如何通过循环同时读取多文件夹CSV文件并分配至不同对象

Hey there! Let's get your code sorted for reading those CSVs across four folders, and also flag a big efficiency issue you might not have caught yet. First, let's fix a few small syntax mistakes in your original code:

• R uses list.files() to get file listings, not list.dir()
• For Windows paths, use forward slashes / or double backslashes \\ to avoid escape character headaches
• R is case-sensitive, so it's stringsAsFactors, not stringasfactors

Corrected & Complete Nested Loop Code #

Here's the polished version of your nested loop, with proper file reading and storage (so you don't end up with thousands of loose variables cluttering your environment):

# Define your folder paths (fixed path format for cross-OS compatibility)
milk_loc <- "C:/Desktop/milk"
cereal_loc <- "C:/Desktop/cereal"
OJ_loc <- "C:/Desktop/OJ"
cat_loc <- "C:/Desktop/cat"

# Grab only CSV files from each folder (filters out non-CSV clutter)
milk_files <- list.files(milk_loc, pattern = "\\.csv$", full.names = FALSE)
cereal_files <- list.files(cereal_loc, pattern = "\\.csv$", full.names = FALSE)
OJ_files <- list.files(OJ_loc, pattern = "\\.csv$", full.names = FALSE)
cat_files <- list.files(cat_loc, pattern = "\\.csv$", full.names = FALSE)

# Initialize a list to store all file combinations (cleaner than loose variables)
all_data <- list()
counter <- 1

# Loop through every combination of files
for(milk_sheet_name in milk_files){
  # Read the current milk CSV (using file.path() for OS-agnostic path building)
  milk_sheet <- read.csv(file.path(milk_loc, milk_sheet_name), stringsAsFactors = FALSE)
  
  for(cereal_file_name in cereal_files){
    cereal_file <- read.csv(file.path(cereal_loc, cereal_file_name), stringsAsFactors = FALSE)
    
    for(OJ_cup_name in OJ_files){
      OJ_cup <- read.csv(file.path(OJ_loc, OJ_cup_name), stringsAsFactors = FALSE)
      
      for(cat_paw_name in cat_files){
        cat_paw <- read.csv(file.path(cat_loc, cat_paw_name), stringsAsFactors = FALSE)
        
        # Store the four data frames as a single entry in our list
        all_data[[counter]] <- list(
          milk = milk_sheet,
          cereal = cereal_file,
          oj = OJ_cup,
          cat = cat_paw,
          combo_label = paste(milk_sheet_name, cereal_file_name, OJ_cup_name, cat_paw_name, sep = "_")
        )
        
        # If you want to call your target function right here, add it:
        # your_target_function(milk_sheet, cereal_file, OJ_cup, cat_paw)
        
        counter <- counter + 1
      }
    }
  }
}

A Critical Warning: #

Wait a second—four folders with 400 files each means your nested loop would run 2.56 BILLION times (400^4). That's impossible to run in any reasonable time, and would eat up all your memory instantly. I'm guessing this isn't actually what you want!

The Far More Practical Alternative #

If your files are matched by filename (e.g., milk_001.csv pairs with cereal_001.csv, OJ_001.csv, and cat_001.csv), use a single loop instead. This makes way more sense and is completely manageable:

# Assuming all folders have the same number of matching files
file_count <- length(milk_files)
matched_data <- list()

for(i in 1:file_count){
  # Read corresponding files from each folder
  milk_data <- read.csv(file.path(milk_loc, milk_files[i]), stringsAsFactors = FALSE)
  cereal_data <- read.csv(file.path(cereal_loc, cereal_files[i]), stringsAsFactors = FALSE)
  oj_data <- read.csv(file.path(OJ_loc, OJ_files[i]), stringsAsFactors = FALSE)
  cat_data <- read.csv(file.path(cat_loc, cat_files[i]), stringsAsFactors = FALSE)
  
  # Store the matched set in our list
  matched_data[[i]] <- list(
    milk = milk_data,
    cereal = cereal_data,
    oj = oj_data,
    cat = cat_data,
    group_id = gsub("\\.csv$", "", milk_files[i]) # Extract the shared filename prefix
  )
  
  # Call your target function with the matched data
  # your_target_function(milk_data, cereal_data, oj_data, cat_data)
}

Extra Tips for Better Performance: #

• Use file.path() instead of paste0() to build paths—it works on Windows, Mac, and Linux without changes
• For large CSVs, swap read.csv() with data.table::fread()—it's way faster and uses less memory
• If you need speed, use the foreach package with doParallel to read files in parallel

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