From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Alvaro Herrera <alvherre(at)commandprompt(dot)com> |
Cc: | Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>, Robert Haas <robertmhaas(at)gmail(dot)com>, Andrew Dunstan <andrew(at)dunslane(dot)net>, David Fetter <david(at)fetter(dot)org>, Hitoshi Harada <umi(dot)tanuki(at)gmail(dot)com>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: TODO items for window functions |
Date: | 2008-12-31 16:04:41 |
Message-ID: | 26955.1230739481@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:
> Heikki Linnakangas escribi:
>> Tom Lane wrote:
>>> pg_catalog | nth_value | anyelement | anyelement, integer OVER window
>>
>> That looks like "OVER window" is associated with the "integer", like
>> DEFAULT. I don't have any better suggestions, though.
> pg_catalog | nth_value | anyelement | (anyelement, integer) OVER window
Yeah, I had considered that too, and it has a distinct advantage for
parameterless functions like rank():
Schema | Name | Result data type | Argument data types
------------+------+------------------+---------------------
pg_catalog | rank | bigint | OVER window
pg_catalog | rank | bigint | () OVER window
The latter is definitely clearer about what you're supposed to do.
However, it seems kind of inconsistent to do this for window functions
unless we also make \df start putting parens around the argument lists
for regular functions. Comments?
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Bruce Momjian | 2008-12-31 16:22:50 | Re: version() output vs. 32/64 bits |
Previous Message | Tom Lane | 2008-12-31 16:01:25 | Re: pg_pltemplate entries for external PLs |