Can I nest apply functions in R? - r

Question

I have a series of CSV files that I want to prepare to append together. My appended file will be large, so I'd like to convert some string variables to numeric and date formats in the individual files rather than the larger appended file.
With other software, I would have one for loop that opens the file and nested for loops that would iterate over certain groups of variables. For this project, I am attempting to use R and apply functions.
I have mapply and lapply functions that work independently. I'm now trying to figure out how to combine them. Can I nest them? (See below for the independent parts and the nesting.)
(This code references code in the answer to How do I update data frame variables with sapply results?)
(Is it customary to provide an example CSV to give a reproducible example? Does R have built-in example CSVs?)
These work separately:
insert.division <- function(fileroot, divisionname){
ext <- ".csv"
file <- paste(fileroot, ext, sep = "")
data <- read.csv(file, header = TRUE, stringsAsFactors = FALSE)
data$division <- divisionname
write.csv(data, file = paste(fileroot, "_adj3", ext, sep = ""),
row.names = FALSE)
}
files <- c(
"file1",
"file2",
"file3",
"file4",
"file5"
)
divisions <- c(1:5)
#Open the files, insert division name, save new versions
mapply(insert.division, fileroot = files, divisionname = divisions)
#Change currency variables from string to numeric
currency.vars <- c(
"Price",
"RetailPrice"
)
df[currency.vars] <- lapply(
df[currency.vars],
function(x) as.numeric(sub("^\\(","-", gsub("[$,]|\\)$","", x)))
)
Combined version:
file.prep <- function(fileroot, divisionname, currency.vars){
ext <- ".csv"
file <- paste(fileroot, ext, sep = "")
data <- read.csv(file, header = TRUE, stringsAsFactors = FALSE)
data$division <- divisionname
df[currency.vars] <- lapply(
df[currency.vars],
function(x) as.numeric(sub("^\\(","-", gsub("[$,]|\\)$","", x)))
)
write.csv(data, file = paste(fileroot, "_adj", ext, sep = ""),
row.names = FALSE)
}
#Open the files, insert division name, change the currency variables,
#save new versions
mapply(file.prep, fileroot = files, divisionname = divisions,
currency.vars = df[currency.vars])

Answer 1

I'm not really sure why you're writing it back to file after changing the data, but here's an example of how I might approach the problem.
## Set up three csv files
set.seed(1)
DF <- data.frame(
w = paste0("($", sample(1500, 30) / 100, ")"),
x = Sys.Date() + 0:29,
y = sample(letters, 30, TRUE),
z = paste0("($", sample(1500, 30) / 100, ")")
)
fnames <- paste0("file", 1:3, ".csv")
Map(write.csv, split(DF, c(1, 10, 20)), fnames, row.names = FALSE)
Using your file.prep() function, you could adjust it a little and do
file.prep <- function(fileroot, divname, vars) {
ext <- ".csv"
file <- paste0(fileroot, ext)
data <- read.csv(file, stringsAsFactors = FALSE)
data$division <- divname
data[vars] <- lapply(data[vars], function(x) {
type.convert(gsub("[()$]", "", x))
})
write.csv(data, row.names = FALSE, file = paste0(fileroot, "_adj", ext))
}
divname <- 1:3
fnames <- paste0("file", divname)
Map(file.prep, fnames, divname, MoreArgs = list(vars = c("w", "z")))