Re: proposal: variadic argument support for least, greatest function - Mailing list pgsql-hackers

From Pavel Stehule
Subject Re: proposal: variadic argument support for least, greatest function
Date
Msg-id CAFj8pRCd5httH0zZG1_A127a3TjobzCHfvMavucrKAPiryv-wg@mail.gmail.com
Whole thread Raw
In response to Re: proposal: variadic argument support for least, greatest function  (Chapman Flack <chap@anastigmatix.net>)
Responses Re: proposal: variadic argument support for least, greatest function
List pgsql-hackers


so 23. 2. 2019 v 20:23 odesílatel Chapman Flack <chap@anastigmatix.net> napsal:
On 02/23/19 13:35, Pavel Stehule wrote:
> please, see, attached patch

It is getting close, for my purposes. There is still this:

>> Can the sentence added to the doc be changed to "These functions support
>> passing parameters as an array after <literal>VARIADIC</literal>
>> keyword."? That is, s/supports/support/ and s/a/an/. I've done that
>> after a couple of recent patches, but it seems to be on springs.

fixed




> I know so is important to understand to things, but nobody can understand
> to all. And then it is nice, so the things just works
>>
>> The approach with more parsimony indoors would be to just create a few
>> new ordinary functions, and add to the doc explaining why they are
>> different, and that would be a patch only needing to touch a couple files.
>>
>> I have a feeling that the final decision on whether the indoor or outdoor
>> parsimony matters more will be made by Tom, or another committer; I find
>> myself seeing both sides, and not feeling insider-y enough myself to
>> pick one.
>
> sure, every time it is commiter's decision.


A part of me waits to see if another voice chimes in on the high-level
want/don't-want question ... I think enough of the patch is ready for
that question to be ripe, and if the answer is going to be "don't want"
it would be ideal to know that before additional iterations of work on it.

sure

Regards

Pavel


-Chap
Attachment

pgsql-hackers by date:

Previous
From: Jeff Janes
Date:
Subject: Re: Bloom index cost model seems to be wrong
Next
From: Tom Lane
Date:
Subject: Re: [Bug Fix] ECPG: could not use set xxx to default statement