On Sun, 19 Sep 2004, Greg Stark wrote:
>
> Dennis Bjorklund <[EMAIL PROTECTED]> writes:
>
> > Still, we want the final solution to be what the sql standard specify. A
> > function as the proposed is however useful until a standard sql solution
> > is implemented. I've used a similar function for some projects and it have
> > worked well for me also.
>
> I haven't read the standard in this area (and the sections I have read don't
> lead me to think reading this section would be a trivial task), but from what
> I've seen people describe here I think the standard behaviour will be nigh
> useless anyway. From what I understand the standard has you declare a locale
> per column. That would mean the entire column can only store strings from a
> single locale.
AFAICT, it's one default collation per column, but you can specify a
collation to use on sorts and groups and comparisons.
I think example statement parts would be like:
ORDER BY name COLLATE de_DE
WHERE name COLLATE de_DE < 'Smith'
WHERE name < 'Smith' COLLATE de_DE
There are limitations, like I believe the following is an error
WHERE name COLLATE de_DE < 'Smith' COLLATE en_US
because both have different explicitly given collations.
---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
