split_pdf on Windows PC

[pohndorff]

I have an issue migrating my code from MacOS to Windows devices.
The following code snippet works just fine on my MacOS but hangs itself after the first iteration:

include
path <- list.files(path = paste0(getwd(),"/raw"),
                   pattern = "*.pdf",
                   recursive = TRUE,
                   full.names =  TRUE)
file <- list.files(path = paste0(getwd(),"/raw"),
                   pattern = "*.pdf",
                   recursive = TRUE)

for(i in 1:length(file)){
  file.path <- str_extract(file[i],".*(?=(KV[0-9]+\\.pdf))")
  dir.create(file.path(paste0(getwd(),"/output/",file.path)), showWarnings = FALSE)
  split_pdf(input_filepath = path[i],
            output_directory = file.path,
            prefix = paste0("p","_"))
}

I installed pdftk on both devices and they work fine, so I guess it's an issue regarding staplr.

Any suggestions to quickly solve the issue?

[oganm]

Probably related to slash directions on the filepath

[pohndorff]

Yeah, stupid me 🤦‍♂️
Can be closed now.

[oganm]

no no. it's our fault if R/unix style slashes break things on windows. I just wrote it as the potential issue to return for a fix later

[oganm]

About it hanging itself, does it happen on the mac? See #7 if it does.

About the slash issue, were you actually able to fix it by changing slash directions? Because I can't replicate the issue on my windows

[pohndorff]

The infinite loop only occurred on Windows so far.

Regarding the solution: I used the file.path() function on it so it autocorrected on any OS. My misconception was due to lacking documentation. I’d suggest to describe to attribute types more precisely. For this take reference to Hadley Wickhams book on R packages http://r-pkgs.had.co.nz/man.html#man-functions

[oganm]

Could you give me the path that failed. Because based on my experimentation in windows, it works with any combination of \s and /s.

I tried to replicate your code by making some assumptions about your file structure. I placed a pdf at the path raw/dir/KV111.pdf

Based on this structure your code doesn't and shouldn't work regardless of the OS as it is trying to place the output at dir directory which doesn't exist because the directory it created is actually called output/dir (assuming output directory already exists as recursive = FALSE in dir.create). I imagine some pieces were lost during the copy paste process here but it's making it hard for me to understand what exactly didn't work in the first place.

If I understand what went wrong to begin with I can see if some explanation could be added to the docs