R脚本可选工作目录参数解析报错，如何实现可选参数设置？

Let's break down the issues in your current code and fix them to make the --working_dir parameter optional, while resolving the setwd error and other hidden problems:

Key Issues in the Original Code #

1. Duplicate Parser Initialization: You overwrite the parser right after setting its prog and description, losing those critical settings.
2. NULL Working Directory Handling: When --working_dir isn't provided, args$working.dir is NULL. Your merge_em function checks missing(working.dir), which returns FALSE because you passed NULL explicitly (instead of omitting the argument). This leads to setwd(NULL) which throws the character argument error.
3. Passing Filenames Instead of Data Frames: Your merge_em function expects data frames for x and y, but you're passing filename strings directly—this would cause a merge error once the setwd issue is fixed.
4. Unnecessary setwd Usage: Changing the working directory can have unintended side effects; it's better to use full file paths instead to avoid disrupting other script operations.

Corrected Script #

library("argparse")
library("R.utils")

merge_em <- function(x_df, y_df, output_dir) {
  # Merge data frames using common column names
  merged_df <- merge(x_df, y_df, by = intersect(names(x_df), names(y_df)))
  
  # Construct full path for the output file (avoids changing working directory)
  output_path <- file.path(output_dir, "merged.txt")
  
  # Write merged data to the specified directory
  write.table(merged_df, output_path, 
              col.names = FALSE, row.names = FALSE, 
              sep = "\t", quote = FALSE)
  
  message("Successfully wrote merged file to: ", output_path)
}

main <- function() {
  options(error = traceback, warn = 1)
  
  # Initialize parser with correct script metadata
  parser <- ArgumentParser(prog = "merge_em.r", 
                           description = "Merge two tab-separated data frames based on common columns")
  
  # Add required positional arguments (input file paths)
  parser$add_argument("x", help = "Path to first input data frame file")
  parser$add_argument("y", help = "Path to second input data frame file")
  
  # Add optional output directory argument
  parser$add_argument(
    "--working_dir", dest = "output_dir", type = "character",
    metavar = "DIR", required = FALSE,
    help = "Directory to write the merged output file (default: current working directory)"
  )
  
  # Parse command line arguments
  args <- parser$parse_args()
  
  # Handle optional directory: default to current working directory if not provided
  if (is.null(args$output_dir)) {
    args$output_dir <- getwd()
    message("No output directory specified, using current working directory: ", args$output_dir)
  }
  
  # Resolve input file paths (convert relative paths to absolute using output directory)
  x_path <- args$x
  if (!isAbsolutePath(x_path)) {
    x_path <- file.path(args$output_dir, x_path)
  }
  
  y_path <- args$y
  if (!isAbsolutePath(y_path)) {
    y_path <- file.path(args$output_dir, y_path)
  }
  
  # Read input files into data frames (adjust read.table parameters if your files use a different format)
  tryCatch({
    x_df <- read.table(x_path, header = TRUE, sep = "\t", stringsAsFactors = FALSE)
    y_df <- read.table(y_path, header = TRUE, sep = "\t", stringsAsFactors = FALSE)
    
    # Execute merge function
    merge_em(x_df, y_df, args$output_dir)
  }, error = function(e) {
    stop("Error processing files: ", e$message)
  })
}

main()

Key Changes Explained #

1. Fixed Parser Setup: We keep only one ArgumentParser call to preserve the script name and description.
2. Robust Directory Handling:
   • If --working_dir isn't provided, we default to the current working directory using getwd().
   • We eliminate setwd entirely by constructing full paths for both input (if relative) and output files.
3. Proper Data Frame Reading: We read input files into data frames before passing them to merge_em, fixing the hidden merge error.
4. Clear Error Handling: Added a tryCatch block to provide meaningful error messages if file reading or merging fails.
5. Clarified Naming: Renamed working.dir to output_dir for clarity, since the directory is primarily used for writing the output (input paths can be relative to it or absolute).

How to Use #

• Default (use current directory):

  Rscript merge_em.r dataframe1.txt dataframe2.txt
  

• Specify output directory:

  Rscript merge_em.r dataframe1.txt dataframe2.txt --working_dir /path/to/output
  

This script will now handle optional working directory input correctly, resolve file paths properly, and merge your data frames without errors.

内容的提问来源于stack exchange，提问作者Achal Neupane