Edit several dataframes in loop with column names and adding columns

Question

I have 10 datasets in a folder, with 4 columns, which I wish to read in as seperate dataframes in r, for which I use the following to do:

temp = list.files(pattern="*.csv")
for(i in 1:length(temp)){
  assign(paste("name",i,sep = ""), as.data.frame(read.table(temp[i])))
}

Then if i want to change the column names as well as adding a new column V5 <- V3**2 in either the same loop or a different loop, how could this be done.

The other suggestions for changing column names i've seen here in stackoverflow suggest creating a list of columns and then changing them. But they dont change the data in the global environment.

Could any of you help with this?

Many thanks.

I discourage the use of assign in almost every situation. In this case, I'd suggest the data be in a list, ala x <- lapply(temp, read.table). If you need to add columns, you can do x <- lapply(x, function(L) transform(L, V5=V3^2)). — r2evans
– r2evans, Commented Feb 16, 2019 at 18:09
thanks, could one then also just use lapply to change column names of those file columns? — RAHenriksen
– RAHenriksen, Commented Feb 16, 2019 at 18:14
Certainly. You can do whatever you want. If you want to change the names in just one of them, then you can do colnames(x[[3]]) <- c(...). If you want to change the second column name in all of them, then x <- lapply(x, function(L) { colnames(L)[2] <- "quux"; L; }). — r2evans
– r2evans, Commented Feb 16, 2019 at 18:24

Soren · Accepted Answer · 2019-02-16 18:23:42Z

1

The following will read-in the .csv files in "path" , unifying their column names and adding an additional computational column and then combine them all into a single data fame.

path <- ""
temp <- list.files(path=path,pattern="*.csv",full.names = T)
dfs <- lapply(temp,function(x)
  {
    df <- read.csv(x,stringsAsFactors = F,col.names=c("col1","col2","col3","col4"))
    df$col5 <- 1*2
    df
  })

do.call("rbind",dfs)

answered Feb 16, 2019 at 18:23

Soren

2,4851 gold badge18 silver badges18 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

r2evans Over a year ago

I think there is a slight blur of concepts here. The OP asked about dealing with multiple frames but never suggested that they be combined, so the do.call at the end is presumptuous. (It is certainly valid in many situations, I don't know that it is here.) Good safe use of full.names, I always encourage it in a defensive coding posture.

Amit Gupta · Accepted Answer · 2019-02-16 18:40:14Z

0

Rename all the datasets in an order like df-01, df-02... df-10 and read like following

   for(ii in 2:5){
       input_csv <- sprintf('sample_-%02d.csv', ii)
       read.csv(input_csv, stringsAsFactors = F,col.names=c("col1","col2","col3","col4"))
       print(input_csv)
       df$V5 <- df$V3**2
    }

answered Feb 16, 2019 at 18:40

Amit Gupta

2,9784 gold badges32 silver badges41 bronze badges

Collectives™ on Stack Overflow

Edit several dataframes in loop with column names and adding columns

2 Answers 2

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related