> IIRC, we were debating whether we should consider collation to be an
> attribute of the datatype (think typmod) or an attribute of individual
> values (think field added to values of textual types). In the former
> case, a function like this would only work if we allowed its result to
> be declared as having the right collate attribute. Which is not
> impossible, but we don't currently associate any typmod with function
> arguments or results, and so I'm not sure how painful it would be.
> With the field-in-data-value approach it's easy to see how it would
> work. But another byte or word per text value might be a high price
> to pay ...
I think the price is not so high. To give the collation info to text
data types, it's enough to store the info in the
pg_attribute. ie. only additional several bytes per column are
required, not per instance. Of course we would need to add some extra
bytes to the in-memory string data, it's just a temporary data anyway.
--
Tatsuo Ishii