On 6/6/18 12:14, Alvaro Herrera wrote:
> On 2018-May-17, Peter Eisentraut wrote:
>
>> The items that are still open from the original email are:
>>
>> 2) jsonb scalar values are passed to the plperl function wrapped in not
>> one, but _two_ layers of references
>>
>> 3) jsonb numeric values are passed as perl's NV (floating point) type,
>> losing precision if they're integers that would fit in an IV or UV.
>>
>> #2 appears to be a quality of implementation issue without any
>> user-visible effects.
>>
>> #3 is an opportunity for future improvement, but works as intended right
>> now.
>>
>> I think patches for these issues could still be considered during beta,
>> but they are not release blockers IMO.
>
> It appears to me that item #2 definitely would need to be fixed before
> release, so that it doesn't become a backwards-incompatibility later on.
The way I understand it, it's only how things are passed around
internally. Nothing is noticeable externally, and so there is no
backward compatibility issue.
At least that's how I understand it. So far this is only a claim by one
person. I haven't seen anything conclusive about whether there is an
actual issue.
> I'm not sure I agree that #3 is just a future feature. If you have
> functions working with jsonb numeric giving exact results, and later
> enable transforms for plperl, then your function starts giving inexact
> results? Maybe I misunderstand the issue but this doesn't sound great.
It would be the other way around. Right now, a transform from jsonb to
Perl would produce a float value in Perl. The argument is that it could
be an integer value in Perl if the original value fits. That's a change
worth considering, but the current behavior is consistent and works as
designed.
I took a brief look at this, and it seems there are some APIs needed to
be exposed from numeric.c to know whether a numeric is an integer.
--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
