Re: [HACKERS] regression failures]

> >
> >
> > I am getting a huge number of regression failures in the past few days,
> > perhaps 80% failure when run.
> >
> > It may be related to some patches installed. Anyone else seeing it.
> > The actual checkresults don't look too bad, though.
>
> Seen them too.
>
> The elog() messages have one more whitespace between ERROR
> and the message text itself. checkresults takes care about
> that.

Ah, I see. checkresults uses 'diff -uw', so it does not see the space
change from "ERROR: text" to "ERROR: text", but the report of ok/failed
does sees the space. The two space was removed by Massimo patch as part
of the the tprintf additions.

I have changed regress.sh to do a 'diff -w' rather than just a diff, so
it will properly report failure and success. I feel better. I was
going to research the failures today.

Do people like the removal of the extra space after the colon, so there
is only one space now?

--
Bruce Momjian | 830 Blythe Avenue
maillist(at)candle(dot)pha(dot)pa(dot)us | Drexel Hill, Pennsylvania 19026
+ If your life is a hard drive, | (610) 353-9879(w)
+ Christ can be your backup. | (610) 853-3000(h)