HashAgg's batching counter starts at 0, but Hash's starts at 1.
While working on 40efbf870 I noticed that when performing a Hash Join
that we always start out by setting nbatch to 1. That seems
reasonable as it's hard to imagine being able to complete any non-zero
amount of work in fewer than 1 batch.
In the HashAgg case, since 40efbf870, we'll display:
"HashAgg Batches": 0,
if you do something like: explain(analyze, format json) select
distinct oid from pg_class;
I'd rather this said that the number of batches was 1.
Does anyone have any objections to that being changed?