From: siretart@gmail.com
On Sun, Sep 21, 2025 at 6:23AM Alexander Kjll
wrote:
> >
> > And to spell it out explicitly we would prefer if you changed from
> > A)
> > Upload $package + its missing dependencies to unstable
> >
> > (Causing unstable containing broken packages for $random, posibly very
> > very long time))
> >
> > to
> >
> > B)
> > * Upload $package + its missing dependencies to experimental
> > * Wait for all parts to pass through NEW
> > * Re-upload everything to unstable
> >
> > cu Andreas
>
> Sure.
>
> This sounds to me like it should be a Debian wide policy. If this is a
> problem, why are uploads to unstable that need to go through NEW even
> allowed in the first place?
>
Because most uploads to unstable that require NEW processing don't depend
on packages that aren't in the archive yet.
Let's please not ask ftp-master to implement a policy to reject or delay
packages that have such missing dependencies. I'm concerned that would
delay NEW processing even more.
--
regards,
Reinhard
>
> And to spell it out explicitly we would prefer if you changed from
> A)
> Upload $package + its missing dependencies to unstable
>
> (Causing unstable containing broken packages for $random, posibly
very
> very long time))
>
> to
>
> B)
> * Upload $package + its missing dependencies to
experimental
> * Wait for all parts to pass through NEW
> * Re-upload everything to unstable
>
> cu Andreas
Sure.
This sounds to me like it should be a Debian wide policy. If this is a
problem, why are uploads to unstable that need to go through NEW even
allowed in the first place?
Because most uploads to unstable that
require NEW processing don't depend on packages that aren't in the
archive yet.
Let's please not ask ftp-master to implement a
policy
to reject or delay packages that have such missing dependencies. I'm
concernedthat would delay
NEW processing even more.
-- regards, Reinhard
--- SoupGate-Win32 v1.05
* Origin: you cannot sedate... all the things you hate (1:229/2)
|