ghc seems to normally be pretty good at suppressing "parasitic" type
errors, but in this following case it seems to get a tad confused...
Given two "overlapping" instance declarations (ones that don't actually
overlap, in fact, but that's the Haskell report fur ya), I get three
copies of the original error (a mysterious-seeming number), followed
by about 40 messages complaining about the lack of an instance declaration
for one of the "overlapped" instance declarations. Not particularly
unclear, but a tad excessive...
Intervals.hs:1: Duplicate or overlapping instance declarations
for `Num [n]' at Intervals.hs:491 and Intervals.hs:517
Intervals.hs:1: Duplicate or overlapping instance declarations
for `Num [n]' at Intervals.hs:491 and Intervals.hs:517
Intervals.hs:1: Duplicate or overlapping instance declarations
for `Num [n]' at Intervals.hs:491 and Intervals.hs:517
Intervals.hs:204: No instance for: `Show (Limit ta1QW)'
arising from use of `show' at Intervals.hs:204
When checking signature(s) for: `fintop'
Intervals.hs:204: No instance for: `Eq (Limit ta1QW)'
arising from use of `minimum' at Intervals.hs:204
When checking signature(s) for: `fintop'
Intervals.hs:194: No instance for: `Eq (Limit ta1Vo)'
arising from use of `minimum' at Intervals.hs:194
When checking signature(s) for: `intop'
Intervals.hs:181: No instance for: `Eq (Limit ta1YM)'
arising from use of `signum' at Intervals.hs:181
When checking signature(s) for: `sign'
... ad nauseum.
Cheers,
Alex.
