Hi,
this test fails newly, because it has "Jun 30, 2001" hard-coded and
assumes this date to be in the future which obviously became false
last Weekend. It took me a lot of time to track it down, because I
suspected everything but the test itself to be the reason ("It can't
be wrong, because last Friday it worked just fine") and therefore
didn't have a thorough look at the sql file first.
Is it possible to generate this date dynamically so that it is
guaranteed to always be in the future? If not, I suggest to set it to
something near infinity.
Furthermore, I suggest to generate regression.diffs with some more
lines of context. I would have helped a lot in this case, if the query
had shown up in the diff's context.
cu
Reinhard
--- expected/abstime.out Thu May 3 19:00:37 2001
+++ results/abstime.out Tue Jul 3 09:53:47 2001
@@ -43,18 +43,17 @@
SELECT '' AS six, ABSTIME_TBL.*
WHERE ABSTIME_TBL.f1 < abstime 'Jun 30, 2001';
six | f1
-----+------------------------------
| Sun Jan 14 03:14:21 1973 PST
| Mon May 01 00:30:30 1995 PDT
| epoch
- | current
| -infinity
| Sat May 10 23:59:12 1947 PST
-(6 rows)
+(5 rows)
SELECT '' AS six, ABSTIME_TBL.*
WHERE ABSTIME_TBL.f1 > abstime '-infinity';
six | f1
-----+------------------------------
| Sun Jan 14 03:14:21 1973 PST
| Mon May 01 00:30:30 1995 PDT