Reading multiple .dat files as a list and saving as .RDATA files in R

I want to import multiple .DAT files from a directory and make them as a list elements and then save them as .RDATA files.

I tried the following code

files <- dir(pattern = "*.DAT")
library(tidyverse)
Data1 <- 
  files %>%
    map(~ read.table(file = ., fill = TRUE))

which works sometimes and fails others. The files are also available on this link. I want to read all files and them save them as .RDATA with the same names.

Solution

Since the data of the link are partly a little bit unclean, I show you the solution of the core problem of your question on the basis of this example data:

(name1 <- name2 <- name3 <- name4 <- name5 <- data.frame(matrix(1:12, 3, 4)))
#   X1 X2 X3 X4
# 1  1  4  7 10
# 2  2  5  8 11
# 3  3  6  9 12

We save the data into a sub directory of your working directory named "test".

l <- mget(ls(pattern="^name"))
DIR <- "test"
# dir.create(DIR)  # leave out if dir already exists
sapply(1:length(l), function(x) 
  write.table(l[[x]], file=paste0(DIR, "/", names(l)[x], ".dat"), row.names=FALSE))

Now we look what's inside "test".

dir(DIR)
# [1] "name1.dat" "name2.dat" "name3.dat" "name4.dat" "name5.dat"

Now we import the files in the directory by pattern. I use rio::import_list, which nicely imports the files into a list an uses data.table::fread inside. But your own code also would work fine.

# rm(list=ls())  # commented out for user safety
L <- rio::import_list(paste0(DIR, "/", dir(DIR, pattern="\\.dat$")), format="tsv")

To save them as .Rdata we want to assign names dynamically which we achive with the list option within save().

sapply(seq_along(L), function(x) {
  tmp <- L[[x]]
  assign(names(L)[x], tmp)
  save(list=names(L)[x], file=paste0(DIR, "/", names(L)[x], ".Rdata"))
})

When we list the directory we see that the data was created.

dir(DIR)
# [1] "name1.dat"   "name1.Rdata" "name2.dat"   "name2.Rdata" "name3.dat"   "name3.Rdata"
# [7] "name4.dat"   "name4.Rdata" "name5.dat"   "name5.Rdata"

Now let's look whether the object names also were created correctly:

# rm(list=ls())  # commented out for user safety
load("test/name1.Rdata")
ls()
# [1] "name1"
name1
#   X1 X2 X3 X4
# 1  1  4  7 10
# 2  2  5  8 11
# 3  3  6  9 12

Which is the case.

Add-on option

We alternatively could attempt a more direct approach using rvest. First we fetch the data names:

library(rvest)
dat.names <- html_attr(html_nodes(read_html(
  "https://www2.stat.duke.edu/courses/Spring03/sta113/Data/Hand/Hand.html"),
  "a"), "href")

and create individual links:

links <- as.character(sapply(dat.names, function(x)
  paste0("https://www2.stat.duke.edu/courses/Spring03/sta113/Data/Hand/", x)))

The remainder is basically the same as above:

DIR <- "test"
# dir.create(DIR)  # leave out if dir already exists

library(rio)
system.time(L <- import_list(links, format="tsv") ) # this will take a minute
sapply(seq_along(L), function(x) {
  tmp <- L[[x]]
  assign(names(L)[x], tmp)
  save(list=names(L)[x], file=paste0(DIR, "/", names(L)[x], ".Rdata"))
})

# rm(list=ls())  # commented out for user safety
load("test/clinical.Rdata")  # test a data set
clinical
#    V1  V2  V3
# 1  26  31  57
# 2  51  59 110
# 3  21  11  32
# 4  40  34  74
# 5 138 135 273

However, as noted earlier in the introduction, the data are partly a little bit unclean and you probably will have to handle them individually and adapt the code case-wise.