ddply error when the aggregation function is defined within another function

Question

The point of the following almost minimal code is the application of ddply within a function f where the aggregation function (helper) of ddply is self-defined within the definition of f.

Unfortunately, i don't understand why sourcing the entire snippet produces an Error in eval(expr, envir, enclos) : could not find function "helper". The code works when the helper function is run independently of the function f. When i replace the ddply call with the uncommented call of by, the code runs without error.

Can you explain the error and provide a solution or a workaround? [Tested with plyr 1.8.1 and R 3.0.3]

rm (list = ls())
library(plyr)

f <- function() {

  dfx <- data.frame(
    group = c(rep('A', 8), rep('B', 15), rep('C', 6)),
    sex = sample(c("M", "F"), size = 29, replace = TRUE),
    age = runif(n = 29, min = 18, max = 54)
  )

  helper <- function(x) {
    return(max(x))
  }

  result <- ddply(dfx, .(group, sex), summarize, max_age = helper(age))
  #result <- by(dfx$age, dfx[,c("group", "sex")], helper)

  return(result)
}

print(f())

BrodieG BrodieG · Accepted Answer · 2014-04-22T19:47:11

Try:

result <- ddply(dfx, .(group, sex), here(summarize), max_age = helper(age))

From the help page for here:

This function captures the current context, making it easier to use **ply with functions that do special evaluation and need access to the environment where ddply was called from.

summarize is one such special function.

ddply error when the aggregation function is defined within another function

1 Answers