Re: AW: [HACKERS] misc and triggers regression tests failed on 6.5bet a1

I have applied a new refint from Massimo. Can people check that and
make the needed fixes to it and trigger.sql?

It seems that virtually all platforms fail with misc and trigger
regression tests.

The function check_primary_key in contrib/spi/refint.c was changed to take
an
additional obligatory argument automatic or dependent. This change was not
made in test/regress/sql/trigger.sql.
The error message in check_primary_key is in the wrong place thus
the misleading error msg: even number of arguments expected.

I supply a patch so the error shows, but I don't supply a patch for
trigger.sql,
since I think possibly the code should be reviewed and changed to use a
default argument for action: dependent.

<<refint.patch>>

Are these bad news for 6.5beta? Or should we just update expected
files?

No :-) No !

Andreas

[Attachment, skipping...]

-- 
  Bruce Momjian                        |  http://www.op.net/~candle
  maillist@candle.pha.pa.us            |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026