Re: [HACKERS] Bug in to_timestamp().

On Thu, Aug 2, 2018 at 9:06 PM Alexander Korotkov
<a(dot)korotkov(at)postgrespro(dot)ru> wrote:
> On Thu, Aug 2, 2018 at 6:17 PM Alexander Korotkov
> <a(dot)korotkov(at)postgrespro(dot)ru> wrote:
> > After some experiments I found that when you mix spaces and separators
> > between two fields, then Oracle takes into account only length of last
> > group of spaces/separators.
> >
> > # SELECT to_timestamp('2018- -01 02', 'YYYY---- --- --MM-DD') FROM
> > dual2018-01-01 00:00:00 -10:00
> > (length of last spaces/separators group is 2)
> >
> > # SELECT to_timestamp('2018- -01 02', 'YYYY---- --- --MM-DD') FROM dual
> > 2018-01-01 00:00:00 -10:00
> > (length of last spaces/separators group is 3)
> >
> > # SELECT to_timestamp('2018- -01 02', 'YYYY---- -- ---MM-DD') FROM dual
> > 02.01.2018 00:00:00
> > (length of last spaces/separators group is 2)
>
> Ooops... I'm sorry, but I've posted wrong results here. Correct
> version is here.
>
> # SELECT to_timestamp('2018- -01 02', 'YYYY---- --- --MM-DD') FROM dual
> ORA-01843: not a valid month
> (length of last spaces/separators group is 2)
>
> # SELECT to_timestamp('2018- -01 02', 'YYYY---- -- ---MM-DD') FROM dual
> 02.01.2018 00:00:00
> (length of last spaces/separators group is 3)
>
> So length of last group of spaces/separators in the pattern should be
> greater or equal to length of spaces/separators in the input string.
> Other previous groups are ignored in Oracle. And that seems
> ridiculous for me.

BTW, I've also revised documentation and regression tests. Patch is attached.

------
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company