClause (nested class). From the javadoc for that
class
{panel}
For ease of handling order by clause, offset clause and fetch first/next
clauses in subclasses. Clauses on the same nesting level have the same index in
the lists, so at any level, any of the lists' elements may be empty. For
example, (select * from t order by a fetch next 5 rows only) order by b would
have
{noformat}
obl[0] = "order by a",
offset[0] = null,
fetchFirst[0] = "next 5 rows"
{noformat}
and
{noformat}
obl[0} = "order by b",
offset[1] = null
fetchFirst[1] = null
{noformat}
When starting a new level, always remember to do a push() before adding clauses
via setOffset(org.apache.derby.impl.sql.compile.ValueNode),
setFetchFirst(org.apache.derby.impl.sql.compile.ValueNode),
setOrderByList(org.apache.derby.impl.sql.compile.OrderByList) and
setHasJDBCLimitClause(java.lang.Boolean).
{panel}
Added tests to OrderByAndOffsetFetchInSubqueries in #testDerby6378, running
regressions.
> OFFSET/FETCH NEXT ignored when query is enclosed in parentheses
> ---------------------------------------------------------------
>
> Key: DERBY-6378
> URL: https://issues.apache.org/jira/browse/DERBY-6378
> Project: Derby
> Issue Type: Bug
> Components: SQL
> Affects Versions: 10.10.1.1
> Reporter: Knut Anders Hatlen
> Assignee: Dag H. Wanvik
> Attachments: derby-6378-1.diff, derby-6378-2.diff, derby-6378-2.status
>
>
> ij version 10.10
> ij> connect 'jdbc:derby:memory:db;create=true';
> ij> create table t(x int);
> 0 rows inserted/updated/deleted
> ij> insert into t values 1,2,3;
> 3 rows inserted/updated/deleted
> ij> select * from t offset 1 row fetch first 1 row only;
> X
> -----------
> 2
> 1 row selected
> ij> (select * from t offset 1 row fetch first 1 row only);
> X
> -----------
> 1
> 2
> 3
> 3 rows selected
> I would have expected that both of the queries had returned a single row.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
