CODEWITHSUNDEEP
rcsv
Sonu Mishra
Sonu Mishra

Reputation: 1779

Reading a long comma separated line to a column in R

I have a very long comma separated string that I want to read as a column vector in R. I tried to read it as a table, wondering if I could just transpose that, but reading into table takes forever. 

col = read.table("filename.csv", sep = ",", header=FALSE)
colT = t(col)

The string is huge and has 2.7M entries separated by commas. Therefore it cannot be handled in a text editor and trying to replace ',' by '\n' was futile. Is there a way I can do that in R?

Upvotes: 2

Views: 1107

Answers (2)

akrun
akrun

Reputation: 887531

We can use fread

library(data.table)
fread("filename.csv", header=FALSE)

tmp <- paste(paste0(letters,1:2.7e6),collapse="\n")
system.time(fread(tmp, header=FALSE))
#  user  system elapsed 
#  0.87    0.00    0.88

If the OP's data have , and not \n, we can use gsub as @thelatemail mentioned

fread(gsub(",","\n",tmp),header=FALSE)

Upvotes: 3

thelatemail
thelatemail

Reputation: 93908

?scan will be quite quick for this sort of thing.

tmp <- paste(paste0(letters,1:2.7e6),collapse=",")
system.time(scan(text=tmp, what=character(1), sep=","))
#Read 2700000 items
#   user  system elapsed 
#   1.15    0.00    1.16

Upvotes: 3

Related Questions