On Fri, Jun 15, 2012 at 10:38 AM, D.Morgan <[email protected]> wrote: > On Fri, Jun 15, 2012 at 10:53 AM, Colin Guthrie <[email protected]> wrote: >> 'Twas brillig, and David Walser at 15/06/12 01:42 did gyre and gimble: >>> So, maybe we should do something about this. My preference would >>> have rpm still complain about unpackaged %excluded files, but not >>> die, just give a warning instead. >> >> Personally, I don't see the point in using %exclude to exclude files >> completely from all sub packages. We can just use "rm" in the the >> %install section for that (we typically already do that to remove *.la >> files for example). >> >> So for me %exclude should only operate on subpackage file list and >> should only be useful to undo any wildcard inclusions: >> >> e.g. >> %files >> %{_libdir}/foo/* >> %exclude %{_libdir}/foo/something-in-another-sub-package >> >> %files -n sub >> %{_libdir}/foo/something-in-another-sub-package >> >> >> That kind of thing. >> >> That, to me, seems most logical. I'm not sure what the upstream >> behaviour is, but I would agree that we should follow it all the same >> even if the goal would be to push for upstream changes when needed. >> >> Cheers >> >> Col > > Hi, > > see with upstream. We won't change the behaviour of rpm for this "only > in mageia" and be inconsistent with fedora/opensuse/...
Isn't it the upstream behavior already that he is describing?
