重命名R中的多个列

Question

I have 1000 files with column similar column names.for example :

df1

DATE PRICE CLOSE           

df2

DATE PRICE CLOSE 

and so on...

If I try to merge them based by date they do get merge but the columns have retained their old names and I want to rename them in a loop

so merge data set looks like this

Date Price Close PRICE CLOSE

I want something like

DATE PRICE1 CLOSE1 PRICE2 CLOSE2.

Is there any easy way to do it? I have tried couple of things which is not giving me correct output

this is using plyr package:

mod_join = function(mypath){
  filenames=list.files(path=mypath, full.names=TRUE)
  datalist = lapply(filenames, function(x){read.csv(file=x,header=T)[,c('Date','High','Low')]})
  join_all(datalist,by = "Date")
} 

this is using merge command on all data frame:

merge2 = function(mypath){
  filenames=list.files(path=mypath, full.names=TRUE)
  datalist = lapply(filenames, function(x){read.csv(file=x,header=T)[,c('Date','High','Low')]})
  Reduce(function(x,y) {merge(x,y,by.x= "Date",by.y = "Date",all=T)}, datalist)}

}  

I tried using for loop by making the data frame lead then using each data frame to subset and merge subsequently but somehow its not subsetting the dataframes:

for (i in 1:1000){
  data_subset <- sprintf('data_%d',i)
  mydata_subset <- data.frame(,data_subset["Date"],data_subset["High"],data_subset["DayLow"])
  obj_name <- paste('subset_Pricedata',i,sep ="_")
  assign(obj_name,value = mydata_subset)
} 

Any help will be great. Thanks

Answer 1

Hopefully, this will do your job:

library(plyr)
df1 = rename(df1,c("PRICE"="PRICE1","CLOSE"="CLOSE1"))
df2 = rename(df2,c("PRICE"="PRICE2","CLOSE"="CLOSE2"))
new = merge(df1,df2,all=TRUE)

Please comment if you face any difficulties.

Answer 2

What about this approach? It should be fast as it uses data.table and its fread-function

library(data.table)
merge2 <- function(mypath){
     filenames <- list.files(path=mypath, full.names=TRUE)
     fileslist <- lapply(filenames, function(nam){
            # reads the file
            file <- fread(nam)
            setnames(file, 2, "price") # renames the second col to "price"
            setnames(file, 3, "close") # third to "close"
            return(file)
     })
     dat <- rbindlist(fileslist)
     return(dat)
}  

EDIT

I just realised that you want to merge your data instead of having it in the long format. What you can do is just add a variable with a name to the data.table "file" before returning the file by adding:

file[, varnam := nam]

and then cast the final data.table "dat" before returning it, using the reshape2 library and its dcast function.

Answer 3

I had a similar problem. Here's what I ended up using, although there is likely a cleaner way.

The function suffix_col_names will add a suffix to a subset of columns. I use this because I eventually merge week1 and week 2 data on columns 1-10.

#function called suffix_col_names
suffix_col_names<-function(your_df, start_col, end_col, your_str, your_sep){
  for (i in start_col:end_col){
    colnames(your_df)[i]<-paste(colnames(your_df)[i], sep=your_sep,your_str)
  }
  return(your_df)
}
#call function to rename columns in week1 and week2
week_1_data<-suffix_col_names(week1,11,24,"1",".")
week_2_data<-suffix_col_names(week2,11,24,"2",".")