Skip to contents

File Management

Source: vignettes/file-management.Rmd
file-management.Rmd

This packages has a couple functions that will help managing how you import files into R.

get_file_path()

Basic usage

This function is useful when you have multiple versions of a file in a folder (think: cumulative exports saved weekly) and want to retrieve the path for a specific file.

get_file_path() has five arguments:

  • directory: a string specifying the directory in which you want to look. It defaults to the working directory (getwd())
  • pattern: a string and/or regular expression specifying which files you’re interested in
  • sort_method: “created date” (the default), “modified date” or”accessed date”
  • sort_type: “newest” (the default) or “oldest”
  • include_directories: boolean to indicate whether or not directories should be included in your results. It defaults to FALSE

If you run the function without specifying any arguments (get_file_path()), it will return the file with the newest create date in your current working directory.

Additional options

Some options for using this function:

Get the file that was most recently accessed in a directory:

get_file_path(sort_method = "accessed date", sort_type = "newest")

Get the first file that was created in a different directory:

get_file_path(directory = "data", sort_type = "newest")

Get the most recently modified parquet file in a directory

get_file_path(directory = "data", pattern = "*.parquet",
              sort_method = "modified date", sort_type = "newest")

read_tsv_cr()

Basic usage

This is a wrapper function build on utils::read.csv2(). It is meant to be used for large tab separated value (.tsv) files, including CalREDIE tab separated files from the DDP.

If you’re reading in a large .tsv file, you can just pass the path of the file through the function:

data <- read_tsv_cr("export.tsv")

Additional options

You can also pass any of the utils::read.csv2() arguments to read_tsv_cr() though if you want to specify a lot of arguments, you might be better off just using utils::read.csv2().