From: | "Hitoshi Harada" <umi(dot)tanuki(at)gmail(dot)com> |
---|---|
To: | "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | "Gregory Stark" <stark(at)enterprisedb(dot)com>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Status of DISTINCT-by-hashing work |
Date: | 2008-08-06 14:40:52 |
Message-ID: | e08cc0400808060740y5456eb96u8676b096ffba8285@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
2008/8/6 Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>:
> "Hitoshi Harada" <umi(dot)tanuki(at)gmail(dot)com> writes:
>> In my patch nodeWindow.c, some functions reach for its parent state
>> node to get info of sort keys by using fcinfo->context. This works but
>> is completely ugly.
>
> Isn't that the same thing nodeAgg does: pass its AggState to aggregate
> functions? I don't see anything ugly about it --- at least, any
> alternative you care to name is likely to be far worse.
Correct. I learned that method from nodeAgg. What I meant was that
window functions reach for plan node and its sort keys through
winstate->ss.ps. The mechanism is not ugly but some macros or exposed
API seem more comfortable, especially if thinking about formulating
window functions as user-defined functions.
Regards,
--
Hitoshi Harada
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2008-08-06 14:42:52 | Re: Status of DISTINCT-by-hashing work |
Previous Message | Dimitri Fontaine | 2008-08-06 14:35:04 | Re: Automatic Client Failover |