Open moodymudskipper opened 4 years ago
funs
could have a pattern
argument, it should contain {fun}
and {col}
, or none, in the latter case it will be used as a separator the same way as dplyr/tidyr do.
It should work with spread too, because pivot to wider offer a multi spread with various functions and various variables that we don't support, and should!
dplyr does it with scoped mutate variants.
One issue is that we first look for
.
, solist(a=.+2,b=mean)
will not be seen as a list of functions, we could usefuns(a=.+2,b=mean)
and NSE to spot it.