Re: Precision loss casting float to numeric

On Mon, Feb 26, 2018 at 1:29 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> The bigger question here is whether people actually want this behavioral
> change. I think there's probably a bigger chance of complaints that
> "casting 1.1::float8 to numeric now produces some weird,
> incorrectly-rounded result" than that we make anyone happier.

Yeah, I worry about that, too.

Of course, as you know, I also have a deep and abiding skepticism
about IEEE binary floats as a concept. Anomalies are unavoidable; we
can choose exactly which set users experience, but we can eliminate
them because the underlying storage format is poorly-adapted to the
behavior people actually want. It's too bad that IEEE decimal floats
weren't standardized until 2008.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company