| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Alvaro Herrera <alvherre(at)commandprompt(dot)com> |
| Cc: | Dimitri Fontaine <dfontaine(at)hi-media(dot)com>, Robert Haas <robertmhaas(at)gmail(dot)com>, Martijn van Oosterhout <kleptog(at)svana(dot)org>, Hitoshi Harada <umi(dot)tanuki(at)gmail(dot)com>, Erik Rijkers <er(at)xs4all(dot)nl>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: review: More frame options in window functions |
| Date: | 2010-02-12 16:55:40 |
| Message-ID: | 5075.1265993740@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Alvaro Herrera <alvherre(at)commandprompt(dot)com> writes:
> Tom Lane escribi:
>> Yeah, I don't immediately see anything that would justify going to
>> that level of effort. Adding +/- as support functions for btree
>> seems like the thing to do.
> Would it work to use a fake access method instead?
Then you'd have to duplicate all the information in a btree opclass;
*and* teach all the stuff that knows about btree to know about fakeam
instead. Doesn't seem like there's any win there.
> If we add it to
> btree, will we be able to backtrack and move that to a separate catalog
> if we ever determine that it would have been better?
Backwards compatibility with existing user-type definitions is one big
reason to *not* try to pull ORDER BY information out of btree.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Euler Taveira de Oliveira | 2010-02-12 17:13:06 | Re: TCP keepalive support for libpq |
| Previous Message | Heikki Linnakangas | 2010-02-12 16:10:05 | Re: Re: [COMMITTERS] pgsql: Make standby server continuously retry restoring the next WAL |