How to append input file name to corresponding list output index in R

Question

How do I append the name of corresponding file name to the list output of my code. The goal is to be able to trace the outputs in the list to the input csv files. Currently, the list output returns a list index [[1]],[[2]],...,[[5]] (see the snapshot below). I want the corresponding file name included, something like this CA_three , FL_three,...., NY_two

@ Akrun, I want each page to have a corresponding file name

Below there are two codes

Code 1 : code that loops through 5 csv files and returns a list of outputs [[]] (I need help here)
Code 2: code to generate 5 csv files used in Code 1

# Code 1
library(multcompView) # for tukeyHD

# list of 20 csv files in a folder

files <- list.files("C:\\mypath\\", pattern="*.csv", full.names = T)

out <- lapply(1:length(files), function(x) {  # use lapply to loop over all files
  
  this_data <- read.csv(files[x], header = TRUE)
  aov_mod <- aov(revenue ~ dept, data = this_data)
  tuk<-TukeyHSD(x= aov_mod)
  tuk
  

})

out


# Code 2
# In order to generate 5 csv files, copy and paste the code below and save in a new folder
# 

state <-c("NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY")

dept <- c("energy","energy","energy","energy","works",'works','works','works','fin','fin','fin','fin','parks','parks','parks','parks','trans','trans','trans','trans')
year <- c("two","two","two","two","two","two","two","two","two","two","two","two","two","two","two","two","two","two","two","two")
revenue <-c(1212.9,1253,1244.4,5123.5,1312,3134,515.8,2449.9,3221.6,3132.5,2235.09,2239.01,3235.01,5223.01,4235.6,2204.5,2315.5,6114,4512,3514.2)
NY_two <-data.frame(state,dept,year,revenue)
# dataset 2
state <-c("NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY","NY")

dept <- c("energy","energy","energy","energy","works",'works','works','works','fin','fin','fin','fin','parks','parks','parks','parks','trans','trans','trans','trans')
year <- rep("five",20)
revenue <-c(1212.9,1253,1244.4,5123.5,1312,3134,515.8,2449.9,3221.6,3132.5,2235.09,2239.01,3235.01,5223.01,4235.6,2204.5,2315.5,6114,4512,3514.2)
NY_five <-data.frame(state,dept,year,revenue)



state <- rep("FL",20)
dept <- c("energy","energy","energy","energy","works",'works','works','works','fin','fin','fin','fin','parks','parks','parks','parks','trans','trans','trans','trans')
year <- rep("three",20)
revenue <-c(112.9,123,124,523.5,112,334,55,449,221.6,332,235,239,235,223,235.6,204,315.5,614,512,514.2)
FL_three <- data.frame(state,dept,year,revenue)



state <- rep("CA",20)
dept <- c("energy","energy","energy","energy","works",'works','works','works','fin','fin','fin','fin','parks','parks','parks','parks','trans','trans','trans','trans')
year <- rep("three",20)
revenue <-c(1102.9,1023,1024,5203.5,1012,3034,505,4049,2021.6,3032,2035,2039,2035,2023,2035.6,2004,3015.5,6014,5012,5014.2)
CA_three <- data.frame(state,dept,year,revenue)


state <- rep("KY",20)
dept <- c("energy","energy","energy","energy","works",'works','works','works','fin','fin','fin','fin','parks','parks','parks','parks','trans','trans','trans','trans')
year <- rep("one",20)
revenue <-c(1102.9,1023,1024,5203.5,1012,3034,505,4049,2021.6,3032,2035,2039,2035,2023,2035.6,2004,3015.5,6014,5012,5014.2)
KY_one <- data.frame(state,dept,year,revenue)



setwd("C:\\define your path\\")

# writing out the file to a newly created folder
write.csv(NY_two,"NY_two.csv",row.names = FALSE)
write.csv(CA_three,"CA_three.csv",row.names = FALSE)
write.csv(FL_three,"FL_three.csv",row.names = F)
write.csv(NY_five,"NY_five.csv",row.names = FALSE)
write.csv(KY_one,"KY_one.csv",row.names = F)

Please share your code, thanx in advance!

