Re: 6.5.0 datetime bug?

> > Hmm, I happen to have a 6.5.0 system sitting here: It works there, so I suspect
> > something with your local operating system config.
> now, this is weird.

I should have asked originally: what time zone are you running in?
From your mailing address I'll bet that you are on the other side of
GMT from where I run my tests:

postgres=# set time zone 'Europe/Amsterdam';
SET VARIABLE
postgres=# select date_part('day',
'3-27-2000'::timestamp-'3-6-2000'::timestamp) as days;days 
------  20
(1 row)

OK, I see the problem in current sources :(

Thanks for pursuing this; I'll take a look at it.

btw, if we were to add some "other side of GMT" time zone testing to
our regression test, what time zone would be the most likely to be
universally supported? We know that PST8PDT works pretty well, but I'm
not sure of the best candidate for the other side...
                   - Thomas

-- 
Thomas Lockhart                lockhart@alumni.caltech.edu
South Pasadena, California