On Sat, Dec 16, 2006 at 01:16:08PM +0000, José Matos wrote: > On Saturday 16 December 2006 1:09 pm, Axel Thimm wrote: > > egg is a packaging method that is orthogonal to what we use. Leaving > > the eggs around may get users to start using egg-installation and get > > files on the system unregistered by rpm. > > > > Or not? If the above is correct eggs should even be banned just as > > other non-native package formats are banned (debs or tarballs for > > example). > > On one hand I agree with Axel, on the other this allows installations per > user, no? > > What is our policy regarding other programming language ( with their > repositories cpan, cran, ctan)... A user may download tarballs, cpan/ctan, alienated debs and install them under $HOME or /usr/local, we don't mandate anything about that places. But we wouldn't start shipping rpm-foreign bits to endorse people doing so - especially not, if they would override the rpm itself that is shipping the egg for example. E.g. the user has an issue with the package (maybe a real bug, maybe just not rtfm), sees the egg, erroneously thinks that he needs to complete the rpm installation by using it and does so. Later package updates are not seen anymore by the user, since he's using a local copy of the previous package. > This should probably be a general policy, no? Well, even if it is not written down somewhere explicitely it is rather assumed. It is also a difficult wording, of course the user is free to use debs, ctan, eggs and so on, and we won't explicitely forbid this, but we shouldn't push users to do so either. Perhaps: "Packaging tarballs or other non-rpm packaging formats within the package is strongly discoursaged as it may misguide users to override the package by using them" -- Axel.Thimm at ATrpms.net
Attachment:
pgpPUPUE93vC3.pgp
Description: PGP signature
-- Fedora-packaging mailing list Fedora-packaging@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-packaging
