Postgres Pro
Downloads
Home   >   mailing lists

Re: jsonb_set array append hack? - Mailing list pgsql-hackers

From Dmitry Dolgov
Subject Re: jsonb_set array append hack?
Date
Msg-id CA+q6zcV=z_daXD8U-Kgt9hU9Xo=4Le3pKmKT6T6O9OBBvj_CVw@mail.gmail.com
Whole thread Raw
In response to Re: jsonb_set array append hack?  (Thom Brown <thom@linux.com>)
Responses Re: jsonb_set array append hack?  (Andrew Dunstan <andrew@dunslane.net>)
List pgsql-hackers
Tree view
> I would expect some kind of error.  We're trying to address a position in an array, and we're instead passing a key.  If it completes successfully, the chances are it isn't what the user intended.

Thanks for the explanation. So, basically, it should be like this, am I right?

postgres=# SELECT jsonb_set(
    '{"name": "Joe", "vehicle_types": ["car", "van"]}'::jsonb,
    '{vehicle_types, nonsense}',
    '"motorcycle"', true);
ERROR:  path element at the position 2 is not an integer

On 20 September 2015 at 23:50, Thom Brown <thom@linux.com> wrote:
On 20 September 2015 at 16:17, Dmitry Dolgov <9erthalion6@gmail.com> wrote:
I'm sorry, but I'm not sure, what behavior is expected in this case?
Right now the following logic was implemented:
    "we trying to set an element inside an array, but we've got a non-integer path item
    ("nonsense" in this particular case), so we're going to add a new element at the end of array by default"

If it's wrong, should we refuse to perform such kind of operations, or should we replace
    "vehicle_type": ["car", "van"]
to
    "vehicle_type: {"nonsense": "motorcycle"}
?

(please bottom-post)

I would expect some kind of error.  We're trying to address a position in an array, and we're instead passing a key.  If it completes successfully, the chances are it isn't what the user intended.

Thom

Attachment

pgsql-hackers by date:

Previous
From: Dean Rasheed
Date:
Subject: Re: Bug in numeric multiplication
Next
From: Tom Lane
Date:
Subject: Re: Bug in numeric multiplication