Skip to content

promoting scalar arguments to vectors in a function

4 messages · JLucke at ria.buffalo.edu, S Ellison, Jeff Newmiller +1 more

JLucke at ria.buffalo.edu
# 
R users: 

Suppose I have a function that takes three numeric arguments x, y,  and z, 
any of which may be scalars or vectors.
Suppose further that the user takes one of the arguments, say y, as a 
vector with the other two as scalars.
Is there an existing R function that will promote the other two arguments 
to vectors of the same size?

I know in most cases this is not a problem as the promotion is automatic. 
I need this for a function I am writing
and I don't want to write any additional code to do this if I can help it.

Joe
Joseph F. Lucke, PhD
Senior Statistician
Research Institute on Addictions
University at Buffalo
State University of New York
1021 Main Street
Buffalo, NY  14203-1016
S Ellison
# 
This depends on what you mean by 'promote' ...

If you want to do something like 'recycling', as is often done for plot colours, but need things the same length for subsetting etc, one crude but fairly straightforward way I will confess to using in a function is something like

L <- max( length(x), length(y), length(z) )
x <- rep(x, length.out=L) 
y <- rep(y, length.out=L) 
z <- rep(z, length.out=L)

If you wanted to zero-fill to the same length, or fill with NA, that'd be something else ...

S Ellison




*******************************************************************
This email and any attachments are confidential. Any use...{{dropped:8}}
Jeff Newmiller
# 
Scalar values in R are just vectors of length 1. The "promotion" you are thinking of is "recycling" that automatically occurs when vectors of different lengths are supplied to certain operations. 

One shortcut might be to put them all into a data frame using the data.frame() function.  Otherwise, I think the rep() function and some conditional code would be needed.
Marc Schwartz
#
On May 23, 2016, at 10:59 AM, JLucke at ria.buffalo.edu wrote:
For a general overview of R's recycling approach, see:

  https://cran.r-project.org/doc/manuals/r-release/R-intro.html#Vector-arithmetic

and:

  https://cran.r-project.org/doc/manuals/r-release/R-intro.html#The-recycling-rule

The question is what type of error checking you want to embed in your function, if the lengths of each argument are not the same and that would be a problem. Do you want to presume that a user will pass arguments that fit your a priori expectation, or protect against the possibility that they do not?

What would you do if y is a vector of length 6, x is a vector of length 1 and z is a vector of length 8?

R's default rules would replicate 'x' 8 times (e.g. rep(x, 8)), while extending 'y' by 2 (e.g. y <- y[c(1:6, 1:2)]), so that both have length(z).

You can use something along the lines of:

Max.Len <- max(length(x), length(y), length(z))
x <- rep(x, length.out = Max.Len)
y <- rep(y, length.out = Max.Len)
z <- rep(z, length.out = Max.Len)

which will recycle each as may be needed so that they have a common length.

Only you will know, given your function, if that might result in problems in whatever result your function is intended to generate.

Regards,

Marc Schwartz