Operators bind tighter than function application

[jstolarek]

If I say

fst (raise "foo") + 2

this parses as fst ((raise "foo") + 2) rather than (fst (raise "foo")) + 2. I've run into similar issues several times before. The problem is that operators bind tighter than application but it should be the other way around. Perhaps this can be solved by modifying the parser. If not then we might need a separate pass to restore correct operator precedence (I think GHC does something like that).

[jamescheney]

Ugh, yeah. I now remember running into this and thinking it would be good to fix sometime...