Re: [WIP] patch - Collation at database level

Tom Lane napsal(a):
> Zdenek Kotala <Zdenek(dot)Kotala(at)Sun(dot)COM> writes:
>> ... And of course shared tables need also collation for their indexes.
>
> No, they don't, because the only textual indexes on shared catalogs are
> on "name" columns, which are intentionally not locale aware, and
> wouldn't be collation aware either.

Yeah, name uses strcmp, which is not locale aware but from ANSI perspective
there is collation SQL_IDENTIFIER for it which is fortunately implementation
defined.

What I see now as the problem is that we need also to know correct collation for
ORDER - for example:

select * from pg_shdescription order by description;

...thinking...

but it should solve by collation per column which will work well with
pg_attribute cloning for new database as Martijn mentioned.

Zdenek

--
Zdenek Kotala Sun Microsystems
Prague, Czech Republic http://sun.com/postgresql