On 17 October 2011 13:29, Max Bolingbroke <batterseapo...@hotmail.com> wrote: > I *think* this is what Duncan had in mind.
I spoke to him on IRC today and found that he actually had something different in mind, which I've implemented as the "hackage-server test-backup" command. This approach doesn't generate any arbitrary data at all - it just tests that it is possible to roundtrip the *current* server's data through a tarball. Generation of arbitrary data to stress-test the roundtripping functionality could be an interesting project, though, I guess. Max _______________________________________________ cabal-devel mailing list cabal-devel@haskell.org http://www.haskell.org/mailman/listinfo/cabal-devel