How to extract a part of the filename from the `read.table` command and define as a new column

Question

I'm looking for a way to automatically extract a part of the file name and populate a column with it while reading-in that file into R. Here's an example:

library(stringr)
df <- read.table("[MYPATH]/Mon1_2016-10-06_@exyz66698.txt", 
                 comment.char = "", header = FALSE, quote="", sep="\t", stringsAsFactors = F, encoding = "UTF-8") %>%
  mutate(Rec = str_extract(myFileNames[1], ".*\\d"))

This however works only because I have prepared in advance the vector myFileNames with all relevant file names and happen to know that the file name Mon1_2016-10-06_@exyz66698.txt" is the first element in that vector.

How can I achieve the same result without this detour to an external vector?

Answer 1

As I suggested in my comment (untested code):

readFileAndAddFileNameAsColumn <- function(fName) {
  read.table(
    fName, 
    comment.char = "", 
    header = FALSE, 
    quote="", 
    sep="\t", stringsAsFactors = F, encoding = "UTF-8"
  ) %>%
  mutate(Rec = str_extract(fName, ".*\\d"))
}

readFileAndAddFileNameAsColumn("[MYPATH]/Mon1_2016-10-06_@exyz66698.txt")

Alternatively, you may be able to use the .id argument of bind_rows if you pass your myFileNamesList to lapply. Something like

lapply(
  myFileNamesList,
  read.table,
  comment.car = "",
  header = FALSE,
  quote = "",
  sep = "\t",
  stringsAsFactors = FALSE,
  encoding = "UTF-8"
) %>%
bind_rows(.id = "Rec")