Thread: Query with ISO caracters: wrong order?

Hello,

Our company is french based so we use ISO characters (accents) in our
data. It looks like all queries made on those characters don't sort out
OK... Especially if the first character is not standard english. The
word is put at the end, after "z" !!!

So, United States in french is "États-Unis" (hope you can see the first
letter ok!)
and it should be next to a country like "Equator" but it's not. THis is
a real problem since people using the database through PHP think it is
not there and try to add a new country! (or something).

Is there a workaround? Any suggestions?
Postgres 6.4.2
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Marc Andre Paquin
I.T. supervisor

Le Réseau Touristique INTER-Rés@
The INTER-Res@ tourism network
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

On Thursday 27 January 2000, at 16 h 9, the keyboard of Web Manager
<web@inter-resa.com> wrote:

> Our company is french based so we use ISO characters (accents) in our
> data. It looks like all queries made on those characters don't sort out
> OK...

May be you should run the backend with the proper value of locale for
LC_COLLATE (no, I didn't try myself yet, neither I checked in the PostgreSQL
code). But see the FAQ, too <http://www.postgresql.org/docs/faq-english.html#4.
1>.

> Especially if the first character is not standard english. The
> word is put at the end, after "z" !!!

If you read a phone directory in Sweden, you will discover that not all the
European languages sort the same, even when they use the same charset
(Latin-1).

> So, United States in french is "États-Unis" (hope you can see the first
> letter ok!)
> and it should be next to a country like "Equator"

In French, but not for all the Latin-1 users.