Checking new packages (Re: Status of Potato)

[Changwoo Ryu <cwryu@adam.kaist.ac.kr>]

Chris Leishman <masklin@debian.org> writes:

> On Fri, Nov 05, 1999 at 12:21:11AM +0100, Sami Dalouche wrote:
> > 
> > This incoming stuff is completly stupid ! Why aren't the uploaded packaged
> > directly moved to the directory structure ! The developpers aren't funny
> > enough to put a false section or something like that !
> > 
> 
> I don't maintain the incoming system, but IMHO it is a very necessary stage=
>  to
> 
> a) make sure that packages are uploaded correctly (ie, all files are present
>    and complete) before they get installed into the tree.
> b) the maintainer signature is correct.
> 
> 
> I'm sure there are other advantages to the system as well (that I just don't
> know about).

a) and b) are automatically done by dinstall.  AFAIK, new packages are
checked by human beings for (1) DFSG compliance and (2) the correct
section.

But I don't see why these two checks are needed.  The checks are not
complete anyway; we often see license-related grave bugs filed.  New
package upload could be disabled at freezing time, but I don't
understand why this bottleneck exists at the development dist.

-- 
Changwoo Ryu