Re: jsonb_set() strictness considered harmful to data

From: Steve Atkins <steve(at)blighty(dot)com>
To: pgsql-general(at)lists(dot)postgresql(dot)org
Subject: Re: jsonb_set() strictness considered harmful to data
Date: 2019-10-20 09:13:04
Message-ID: 5d3488c4-fd04-ee99-fb1e-e88cdb485238@blighty.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general pgsql-hackers


On 19/10/2019 07:52, Ariadne Conill wrote:
>
> I would say that any thing like
>
> update whatever set column=jsonb_set(column, '{foo}', NULL)
>
> should throw an exception. It should do, literally, *anything* else
> but blank that column.

steve=# create table foo (bar jsonb not null);
CREATE TABLE
steve=# insert into foo (bar) values ('{"a":"b"}');
INSERT 0 1
steve=# update foo set bar = jsonb_set(bar, '{foo}', NULL);
ERROR:  null value in column "bar" violates not-null constraint
DETAIL:  Failing row contains (null).
steve=# update foo set bar = jsonb_set(bar, '{foo}', 'null'::jsonb);
UPDATE 1

I don't see any behaviour that's particularly surprising there? Though I
understand how an app developer who's light on SQL might get it wrong -
and I've made similar mistakes in schema upgrade scripts without
involving jsonb.

Cheers,
  Steve

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Luca Ferrari 2019-10-20 09:35:04 UTC-6 or UTC+6?
Previous Message Floris Van Nee 2019-10-20 08:39:58 jsonb_set() strictness considered harmful to data

Browse pgsql-hackers by date

  From Date Subject
Next Message Andreas Karlsson 2019-10-20 11:30:33 Re: Declaring a strict function returns not null / eval speed
Previous Message Floris Van Nee 2019-10-20 08:39:58 jsonb_set() strictness considered harmful to data