[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: quality of the cd images we are offering



On Thu, 15 Mar 2001, Santiago Garcia Mantinan wrote:

> > then check the diff. That might give a clue about the cause of non-inclusion. 
> > Note that dpkg-scanpackages will not include two .debs with the same package
> > name, this might be the case here. 
> 
> Even though what you say is similar to what I had already done and reported
> to the list on my message with subject "Testing yesterday's commit"
> explining the files that I have found to be missing from one of the woody
> cds I had made with cvs scripts, I have redone the tests.
> 
> I could not do the tests on the ftp.hu.debian.org image as I no longer have
> it (deleted it accidentally), so I have created a new set of images with
> current cvs code, and after mounting the first cd on /cdrom I run:
> 
> pule:/cdrom# find . -type f -name '*.deb'|wc -l
>     989
> pule:/cdrom# find . -name Packages -exec grep '^Package: ' {} \;|wc -l
>     983
> pule:/cdrom# find . -name Packages -exec grep '^Filename: ' {} \; | cut -d '
> ' -f 2- | sort > /tmp/packdebs
> pule:/cdrom# find . -type f -name '*.deb'|cut -b 3-|sed
> 's/^\(.*\)binary-all\(.*\)$/\1binary-i386\2/'|sort > /tmp/realdebs
> pule:/cdrom# diff /tmp/realdebs /tmp/packdebs
> 84d83
> < dists/potato/main/binary-i386/doc/doc-debian_2.2.2.deb
> 112d110
> < dists/potato/main/binary-i386/games/gnome-games_1.0.51-6.deb
> 132d129
> < dists/potato/main/binary-i386/games/task-gnome-games_1.0.5.deb
> 326d322
> < dists/potato/main/binary-i386/news/task-news-server_2.2.2.2000.01.31-4.deb
> 340d335
> < dists/potato/main/binary-i386/sound/esound-common_0.2.17-7.deb
> 399d393
> < dists/potato/main/binary-i386/x11/task-gnome-desktop_1.0.10.deb
> 
> This files are all in binary-all, but as I had changed before binary-all to
> binary-i386 to make the names absolutelly similar, that's why they are shown
> as if they were on binary-i386.

This is so strange... There are quite many packages that woody takes from
potato/binary-all, so that can't be the problem. The potato dpkg-scanpackages
doesn't seem to have problems with these specific packages, but maybe the
woody version does? Something to do with recent perl weirdness?


Regards,
  Anne Bezemer



Reply to: