data.frame Error when attempting to populate specific rows

General
#1

New to R Studio. I'm trying to read in excel file names and populate a data.frame:

#Let's say I've already read in a file name
counter <- 1 #simplifying as this would be a for loop that would start with a counter at 1
file_name <- "excel file name"
extension <- "xlsx"
status <-"successful"

#I initialize the data.frame and label the columns
global_file_list <- data.frame(File.Name = as.character(),
Extension = as.character(),
Status = as.character())

#When I try to populate this first row, I get the following warnings and the data.frame remains empty:
global_file_list[counter,] <- data.frame(cbind(file_name, extension, status))
Warning messages:
1: In [<-.factor(*tmp*, iseq, value = c(file_name = 1L)) :
invalid factor level, NA generated
2: In [<-.factor(*tmp*, iseq, value = c(file_ext = 1L)) :
invalid factor level, NA generated
3: In [<-.factor(*tmp*, iseq, value = c(status = 1L)) :
invalid factor level, NA generated

#However, if I were to do the following, without using the counter, the first row populates:
global_file_list <- data.frame(cbind(file_name, extension, status))

#Then this continues to work but populates the dataframe with N/A values
counter <- counter + 1
global_file_list[counter,] <- data.frame(cbind(file_name, extension, status))

  1. What is the reason the dataframe won't accept the first entry with a counter?
  2. Why does it populate with N/A instead of my variable values?

Thanks for your help!

#2

Here is one way to start.

global_file_list <- data.frame(file_name = "excel file name", extension = "xlsx",status ="successful", stringsAsFactors = FALSE)

If you don't include stringsAsFactors = FALSE all three columns will be created as factors. You may want to change extension and status to factors at some point.

#4

Trying to help you avoid having to discover what value of counter to use each time you want to append rows to the global list. Here is a pattern that use rbind.

global_file_list <- data.frame(file_name = "excel file name", extension = "xlsx",status ="successful", stringsAsFactors = FALSE)
next_file_list <- data.frame(file_name = "excel file name 2", extension = "xlsx",status ="successful", stringsAsFactors = FALSE)
global_file_list <- rbind(global_file_list,next_file_list)

That's the best I can do given you start with this "assumption".

#6

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

#3

Thanks for the response Phil. I tried stringsAsFactors and it works as you've written it but in my program the file_name, extension, and status are variables that change with every excel file read from an input folder. Because of that, I'm trying to add a row to the data.frame with each new file read by identifying the row with the counter variable. As such, I would write it as below (adding the notes with "..." to spare you of the code:

counter <- 1
global_file_list <- data.frame(File.Name = as.character(),
                           Extension = as.character(),
                           Status = as.character())

... start a for loop based on the number of files in the folder...
... read the file based on counter to get the below variables... (e.g. if counter = 1, read the 1st file in the folder, etc.)

file_name <- "some excel file name"
extension <- "could be xlsx or xls"
status <- "successful or unsuccessful"

#with this next line I attempt to input the 3 above variables into a new row in the dataframe

global_file_list[counter,] <- data.frame(cbind(file_name, extension, status),stringsAsFactors = FALSE)

counter <- counter + 1

...end for loop...

This addition of stringsAsFactors doesn't change the warning messages written in the original post and the row is populated with "NA" values.

#5

Thanks again Phil. Combining your input with that of another brought me to the answer. The two issues in my code were:

  1. use stringsAsFactors during the initialization of the data.frame, not when attempting to input data into it (as I had tried that as well)
  2. there is no need to input the data to global_file_list by using the "data.frame" command again as I was doing "data.frame(cbind(..." It works with just the cbind.
1 Like