问题:

In d:\folder, I have many Stata files, such as data_aa_1.dta, data_aa_2.dta, data_aa_3.dta data_bb_1.dta, data_bb_2.dta, data_bb_3.dta, data_cc_1.dta.... I want to convert those files and get as many data frames as the dta files in R. So, I believe I have to loop over c("aa", "bb", "cc") and c(1:3). I tried something like the following:

library(foreign)

for(i in c("aa","bb","cc"){
    for (j in 1:3){
        data_[i]_[j] <-read.dta("d:/folder/data_[i]_[j].dta")
    }
}

But, that looks wrong -- definitely.

Any help will be appreciated.

Thanks!

回答1:

There is undoubtedly a more elegant solution, however

library(foreign)



for(i in c("aa","bb","cc"){
    for (j in 1:3){
      obj_name <- paste('data', i, j, sep ='_')
      file_name <- file.path('d:/folder',paste(obj_name,'dta', sep ='.'))
      input <- read.dta(file_name)
      assign(obj_name, value = input)

    }
}

EDIT

Avoiding the for loop, and using @joran's suggestion of list.files

dta_files <- list.files('d:/folder', pattern = '.dta', full.names = T)

lapply(dta_files, function(fname){
  input <- read.dta(fname) 
  obj_name <- tools::file_path_sans_ext(basename(fname))
  assign(obj_name, value input, env = .GlobalEnv)})

回答2:

Try this,

fl = list.files(pattern = "dta", path = "d:/folder", 
                full.names = TRUE)
dl = lapply(fl, foreign::read.dta)
names(dl) = tools::file_path_sans_ext(fl)
str(dl)