[
https://issues.apache.org/jira/browse/DERBY-2960?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12514243
]
Mamta A. Satoor commented on DERBY-2960:
----------------------------------------
Checked in a fix for this into main (558036) and 10.3 codeline (558049). The
commit comments were as follows
"select" query failed because we were generating SQLVarchar rather than
CollatorSQLVarchar
create table alltypes (c char(10), v varchar(50));
insert into alltypes values ('duplicate', 'is duplicated');
select substr(c||v, 1, 4), count(*) from alltypes group by substr(c||v, 1, 4) ;
To fix the problem, I needed to override StringDataValue getNewVarchar() in
CollatorSQLVarchar. Without this new method, we were generating SQLVarchar in
territory based databased when the collation type was territory based.
> Group by substr() on collated database causes ERROR XJ001: Java exception:
> 'ASSERT FAILED col1.getClass() (class org.apache.derby.iapi.types.SQLVarchar)
> expected to be the same as col2.getClass() (class
> org.apache.derby.iapi.types.CollatorSQLVarchar):
> -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: DERBY-2960
> URL: https://issues.apache.org/jira/browse/DERBY-2960
> Project: Derby
> Issue Type: Bug
> Components: SQL
> Affects Versions: 10.4.0.0
> Reporter: Kathey Marsden
> Assignee: Mamta A. Satoor
> Fix For: 10.3.1.4, 10.4.0.0
>
>
> ij> create table alltypes (i int, s smallint, l bigint, c char(10), v
> varchar(50), lvc long varchar,d double precision,
> r real, dt date, t time, ts timestamp, b char(2) for bit data, bv varchar(8)
> for bit data, lbv long varchar for bit data
> , dc decimal(5,2));
> 0 rows inserted/updated/deleted
> ij> select substr(c||v, 1, 4), count(*) from alltypes group by substr(c||v,
> 1, 4) ;
> 1 |2
> ----------------
> 0 rows selected
> ij> insert into alltypes values (0, 100, 1000000, 'duplicate', 'this is
> duplicated', 'also duplicated',200.0e0, 200.0e0,
> date('1992-01-01'), time('12:30:30'), timestamp('1992-01-01 12:30:30'),
> X'12af', x'0000111100001111', X'1234', 111.11)
> ;
> 1 row inserted/updated/deleted
> ij> insert into alltypes values (0, 100, 1000000, 'duplicate', 'this is
> duplicated', 'also duplicated',200.0e0, 200.0e0,
> date('1992-01-01'), time('12:30:30'), timestamp('1992-01-01 12:30:30'),
> X'12af', x'0000111100001111', X'1234', 111.11)
> ;
> 1 row inserted/updated/deleted
> ij> select substr(c||v, 1, 4), count(*) from alltypes group by substr(c||v,
> 1, 4) ;
> ERROR XJ001: Java exception: 'ASSERT FAILED col1.getClass() (class
> org.apache.derby.iapi.types.SQLVarchar) expected to b
> e the same as col2.getClass() (class
> org.apache.derby.iapi.types.CollatorSQLVarchar):
> org.apache.derby.shared.common.san
> ity.AssertFailure'.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
