Hi, hackers!
I noticed that SELECT results in float4 and float8 tests lack ORDER BY
clauses. This makes test results depend on the current heap/MVCC
implementation.
If I try to run the float8 test on a table created with a different
access method provided by an extension, I'm getting results ordered
differently.
It's not a big problem, but propose a simple fix for the tests. It
just adds ORDER BY 1 to all relevant float4 and floa8 queries. I don't
have a strong opinion about backpatching this, but as the patch
changes only regression tests, it's maybe also worth backpatching.
Regards,
Pavel Borisov,
Supabase.