Skip to content

[newbie] read row from file into vector

7 messages · Rui Barradas, Duncan Murdoch, William Dunlap +3 more

Tom Roche
# 
summary: how to read a row (not column) from a file into a vector (not a data frame)?

details:

I'm using

$ lsb_release -ds
Linux Mint Debian Edition
$ uname -rv
3.0.0-1-amd64 #1 SMP Sun Jul 24 02:24:44 UTC 2011
$ R --version
R version 2.14.1 (2011-12-22)

I'm new to R (having previously used it only for graphics), but have worked in many other languages. I've got a CSV file from which I'd like to read the values from a single *row* into a vector. E.g., for a file such that

$ head -n 2 ~/data/foo.csv | tail -n 1
5718,0.3,0.47,0,0,0,0,0,0,0,0,0.08,0.37,0,0,0.83,1.55,0,0,0,0,0,0,0,0,0,0.00,2.48,2.33,0.17,0,0,0,0,0,0,0.00,10.69,0.18,0,0,0,0

I'd like to be able to populate a vector 'v' s.t. v[1]=5718, ... v[43]=0

I can't seem to do that with, e.g., read.csv(...) or scan(...), both of which seem column-oriented. What am I missing?
Rui Barradas
# 
Tom Roche wrote
Hello,

?readLines

If you want the 2nd line it's fast, if you have a large file, cycle through.

Rui Barradas


--
View this message in context: http://r.789695.n4.nabble.com/newbie-read-row-from-file-into-vector-tp4243952p4244038.html
Sent from the R help mailing list archive at Nabble.com.
Duncan Murdoch
#
On 11-12-29 3:51 PM, Tom Roche wrote:
Those are both column oriented, but you can change the result to a 
vector after reading it.

For example,

x <- read.csv("foo.csv", nrow=1)
x <- as.numeric(x[1,])  # convert to numeric vector

If you don't want the first row, use skip=<something> in the read.csv 
call.

Duncan Murdoch
William Dunlap
# 
Look into connection objects, which let you open a file or other
readable sort of thing and read it piece by piece.  E.g., the
following function opens your file (making a "file connection"),
skips some lines, reads the desired line into a character object,
then reads from that character object (as a "text connection")
to make a numeric object:

f <- function (fileName, lineNumber, ...) 
{
    connection <- file(fileName, "rt")
    on.exit(close(connection))
    if (lineNumber > 1) {
        readLines(connection, n = lineNumber - 1)
    }
    lineText <- readLines(connection, n = 1)
    scan(textConnection(lineText), ...)
}

Here is a self-contained example:
+     "101;102;103", "104,105", "106/107/108")
A data file with a header line
101;102;103
104,105
106/107/108
Read 2 items
[1] 104 105

Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com
Greg Snow
# 
The scan function can be used to read a single row.  If your file has multiple rows you can use the skip and nlines arguments to determine which row to read.  With the what argument sent to a single item (a number or string depending on which you want) it will read each element on that row into a vector.

If you want to do more of the hard work yourself you can read in a whole line as a single string using the readLines function then use the strsplit (or possibly better, tools from the gsubfun package) to split that string into a vector (the unlist function may also be of help).
Tom Roche
# 
Tom Roche 11-12-29 3:51 PM
Duncan Murdoch Thu, 29 Dec 2011 16:45:45 -0500
Aha!

William Dunlap Thu, 29 Dec 2011 21:49:13 +0000
Will do, since what I plan to use R for (mostly) is manipulating very large netCDF files.

thanks, all! Tom Roche <Tom_Roche at pobox.com>
Daniel Nordlund
# 
Tom,

you might also want to see what is already available in R for handling netCDF files.  Install the sos package (if you haven't already) and run the command

findFn('netCDF') 

It returned quite a lot of functions for dealing with netCDF  files/data.


Hope this is helpful,

Dan

Daniel Nordlund
Bothell, WA USA