Michael Paquier <michael(at)paquier(dot)xyz> writes:
> On Mon, Nov 09, 2026 at 12:10:33AM -0500, shihao zhong wrote:
>> 0001 is your fix plus the same change for stanumbers. Now
>> serialize_expr_stats() matches update_attstats() again.
Ugh.
> Tom, any thoughts?
I think it's unfortunate that we have two copies of largely the same
code that have to be kept in sync. At the very least we ought to
install comments warning of that in both places. But I wonder if
it's practical to refactor so that there's just one copy?
regards, tom lane