Re: BUG #14178: output of jsonb_object and json_object doesn't match textually

From: Thomas Munro <thomas(dot)munro(at)enterprisedb(dot)com>
To: Michael Paquier <michael(dot)paquier(at)gmail(dot)com>
Cc: ap(at)zip(dot)com(dot)au, PostgreSQL mailing lists <pgsql-bugs(at)postgresql(dot)org>
Subject: Re: BUG #14178: output of jsonb_object and json_object doesn't match textually
Date: 2016-06-07 07:55:10
Message-ID: CAEepm=09NJt4-NAm9Z6j9fVOGh9nFqHJ852SoJf_dhq3eBeduw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

On Tue, Jun 7, 2016 at 6:54 PM, Michael Paquier
<michael(dot)paquier(at)gmail(dot)com> wrote:
> On Tue, Jun 7, 2016 at 3:48 PM, Thomas Munro
> <thomas(dot)munro(at)enterprisedb(dot)com> wrote:
>> On Tue, Jun 7, 2016 at 2:40 PM, Michael Paquier
>> <michael(dot)paquier(at)gmail(dot)com> wrote:
>>> On Tue, Jun 7, 2016 at 10:42 AM, <ap(at)zip(dot)com(dot)au> wrote:
>>>> I think json_object should provide similar output. Consistency is nice
>>>> aesthetically and as being in line with the principle of least surprise.
>>>> Its lack may lead to possible unpleasant results if anyone tries to deal
>>>> with a JSON structure as TEXT (for whatever reason).
>>>
>>> That's a debatable point, and as any output is legal json it does not
>>> actually hurt one way or another. However, as json_object has been
>>> introduced at the same time as jsonb, I'd tend as well to think that
>>> consistency is a good idea if possible, see the patch attached.
>>
>> (It's a shame that json_build_object and json_object_agg also don't
>> agree on where to put whitespace...)
>
> Right, missed that, as well as json_object_two_arg.

One more difference is that json_object_agg uses "{ " and " }" while
the other functions use "{" and "}".

--
Thomas Munro
http://www.enterprisedb.com

In response to

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message boa 2016-06-07 09:16:18 BUG #14180: Segmentation fault on replication slave
Previous Message dmigowski 2016-06-07 07:50:36 BUG #14179: Not enough info in error message