Re: WIP patch: Collation support - Mailing list pgsql-hackers

From Zdenek Kotala
Subject Re: WIP patch: Collation support
Date
Msg-id 48D90774.3070307@sun.com
Whole thread Raw
In response to Re: WIP patch: Collation support  (Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>)
List pgsql-hackers
Heikki Linnakangas napsal(a):
> Zdenek Kotala wrote:
>> pg_collation catalog is also important for pg_dump, because system 
>> collation names are not compatible over OS and pg_dump output should 
>> be portable. pg_collation adds abstract layer which solve this problem.
> 
> That's a valid point. We'll still need a way to map OS locale to 
> whatever internal names we invent for them, though, so I'm not sure if 
> the pg_collation catalog helps much, but just moves the problem 
> elsewhere. 

It is true. For names we can for example use RFC479 0IANA register) 
http://tools.ietf.org/html/rfc4790#section-7
or use UNICODE terminology CLDR.

> The pg_dump output of the CREATE COLLATION statements still 
> wouldn't be portable from one OS to another.
> 

I don't think so that pg_collation catalog should be dumped (maybe only 
with extra switch).

        Zdenek


pgsql-hackers by date:

Previous
From: Gregory Stark
Date:
Subject: Re: Initial prefetch performance testing
Next
From: Greg Smith
Date:
Subject: Re: Initial prefetch performance testing