Re: [HACKERS] SQL/JSON in PostgreSQL - Mailing list pgsql-hackers

From Sven R. Kunze
Subject Re: [HACKERS] SQL/JSON in PostgreSQL
Date
Msg-id 70d3d53c-4494-07fb-1118-5c63d5f6917e@mail.de
Whole thread Raw
In response to Re: [HACKERS] SQL/JSON in PostgreSQL  (Andres Freund <andres@anarazel.de>)
Responses Re: [HACKERS] SQL/JSON in PostgreSQL  (Oleg Bartunov <obartunov@gmail.com>)
Re: [HACKERS] SQL/JSON in PostgreSQL  (Peter van Hardenberg <pvh@pvh.ca>)
Re: [HACKERS] SQL/JSON in PostgreSQL  (Nico Williams <nico@cryptonector.com>)
List pgsql-hackers
Hi,

about the datetime issue: as far as I know, JSON does not define a 
serialization format for dates and timestamps.

On the other hand, YAML (as a superset of JSON) already supports a 
language-independent date(time) serialization format 
(http://yaml.org/type/timestamp.html).

I haven't had a glance into the SQL/JSON standard yet and a quick search 
didn't reveal anything. However, reading your test case here 
https://github.com/postgrespro/sqljson/blob/5a8a241/src/test/regress/sql/sql_json.sql#L411 
it seems as if you intend to parse all strings in the form of 
"YYYY-MM-DD" as dates. This is problematic in case a string happens to 
look like this but is not intended to be a date.

Just for the sake of completeness: YAML solves this issue by omitting 
the quotation marks around the date string (just as JSON integers have 
no quotations marks around them).

Regards,
Sven



pgsql-hackers by date:

Previous
From: Andres Freund
Date:
Subject: Re: [HACKERS] WIP: [[Parallel] Shared] Hash
Next
From: Mark Dilger
Date:
Subject: Re: [HACKERS] Hash support for grouping sets