On 09/14/2015 03:41 PM, Alvaro Herrera wrote:
> Andrew Dunstan wrote:
>
>> Currently, json_agg, jsonb_agg, json_object_agg and jsonb_object_agg do type
>> classification on their arguments on each call to the transition function.
>> This is quite unnecessary, as the argument types won't change. This patch
>> remedies the defect by caching the necessary values in the aggregate state
>> object.
> Seems a reasonable idea to me. This is 9.6 only, right?
I think we can reasonably backpatch it to 9.5, which is where the jsonb
functions were actually introduced. It's not at all user visible, and
we're still in alpha. Seem fair?
I have addressed your stylistic concerns, but I'll leave the fmgr_info
question Teodor raised for another day. Before I do anything more than
this I want to do some profiling to find out where the time is actually
going for various workloads.
cheers
andrew