Re: remaining sql/json patches

Hi,

> On Tue, Mar 5, 2024 at 12:03 AM Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> wrote:
>> On 2024-Mar-04, Erik Rijkers wrote:
>>
>> > In my hands (applying with patch), the patches, esp. 0001, do not apply.
>> > But I see the cfbot builds without problem so maybe just ignore these FAILED
>> > lines. Better get them merged - so I can test there...
>>
>> It's because of dbbca2cf299b. It should apply cleanly if you do "git
>> checkout dbbca2cf299b^" first ... That commit is so recent that
>> evidently the cfbot hasn't had a chance to try this patch again since it
>> went in, which is why it's still green.
>
> Thanks for the heads up. Attaching rebased patches.

In the commit message of 0001, we have:

"""
Both JSON_VALUE() and JSON_QUERY() functions have options for
handling EMPTY and ERROR conditions, which can be used to specify
the behavior when no values are matched and when an error occurs
during evaluation, respectively.

All of these functions only operate on jsonb values. The workaround
for now is to cast the argument to jsonb.
"""

which is not clear for me why we introduce JSON_VALUE() function, is it
for handling EMPTY or ERROR conditions? I think the existing cast
workaround have a similar capacity?

Then I think if it is introduced as a performance improvement like [1],
then the test at [1] might be interesting. If this is the case, the
method in [1] can avoid the user to modify these queries for the using
the new function.

[1] https://www.postgresql.org/message-id/8734t6c5rh.fsf@163.com

--
Best Regards
Andy Fan