R：從文件夾中讀取多個文本文件，然后將 csv 文件寫入同一文件夾（即，將文本文件的文件夾轉換為 csv 文件的文件夾）

Question

我想從空格分隔的文本文件的單個文件夾中讀取文件，然后將它們作為 csv 文件寫到同一文件夾中。

空格分隔的文本文件很煩人，第二行是破折號，每個文件末尾的三行帶有一些不正確的摘要值（見下文）。 這些被“read_fwf”中的“comment”和“skip”arguments 刪除。 如果有更好的方法來實現這一點，請指教。

    Frame   Clusters   Total Area
    -----------------------------
        0          2          353   
        1          2          233   
        2          1           18   
        3          1           18 
      ...        ...         ... 
     6230          1           38   
     6231          2          183   

    Total qualifying frame count = 6105
    Total file cluster count     = 42005
    Total file biomass area      = 8867980

我正在嘗試使用像 lapply 這樣的 tidyverse 選項而不是編碼循環。

我想我正在閱讀文本文件。 我的失敗似乎在 write.csv 部分代碼。

作為 R 的新用戶，我將非常感謝任何有關語法的建議或鏈接（即括號、縮進等）。 論壇帖子的清晰度提示也會有所幫助。 謝謝你。

    # Clear all variables
    rm(list = ls())

    # Reference tidyverse
    library(tidyverse)

    # Create variable, "folder" for location of text files
    folder <- "C:/Users/stai669/OneDrive - PNNL/Desktop/CRDC_P/DP/DIDSON snip 
    files/largeDataset/Cluster/Aft/N/"

    # Create variable, "file_list" that lists all files in folder for "lapply" to loop through
    filelist <- list.files(path = folder, pattern = "*.txt")

    # Read in each file with "lapply"; other arguments for column names, etc.
    data <- lapply(filelist, function(x) {
                      textfile <- read_fwf(
                      paste(folder, x, sep = ""),
                      fwf_positions(start = c(1, 6, 17), end = c(5, 16, 29),
                      comment = "T",
                      col_names = c("fram", "clust", "biom"),
                      skip = 2,
                      cols(fram = col_integer(), clust = col_integer(), biom =    col_integer())
                      )
    # Write the files to csv in the same folder                          
        write.csv(textfile, file = sub(pattern = "\\.txt$", replacement = ".csv", x = x))
        })

Answer 1

# Clear all variables
rm(list = ls())

# Reference tidyverse
library(tidyverse)

# Create variable, "file_list" that lists all files in folder for "lapply()" to loop through
filelist <- list.files(pattern = "*.txt")

# Read in each file with "lapply"; other arguments for column names, etc.
data <- lapply(filelist, function(x) {
                          textfile <- read_fwf(
                          paste(x, sep = ""),
                          fwf_positions(start = c(1, 6, 17), end = c(5, 16, 29), col_names = c("fram", "clust", "biom")),
                          comment = "T",
                          skip = 2,
                          cols(fram = col_integer(), clust = col_integer(), biom = col_integer())
                          )
# Write the files as csv to the same folder                          
            write.csv(textfile, file = sub(pattern = "\\.txt$", replacement = ".csv", x = x))
})

同樣，請注意“col_names”參數與“fwf_positions”一起使用。