Re: pg_upgrade failure on Windows Server

On Thu, Jan 22, 2015 at 12:06 AM, Bruce Momjian <bruce(at)momjian(dot)us> wrote:
>
> On Wed, Jan 21, 2015 at 03:31:42PM -0300, Alvaro Herrera wrote:
> > Bruce Momjian wrote:
> >
> > > This sounds like the exact right patch. However, since it has a lot
of
> > > Windows-specific code, and we don't have any Windows experts, I am not
> > > sure how this can be applied.
> >
> > Are you saying we will remove the Windows port? That sounds awesome,
> > thanks! If you need help, I will volunteer on the spot, just LMK.
>
> :-)
>
> Well, I _am_ saying that historically patches that touch the innards of
> the Windows API are rarely applied as we can't evaluate or maintain the
> code. I can probably come up with an example if you want.

I think it is true to a great extent that Windows patches receive less
attention, however in many cases the patch finally do get committed.
I think the right thing for this patch is that Author should submit it to
next CF, so that it could be tracked and reviewed, once it is reviewed
by some one having Windows access, it should be taken care by
Committer.

> The
> Windows-specif code we do carry is required and was developed by people
> that are no longer as involved.
>

I have seen many a times that once committer's (those who are not generally
involved in Windows development) get reasonable confidence about patch
and the review done for the same, they commit the patch. It happens both
for the patches where I was Author and where I was Reviewer, although I
agree that it takes more time.

With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com