Re: jsonb_set: update or upsert default? - Mailing list pgsql-hackers

From David E. Wheeler
Subject Re: jsonb_set: update or upsert default?
Date
Msg-id B3D315A2-5553-49F5-AA47-9F57DA6BBE1F@justatheory.com
Whole thread Raw
In response to jsonb_set: update or upsert default?  (Andrew Dunstan <andrew@dunslane.net>)
Responses Re: jsonb_set: update or upsert default?  (Petr Jelinek <petr@2ndquadrant.com>)
List pgsql-hackers
On May 22, 2015, at 7:22 PM, Andrew Dunstan <andrew@dunslane.net> wrote:

> The proposed flag for jsonb_set (the renamed jsonb_replace) in the patch I recently published is set to false,
meaningthat the default behaviour is to require all elements of the path including the last to be present. What that
doesis effectively UPDATE for jsonb. If the flag is true, then the last element can be absent, in which case it's
created,so this is basically UPSERT for jsonb. The question is which should be the default. We got into the weeds on
thiswith suggestions of throwing errors on missing paths, but that's going nowhere, and I want to get discussion back
ontothe topic of what should be the default. 

Here’s JavaScript in Chrome, FWIW:

var f = {}
f["foo"][0] = “bar"
Uncaught TypeError: Cannot set property '0' of undefined   at <anonymous>:2:13   at Object.InjectedScript._evaluateOn
(<anonymous>:895:140)  at Object.InjectedScript._evaluateAndWrap (<anonymous>:828:34)   at
Object.InjectedScript.evaluate(<anonymous>:694:21) 

Best,

David


pgsql-hackers by date:

Previous
From: Andrew Dunstan
Date:
Subject: Re: Re: [COMMITTERS] pgsql: Map basebackup tablespaces using a tablespace_map file
Next
From: Bruce Momjian
Date:
Subject: Re: Run pgindent now?