On Thu, 10 Nov 2016, 12:20 p.m. Johannes Schauer, <[email protected]> wrote:
> Quoting James Clarke (2016-11-10 09:12:50) > > > In pbuilder's case the date of a binNMU is just `date -R` when the > command > > > runs, thus we don't have the same problem as described there. > > > > > > I'm following that discussion to see what it'll lead to, but I'm > > > unconvinced that we need to mangle the date like you suggest. > Especially > > > your last email that talks about m-a:same potential troubles leaves me > > > unbought, as ma:same is afaik only avout file contents, surely not file > > > metadata (timestamps) (which would be different anyway for the "last > > > timestamp+1sec" as you said?!). > > > > Well if SDE varies, binaries, manpages etc can end up having different > > contents... > > and then packages with the same binNMU number will not be co-installable > anymore if they share a file that ends up having different content because > of > S_D_E. > I argue that should not be in a ma:same package. Reproducible manages and such only exist since 1 year or so, ma:same exists since a lot before and before that manages weren't the same. Also, imho ma:same packages should be coinstallable regardless of the number of binNMUs that happened differently across archs (modulo dpkg not being happy, but your idea discards that too).
