This has been deprecated in favour of PairsMacros.jl
PairAsPipe (@pap) is friendly with DataFrames.jl's API.
The macro @pap is designed to transform newcol = fn(:col) to :col => fn => :newcol which is an elegant (ab)use of pairs syntax (a => b) as pipes. Hence the name of the package.
Some examples
using DataFrames, PairAsPipe
df = DataFrame(a = 1:3)
transform(df, @pap b = :a .* 2) # same as transform(df, :a => a->a.*2 => :b)
transform(df, @pap :a .* 2) # same as transform(df, :a => a->a.*2); except for output column name
transform(df, @pap sum(:a)) # same as transform(df, :a => mean); except for output column name
filter(@pap(:a == 1), df) # same as filter([:a] => a -> a == 1, df)