Re: Collations versus record-returning functions - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: Collations versus record-returning functions
Date
Msg-id 1300783141.7698.14.camel@vanquo.pezone.net
Whole thread Raw
In response to Re: Collations versus record-returning functions  (Martijn van Oosterhout <kleptog@svana.org>)
List pgsql-hackers
On sön, 2011-03-20 at 20:26 +0100, Martijn van Oosterhout wrote:
> A rowtype has an order, determined by the fields within it. Those
> fields may be strings and so may have a collation. Doesn't seem
> particularly magical to me.

Yeah, that's answer #4.  The composite types themselves are not
considered collatable, but the fields in them carry collation
individually.  That's what the test case in question represents, and I
think it must work like that if you maintain the analogy between
composite types and  tables (which have columns that carry collation
individually).



pgsql-hackers by date:

Previous
From: Heikki Linnakangas
Date:
Subject: Re: Chinese initdb on Windows
Next
From: Piyush Newe
Date:
Subject: Re: Rectifying wrong Date outputs