Re: Initial review of xslt with no limits patch

David E. Wheeler <david@kineticode.com>

From: "David E. Wheeler" <david@kineticode.com>

To: Tom Lane <tgl@sss.pgh.pa.us>

Cc: Robert Haas <robertmhaas@gmail.com>, Pavel Stehule <pavel.stehule@gmail.com>, Andrew Dunstan <andrew@dunslane.net>, Mike Fowler <mike@mlfowler.com>, pgsql-hackers@postgresql.org

Date: 2010-08-06T20:05:26Z

Lists: pgsql-hackers

On Aug 6, 2010, at 11:13 AM, Tom Lane wrote:

> That would work too, although I think it might be a bit harder to use
> than one alternating-name-and-value array, at least in some scenarios.
> You'd have to be careful that you got the values in the same order in
> both arrays, which'd be easy to botch.
> 
> There might be other use-cases where two separate arrays are easier
> to use, but I'm not seeing one offhand.

Stuff like this makes me wish PostgreSQL had an ordered pair data type. Then you'd just have a function with `variadic ordered pair` as the signature.

I don't suppose anyone has implemented a data type like this…

Best,

David