The following bug has been logged on the website:
Bug reference: 6642
Logged by: T J Carrasco
Email address: tulio.carrasco@gmail.com
PostgreSQL version: 9.1.3
Operating system: Ubuntu 12.04 LTS
Description:=20=20=20=20=20=20=20=20
In Spanish every vowel has the same precedence value even with or without
any type of accent, so: a =3D =C3=A1 =3D =C3=A4 =3D A =3D =C3=81 =3D =C3=84
With OS locale =3D lc_collate =3D lc_ctype =3D es_ES.UTF-8 then:
CREATE TABLE test(id integer PRIMARY KEY, dad text, mum text, name text);
INSERT INTO test VALUES (1, 'le=C3=B3n','valencia', 'jos=C3=A9'), (2, 'leon=
',
'mendoza', 'juan'), (3, 'le=C3=B3n', 'val=C3=A1rd', 'jose'), (4, 'le=C3=B3n=
','v=C3=A1lencia',
'jos'), (5, 'le=C3=B3n', 'mendoza', 'jua'), (6, 'leon', 'valencia', 'josie'=
), (7,
'le=C3=B3n', 'valencia', 'josie'), (8, 'leo','zara', 'juan'), (9,
'le=C3=B3n','Valencia', 'jos');
SELECT * FROM test order by dad, mum, name;=20
We obtain:
id | dad | mum | name
---+------+----------+-------
8 | leo | zara | juan
2 | leon | mendoza | juan
6 | leon | valencia | josie
5 | le=C3=B3n | mendoza | jua
3 | le=C3=B3n | val=C3=A1rd | jose
1 | le=C3=B3n | valencia | jos=C3=A9
7 | le=C3=B3n | valencia | josie
9 | le=C3=B3n | Valencia | jos
4 | le=C3=B3n | v=C3=A1lencia | jos
(9 rows)
Which doesn=C2=B4t accomplish with Spanish collation. It should be somethin=
g like
this:
id | dad | mum | name
---+------+----------+-------
8 | leo | zara | juan
2 | leon | mendoza | juan
5 | le=C3=B3n | mendoza | jua
3 | le=C3=B3n | val=C3=A1rd | jose
4 | le=C3=B3n | v=C3=A1lencia | jos
9 | le=C3=B3n | Valencia | jos
1 | le=C3=B3n | valencia | jos=C3=A9
6 | leon | valencia | josie
7 | le=C3=B3n | valencia | josie
(9 rows)
In which due dad column has the same value it uses the mum column and if
they also have the same value then it should use the name column.
Thanks in advance for your help.