[R] Reading data with 'awk' - basics?

[Gabor Grothendieck]

On Mon, Oct 17, 2011 at 9:23 AM, Brian Smith <bsmith030465 at gmail.com> wrote:
> Hi,
>
> I had a large file for which I require a subset of rows. Instead of reading
> it all into memory, I use the awk command to get the relevant rows. However,
> I'm doing it pretty inefficiently as I write the subset to disk, before
> reading it into R. Is there a way that I can read it into an R object
> without writing to disk? For example, this is what I do currently:
>
> ## write test sample file
> mat1 <- matrix(sample(1:100,16),8,2)
> fname1 <- 'temp1.txt'
> fname2 <- 'temp2.txt'
> write.table(mat1,fname1,sep='\t',row.names=F,col.names=F)
>
> ## Read a subset of rows, write to file, and read from file
> system(paste("awk '(NR > 1 && NR < 4) {print $0}' ",fname1," >
> ",fname2,sep=''))
> mat2 <- read.table(fname2,sep='\t')
>
> print(mat2)
> #####
>
> Is there a way that I can skip writing to disk?
>

See:

http://tolstoy.newcastle.edu.au/R/e5/help/08/09/2129.html

-- 
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com