Jose Kahan writes:
>
> On Thu, Jun 05, 2003 at 10:30:10AM -0500, kent landfield wrote:
> >
> > I'm not sure it really matters if the starting number is 0001 or 0000.
> > In the past it has been 0. Somewhere this bug creeped in then it was
> > both. Maybe we could just go back to 0 and put a warning in the README.
> > Seems like a waste of code to keep worrying about backwards compatibility
> > for bugs.
>
> It does matter if you rebuild your archive and all your messages are
> moved by one number.. this means all your links to them are broken
> (they will point to the wrong message). I can tell you that this is
> indeed serious, specially if you quote from the archives.
>
> If there was a way to automatically detect the problem, I'd go for it.
> But it's harder than that.
>
> That's why I proposed to have that option there. It's easy to have
> a script that rebuilds the archive and knows when to use one or the
> other option. And perhaps all your messages start with 1, rather
> than 0. You don't want to rebuild them starting with 0, if you wanted
> to take advantage of the WAI changes.
>
> -jose
Ok... Not sure I understand the last part about WAI changes, but you do
so that's the important thing. I defer that to you.
So let me understand what you are saying... Because of WAI changes the
base archive numbers need to start with the number 0001. Correct ? If
that's the case this is not just a one time bug. That would drastically
change things...
Again, if we go with the number 0001, we invalidate all existing archives
that are 0000 based. (This is most of the hypermail archives in existence.)
If we go with 0000 then we make the WAI changes harder to use. Is this
an accurate reflection of the problem ?
--
Kent Landfield | HYPERMAIL: http://www.hypermail.org/
Email: kent_at_hypermail.org | RFCS: http://www.faqs.org/rfcs/
Received on Thu 05 Jun 2003 07:49:51 PM GMT