Error Unicode Character as Primary Key

[gloonie]

Hi,

I am encountering a problem with unicode in PostgreSQL 7.3.2 in creating a primary key consisting of of a one-character Unicode column. Definitions work, but when I try to INSERT rows with values higher than u'FB50, PostgreSQL starts rejecting the rows as Duplicates, even though the values are different. Is this a bug in PostgreSQL? It doesn't matter if I define the column as "text" or char(1) - the rows are still rejected. columns with more than 1 character in the value work fine.

Definition:
create table xp_char (
unicode_char char(1) not null,
descr text not null,
constraint xpchar_pkey primary key (unicode_char)
);

Sample inserts:

INSERT INTO xp_char (unicode_char_id, descr, unicode_char)
VALUES ('ۻ','ARABIC LETTER DAD WITH DOT BELOW');
INSERT INTO xp_char (unicode_char, descr)
VALUES ('ۼ',ARABIC LETTER GHAIN WITH DOT BELOW');




Has anyone else had this problem?

TIA,

Glenn