On Wed, Dec 03, 2008 at 03:00:21PM +0100, Lucas Nussbaum wrote:
> > These things are the major slowdowns, at least for me, when doing NEW
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
keeping this in mind ...
> > processing:
> >
> > - package contains files under different license
> > - package contains files with differing copyright headers
> > - package has many copyright holders and these differ randomly between
> > files
> > - package has a large directory hierarchy
> >
> > If you have a package with a single license, a handful of copyright
> > holders and identical copyright headers in every file with a flat
> > source hierarchy, scanning through the package is pretty fast.
>
> Sounds completely logical.
Wonderful, and I agree that this looks really reasonable as a
scenario.
... but then the other half-thread discussing how quality reviews slow
down NEW processing is entirely pointless, as being based on an
initial wrong assumption (i.e., doing that reviews consumes time).
You can argue that maybe it is like this for Kalle and for the other
FTP masters doing NEW review, but that seems unlikely to me.
> > Then, if you've also made sure that you don't get any lintian
> > warnings and your debian-directory is clear (especially
> > debian/rules), the whole process is pretty painless.
>
> Why is that relevant?
I very much agree with previous Lucas observation that it sounds dumb
to have quality reviews (including running lintian and refuse if "E:"
output appears) for the first upload, and then complete freedom after
that.
Hence, the above comments from Kalle hints the obvious solution to the
disparity: make dak run lintian and reject uploads introducing "E:"
output.
If we do that, taking into account the lintian overrides contained
*in* the package being tested, we have a win-win solution:
- no disparity between first and next uploads
- a way to override lintian when it is on crack (which will also
expose the total consciousness of an uploader in doing a mistake, in
case an absurd override is found later on)
- time saving for the FTP masters which will have automatic reject of
packages they are most likely to reject anyhow (at least according
to Kalle)
- time saving for the uploader, which will have quasi-immediate
feedback from dak about the fact she is overlooking a potentially
serious error
Cheers.
--
Stefano Zacchiroli -o- PhD in Computer Science \ PostDoc @ Univ. Paris 7
zack@{upsilon.cc,pps.jussieu.fr,debian.org} -<>- http://upsilon.cc/zack/
Dietro un grande uomo c'è ..| . |. Et ne m'en veux pas si je te tutoie
sempre uno zaino ...........| ..: |.... Je dis tu à tous ceux que j'aime
Attachment:
signature.asc
Description: Digital signature