Re: BUG #4862: different results in to_date() between 8.3.7 & 8.4.RC1

Oracle 9i:



 YEAR MONTH METHOD1     METHOD2

 2009 03   1/03/2009   1/03/2009





Oracle 10g:



YEAR      MONTH                METHOD1           METHOD2

 2009      03          1/03/2009            1/03/2009


Regards,
Jeremy.

On Sat, Jun 20, 2009 at 2:03 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:

> Brendan Jurd <direvus@gmail.com> writes:
> > I hope that answers your question.  to_date() is by nature a weird
> > beast with many strange corners in its behaviour, and it's hard to
> > strike a balance between backwards compatibility and Least
> > Astonishment.  My personal preference would be for a 100% strict
> > interpretation of the format pattern, and a pox on anyone who has been
> > relying on sloppy patterns!  But that's not very practical.  I would
> > welcome any suggestions for further refinements.
>
> My feeling about it is that we usually try to match Oracle's behavior
> for to_date/to_char, so the $64 question is whether Oracle allows a
> leading space in these same cases.  Anyone have it handy to test?
>
>                        regards, tom lane
>