Hi Steven, Steven Chamberlain <steven@pyro.eu.org> (2015-10-07): > Package: anna > Version: 1.52 > Severity: wishlist > Tags: patch > Control: block 345408 by -1 > > Hi, > > If we had multiple udeb sources (Bug#345419), it could mean that > anna reads a Packages file having multiple versions for some udebs. > It's very easy to make net-retriever generate such a file, combining > multiple suites, e.g. stable + stable-proposed-updates; or maybe > stable + stable-backports. TBH, I'm not sure whether this should be dealt with in anna or in net-retriever. The latter is responsible for the addition in the first place, so could be considered as the one responsible for avoiding duplicate entries. Another way to look at it would be: net-retriever knows this limitation in anna, and is fine with trusting anna to do the "only pick the last occurrence" selection, provided n-r cats everything in the right order. ISTR some shell dance in net-retriever from my old patchset; having some C code in anna instead doesn't seem crazier. Taking a step back, one could argue that the logic could be slightly different, like picking the highest version. Even if it's a corner case, we could imagine having bits merged in a point release that's higher than what's in backports. More interestingly, that would mean anna doesn't care about the order in which the files were cat'd together. I'm just throwing some ideas, I don't have strong feelings at the moment. Mraw, KiBi.
Attachment:
signature.asc
Description: Digital signature