On Tue, Aug 28, 2018 at 7:57 AM Nicolas Mailhot
<nicolas.mailhot(a)laposte.net> wrote:
Le 2018-08-08 22:48, Jeff Johnson a écrit :
> My issue is the misdirection discussing lazy filelist downloading as a
> "solution" to the "problem" of huge amounts of data that is
forced to
> be downloaded and loaded.
>
> The issue has been discussed repeatedly without a solution.
>
> Adding -- and maintaining -- patterns or whitelist exceptions, which
> moves file dependencies into primary.xml is actually an approach that
> solves the problem, and scales to multiple repos as well, each of
> which also will have their own patterns and whitelist.
Actually, you do not need to maintain whitelists manually, just having
createrepo automatically whitelist all the file deps it finds referenced
in the indexed repo would solve most of the problem.
If you want to be fancy, you could also add an option to pass it the url
of another repo is should inherit whitelists from (for updates, epel,
etc)
There is a very old joke about a man visiting a tailor and buying a
very nice suit that does not fit, so he lifts one arm, sticks his leg
out to the side, bends to the left, etc. And people look at him and
say "that poor man" and are told, "yes, but look at that wonderful
suit!".
The man is walking funny with all the filters applied. As soon as he
stands up straight, it's apparent that the actual content in the
repository does not match the dnf reported contents. And no, I'm
afraid it does *not* scale since the exclusions from another
repository may not be appropriate to the repository one is currently
building, and automatically including that whitelist is automatically
mismatching the content database to the actual content of the
repository *every time*.