| From: | "Vladimir Sitnikov" <sitnikov(dot)vladimir(at)gmail(dot)com> |
|---|---|
| To: | "Hitoshi Harada" <umi(dot)tanuki(at)gmail(dot)com> |
| Cc: | "David Rowley" <dgrowley(at)gmail(dot)com>, pgsql-hackers(at)postgresql(dot)org, heikki(dot)linnakangas(at)enterprisedb(dot)com |
| Subject: | Re: Windowing Function Patch Review -> Standard Conformance |
| Date: | 2008-11-05 03:22:47 |
| Message-ID: | 1d709ecc0811041922y2af09fderf605eedc098222e4@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
>
>
> Even though I understand the definition, your suggestion of COUNT(*)
> OVER (ORDER BY salary) doesn't make sense.
Why does not that make sense?
I have not read the spec, however Oracle has a default window specification
in case there is only an order by clause. The default window is "range
between unbounded preceding and current row".
"count(*) over (order by salary range between unbounded preceding and
current row)" is perfectly identical to the "number of rows preceding or
peers to R" by the definition, isn't it? I see here a word-by-word
translation from SQL to the English and vice versa.
If the patch returns "row_number" it is wrong since there is no way for
row_number to be a "number of rows preceding or peer with R", is there?
Regards,
Vladimir Sitnikov
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Robert Haas | 2008-11-05 03:49:57 | Re: [WIP] In-place upgrade |
| Previous Message | Hitoshi Harada | 2008-11-05 01:26:59 | Re: Windowing Function Patch Review -> Standard Conformance |