@akrun, Please can you share you code (add comments too). thanks — RayX500, Mar 09 '22 at 16:55
I believe you are reading from the `files` i.e. `files <- list.files("C:\\mypath\\", pattern="*.csv", full.names = TRUE)` Are those file names you wanted i.e. `filenms <- sub("\\.csv$", "", basename(files))` and then when you get the output set the names to `names(out) <- filenms` — akrun, Mar 09 '22 at 16:57
Once you have named it, then it is easier i.e. `purrr::iwalk(out, ~ write.csv(.x, paste0(.y, ".csv"), row.names = FALSE)` — akrun, Mar 09 '22 at 17:01
@akrun, post your code in answer section (instead of the comment section), this will help me understand your logic. Also, so I can vote your response as answered. Really appreciate your effort — RayX500, Mar 09 '22 at 17:06
Thanks, I posted as a solution. I didn't answer earlier as I was not sure whether that is what you wanted — akrun, Mar 09 '22 at 17:10
Just wanted to mention that I updated the solution as your TukeyHSD is a list and it may be better to make it a tibble with `tidy` from `broom` — akrun, Mar 09 '22 at 17:15

akrun · Accepted Answer · 2022-03-09T19:53:37.800

2

The 'out' list doesn't have any names because it was not named. If the names should come from the files part, we may name the output ('out') with the substring of file names

# returns all the file paths for csv
files <- list.files("C:\\mypath\\", pattern="\\.csv$", full.names = TRUE) 
# get the substring of file names without the .csv part
filenms <- sub("\\.csv$", "", basename(files))

Now, we use the same code as in the OP's post or just loop over the files directly

out <- lapply(files, function(x) {  # use lapply to loop over all files
  
  this_data <- read.csv(x, header = TRUE)
  aov_mod <- aov(revenue ~ dept, data = this_data)
  tuk <- TukeyHSD(x= aov_mod)
  # output of TukeyHSD is a list which can be summarised into tibble
  # with tidy from broom
  broom::tidy(tuk)
  

})
# set the names with filenms
names(out) <- filenms

If we want to write the output use imap/iwalk from purrr which is concise as .y returns the names and .x returns the value of the list

purrr::iwalk(out, ~ write.csv(.x, paste0(.y, ".csv"), row.names = FALSE))

If we want pdf file, an option is to use tableGrob

library(grid)
library(gridExtra)
library(gtable)
# reproducible exmaple
out <- list(iris = head(iris), mtcars = head(mtcars))
pdf(file.path(getwd(), "Downloads/filepdfMar9.pdf"))
iwalk(out,  ~ {
     
     title <- textGrob(.y, gp = gpar(fontsize = 50))
     padding <- unit(0.5,"line")
     x1 <- gtable_add_rows(
     tableGrob(.x), heights = grobHeight(title) + padding, pos = 0
       )
    x1 <- gtable_add_grob(
      x1, list(title),
        t = 1, l = 1, r = ncol(x1)
    )
grid.newpage()
grid.draw(x1)
})
dev.off()

-output

edited Mar 09 '22 at 19:53

answered Mar 09 '22 at 17:10

akrun

874,273
37
540
662

Everything works up until the purr::walk, I got this error `Error in as.data.frame.default(x[[i]], optional = TRUE, stringsAsFactors = stringsAsFactors) : cannot coerce class ‘c("TukeyHSD", "multicomp")’ to a data.frame` – RayX500 Mar 09 '22 at 17:17
@RickyTricky can you try the `out` updated. may be you used earlier version – akrun Mar 09 '22 at 17:19
works well thanks alot. One last thing, instead of exporting the final output as csv (`purrr::iwalk`) can you export them as 1 pdf file with multiple pages? – RayX500 Mar 09 '22 at 17:22
@RickyTricky then you may need to convert as image and use `pdf` – akrun Mar 09 '22 at 17:47
i hope I am not asking for too much, can you modify the code to convert as image and export as pdf. Thanks alot – RayX500 Mar 09 '22 at 18:07
@RickyTricky you may try `library(gridExtra); pdf("yourfile.pdf", height=11, width=10); iwalk(out, ~ grid.table(.x)); dev.off()` – akrun Mar 09 '22 at 18:10
where do I place the `pdf("your file.pdf")`....Please include al the suggested codes in the previous code you provided earlier. I will prefer it to be 1 large pdf file with multiple page instead of multiple pdf files. I really appreciate your assistance, I'm still pretty new to R – RayX500 Mar 09 '22 at 18:17
@RickyTricky yes, that is why you have to include the `pdf(...)` command before the loop in `iwalk`, which should generate multiple pages – akrun Mar 09 '22 at 18:18
@RickyTricky please check the update in my post – akrun Mar 09 '22 at 18:25
thanks alot. the output is not showing the corresponding file names ( I have added an example of the output in my original post). How can we include filenames each page.? – RayX500 Mar 09 '22 at 18:57
@RickyTricky you can modify the one I updated. Check this [link](https://stackoverflow.com/questions/11774703/adding-text-to-a-grid-table-plot) as well – akrun Mar 09 '22 at 19:10
How do i automate the `TableGrob` for all the tables? – RayX500 Mar 09 '22 at 19:50

How to append input file name to corresponding list output index in R

1 Answers1