R 用for循环读取文件

Question

I just want to use use 10 files in R. For each I want to calculate something. Exp. file: stat1_pwg1.out stat23_pwg2.out.. stat45_pwg10.out

I try this:

for (i in 1:10){
Data=paste("../XYZ/*_pwg",i,".out",sep="")
line=read.table(Data,head=T)
}

But it does not work? Any hinds?

Answer 1

I suspect your problem comes from the wildcard * . A better way to do this might be to first store the file names using dir , then find the ones you want.

files <- dir("../XYZ",pattern="stat[0-9]+_pwg[0-9]+\.out")
for(f in files) {
  line=read.table(Data,head=T)
}

You could also use one of the apply family of functions to eliminate the for loop entirely.

Answer 2

A few things about your code.

paste is vectorised, so you can take it out of the loop.

paste("../XYZ/*_pwg", 1:10, ".out", sep = "")

(Though as you'll see in a moment, you don't actually need to use paste at all.)

read.table won't accept wildcards; it needs an exact match on the file name.

Rather than trying to construct a vector of the filenames, you might be better using dir to find the files that exist in your directory, filtered by a suitable naming scheme.

To filter the files, you use a regular expression in the pattern argument. You can convert from wildcards to regular expression using glob2rx .

file_names <- dir("../XYZ", pattern = glob2rx("stat*_pwg*.out"))
data_list <- lapply(filenames, read.table, header = TRUE)

For a slightly more specific fit, where the wildcard only matches numbers than anything, you need to use regular expressions directly.

file_names <- dir("../XYZ", pattern = "^stat[[:digit:]]+_pwg[[:digit:]]+\\.out$")

Answer 3

    files <- dir(pattern="*Rip1_*")

    files

    for (F in files){ assign(F , Readfunc(F))}