One advantage of the new pipe operator over magrittr's is that the former
works with substitute().
> f <- function(x, xlab=deparse1(substitute(x))) paste(sep="", xlab, ": ",
paste(collapse=", ",x))
> 2^(1:4) |> f()
[1] "2^(1:4): 2, 4, 8, 16"
> 2^(1:4) %>% f()
[1] ".: 2, 4, 8, 16"
This is because
Luke,
Mostly an aside. I think that pipes are a good addition, and it is clear
that you and
other R-core thought through many of the details. Congratulations on what
appears to be
solid work. I've used Unix since '79, so it is almost guarranteed that I like
the basic
idiom, and I expect
