Ack, done: https://code.google.com/p/ganeti/issues/detail?id=872
On Fri, Jul 4, 2014 at 7:10 PM, Klaus Aehlig <[email protected]> wrote: > On Fri, Jul 04, 2014 at 06:48:44PM +0200, 'Petr Pudlak' via ganeti-devel > wrote: > > We verify that 'readJSON . showJSON == Ok' and that maps are equal iff > > keys map to the same values. > > > > Signed-off-by: Petr Pudlak <[email protected]> > > --- > > Makefile.am | 1 + > > src/Ganeti/Utils/MultiMap.hs | 5 +++ > > test/hs/Test/Ganeti/Utils/MultiMap.hs | 72 > +++++++++++++++++++++++++++++++++++ > > test/hs/htest.hs | 2 + > > 4 files changed, 80 insertions(+) > > create mode 100644 test/hs/Test/Ganeti/Utils/MultiMap.hs > > LGTM. However, ... > > > +instance (Arbitrary k, Ord k, Arbitrary v, Ord v) > > + => Arbitrary (MultiMap k v) where > > + arbitrary = > > + let set = S.fromList <$> listOf arbitrary > > + in (multiMap . M.fromList) <$> listOf ((,) <$> arbitrary <*> set) > > ...the test for extensional equality would be a lot more meaningful, > if all terms from the functions that MultiMap exports were taken. > But since that instantiation is fine for the serialisation test, it's > OK for this patch series. But please file a low-priority issue to replace > this instanciation by a more meaningful one. > > -- > Klaus Aehlig > Google Germany GmbH, Dienerstr. 12, 80331 Muenchen > Registergericht und -nummer: Hamburg, HRB 86891 > Sitz der Gesellschaft: Hamburg > Geschaeftsfuehrer: Graham Law, Christine Elizabeth Flores >
