On 04/24/2018 12:17 PM, Peter Eisentraut wrote:
> On 4/10/18 10:31, Dagfinn Ilmari Mannsåker wrote:
>> Also, it doesn't parse back in as jsonb either:
>>
>> =# select jsonbnan()::text::json;
>> ERROR: invalid input syntax for type json
>> DETAIL: Token "NaN" is invalid.
>> CONTEXT: JSON data, line 1: NaN
>>
>> And it's inconsistent with to_jsonb():
>>
>> =# select to_jsonb('nan'::numeric);
>> ┌──────────┐
>> │ to_jsonb │
>> ├──────────┤
>> │ "NaN" │
>> └──────────┘
>>
>> It would be highly weird if PL transforms (jsonb_plpython does the same
>> thing) let you create spec-violating jsonb values that don't round-trip
>> via jsonb_out/in.
> Yeah this is not good. Is there a way to do this in a centralized way?
> Is there a function to check an internal jsonb value for consistency.
> Should at least the jsonb output function check and not print invalid
> values?
>
The output function fairly reasonably assumes that the jsonb is in a
form that would be parsed in by the input function. In particular, it
assumes that anything like a NaN will be stored as text and not as a
jsonb numeric. I don't think the transform should be doing anything
different from the input function.
There is the routine IsValidJsonNumber that helps - see among others
hstore_io.c for an example use.
cheers
andrew
--
Andrew Dunstan https://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services