Character fields with UNICODE_FSS charset allow to enter 3x more characters than specified column length.
Test case:
create table T (C1 varchar(10) character set UNICODE_FSS);
commit;
-- All these commands will pass without error
insert into T (C1) values ('1234567890');
insert into T (C1) values ('12345678901234567890');
insert into T (C1) values ('123456789012345678901234567890');
commit;
-- Data are there
select * from T;
-- But Next insert will fail as expected
insert into T (C1) values ('1234567890123456789012345678901');
Note that many (GUI) tools will fail with transliteration/data truncation error on attempt to read the "more than expected" data rows, but ISQL will show these rows.
Submitted by: @pcisar
Character fields with UNICODE_FSS charset allow to enter 3x more characters than specified column length.
Test case:
create table T (C1 varchar(10) character set UNICODE_FSS); commit; -- All these commands will pass without error insert into T (C1) values ('1234567890'); insert into T (C1) values ('12345678901234567890'); insert into T (C1) values ('123456789012345678901234567890'); commit; -- Data are there select * from T; -- But Next insert will fail as expected insert into T (C1) values ('1234567890123456789012345678901');
Note that many (GUI) tools will fail with transliteration/data truncation error on attempt to read the "more than expected" data rows, but ISQL will show these rows.