--- Tom Lane <[EMAIL PROTECTED]> wrote:
> Hoo, I'm surprised no one noticed this during 7.4
> development/testing.
> The problem applies for any datatype that requires
> double alignment,
> which includes int8, float8, and timestamp as well
> as most of the
> geometric types. pg_statistic is decl
Joe Conway <[EMAIL PROTECTED]> writes:
> anyarray has been defined this way since 7.3 -- any concerns there?
I don't think so --- we weren't trying to use it as an actual column
datatype back then.
7.4 has a problem though :-( ... this is one of the "damn I wish we'd
caught that before release" o
Shelby Cain <[EMAIL PROTECTED]> writes:
> The select statements return different data for
> most_commons_vals depending on whether n_distinct is
> included in the select clause or not.
> I only seem to get the behavior below against int8
> columns - but I haven't interated through every
> conceivab
