R devel - Jan 2012 - seq_along and rep_along

Hi all,

A couple of ideas for improving seq_along:

* It would be really useful to have a second argument dim:

    seq_along(mtcars, 1)
    seq_along(mtcars, 2)
    # equivalent to
    seq_len(dim(mtcars)[1])
    seq_len(dim(mtcars)[2])

  I often find myself wanting to iterate over the rows or column of a
data frame, and there isn't a particularly nice idiom if you want to
avoid problems with zeros - you have to use seq_len(nrow(df)) etc

* To me, it would seem be very natural to have a rep_along function:

  rep_along <- function(x, y) rep(x, length.out = length(y))

  possibly with more checking for the case where the lengths aren't
integer multiples.

I'd be happy to submit proposed implementations/documentation if there
was interest.

Hadley

-- 
Assistant Professor / Dobelman Family Junior Chair
Department of Statistics / Rice University
http://had.co.nz/

On 12-01-06 1:31 PM, Hadley Wickham wrote:
> Hi all,
>
> A couple of ideas for improving seq_along:
>
> * It would be really useful to have a second argument dim:
>
>      seq_along(mtcars, 1)
>      seq_along(mtcars, 2)
>      # equivalent to
>      seq_len(dim(mtcars)[1])
>      seq_len(dim(mtcars)[2])
>
>    I often find myself wanting to iterate over the rows or column of a
> data frame, and there isn't a particularly nice idiom if you want to
> avoid problems with zeros - you have to use seq_len(nrow(df)) etc
I don't see the benefit of seq_along(mtcars, 1) versus seq_len(nrow(df)) 
in readability.

Duncan Murdoch
>
> * To me, it would seem be very natural to have a rep_along function:
>
>    rep_along<- function(x, y) rep(x, length.out = length(y))
>
>    possibly with more checking for the case where the lengths aren't
> integer multiples.
>
> I'd be happy to submit proposed implementations/documentation if there
> was interest.
>
> Hadley
>