How to open CSV file in R when R says "no such file or directory"?

Question

I have an excel file that I want to open in R. I tried both of these commands after saving the excel file as a csv file or a text file.

read.table() or read.csv()

I think part of the problem is where the file is located. I have it saved on the desk top. What am I missing here?

Here is the R output

In file(file, "rt") :
  cannot open file 'Rtrial.csv': No such file or directory
> help.search("read.csv")
> read.csv("Rtrial.csv")
Error in file(file, "rt") : cannot open the connection
In addition: Warning message:
In file(file, "rt") :
  cannot open file 'Rtrial.csv': No such file or directory
> read.table("tab")

It would in general be useful to provide more information; in this particular case it's hard to make a truly reproducible example, but you could at least show the full command you tried, and the output of getwd() [i.e., show your working directory] — Ben Bolker, Apr 06 '11 at 16:28
for general advice of opening excel files (xls, xlsx), see http://stackoverflow.com/questions/6099243/read-an-excel-file-directly-from-a-r-script/6100369#6100369 — Jeromy Anglim, Oct 10 '12 at 03:46

score 44 · Accepted Answer · answered Apr 06 '11 at 17:54

44

To throw out another option, why not set the working directory (preferably via a script) to the desktop using setwd('C:\John\Desktop') and then read the files just using file names

answered Apr 06 '11 at 17:54

Maiasaura

32,226
27
104
108

+1: this is a great idea; if you change your directory structure then all that needs to change is this one line. – Aaron left Stack Overflow Apr 06 '11 at 20:35
1

Still not clear to me why the original poster was getting a permissions error in one of the iterations, then ... – Ben Bolker Apr 06 '11 at 22:55
for unix based, should be like this ```setwd("~/dir")```. – affhendrawan Nov 02 '17 at 07:21

score 23 · Answer 2 · answered Apr 06 '11 at 15:28

23

Try

f <- file.choose()

to choose the file interactively and save the name in f.

Then run read.csv on the saved filename

d <- read.csv(f)

answered Apr 06 '11 at 15:28

Aaron left Stack Overflow

36,704
7
77
142

3

But then you have to choose the file manually every time you run the script. – Benjamin Apr 06 '11 at 15:29
True. This is more useful for exploratory analysis than scripting. It also can help to figure out the proper name and path of a file, which can then be written out as in Benjamin's answer. – Aaron left Stack Overflow Apr 06 '11 at 15:48

score 14 · Answer 3 · answered Apr 06 '11 at 15:26

14

Sound like you just have an issue with the path. Include the full path, if you use backslashes they need to be escaped: "C:\\folder\\folder\\Desktop\\file.csv" or "C:/folder/folder/Desktop/file.csv".

myfile = read.csv("C:/folder/folder/Desktop/file.csv")  # or read.table()

It may also be wise to avoid spaces and symbols in your file names, though I'm fairly certain spaces are OK.

answered Apr 06 '11 at 15:26

Benjamin

11,560
13
70
119

Error in file(file, "rt") : cannot open the connection In addition: Warning message: In file(file, "rt") : cannot open file 'C:\Documents and Settings\Manley\Desktop': Permission denied > – illbill Apr 06 '11 at 16:09
@illbill: Now it seems to be a permissions issue. Is it read only? Do you have read rights? – Benjamin Apr 06 '11 at 16:10
@Svun is probablity right. R won't read the file if it is already open in Excel. – Benjamin Apr 06 '11 at 20:17

score 8 · Answer 4 · answered May 28 '16 at 08:19

I had to combine Maiasaura and Svun answers to get it to work: using setwd and escaping all the slashes and spaces.

setwd('C:\\Users\\firstname\ lastname\\Desktop\\folder1\\folder2\\folder3')
data = read.csv("file.csv")
data

This solved the issue for me.

score 6 · Answer 5 · answered Apr 06 '11 at 16:28

6

Here is one way to do it. It uses the ability of R to construct file paths based on the platform and hence will work on both Mac OS and Windows. Moreover, you don't need to convert your xls file to csv, as there are many R packages that will help you read xls directly (e.g. gdata package).

# get user's home directory
home = setwd(Sys.getenv("HOME"));

# construct path to file
fpath = file.path(home, "Desktop", "RTrial.xls");

# load gdata library to read xls files
library(gdata);

# read xls file
Rtrial = read.xls(fpath);

Let me know if this works.

answered Apr 06 '11 at 16:28

Ramnath

54,439
16
125
152

1

On WinXP (and possibly other Windows) the `HOME` environment variable points to `My Documents`, and the desktop is in the parent of that folder. You either need `Sys.getenv("USERPROFILE")` if it exists, or `paste(Sys.getenv("HOMEDRIVE"), Sys.getenv("HOMEPATH"), sep = "")`. – Richie Cotton Apr 06 '11 at 17:02
@Richie. Thanks for the pointer. I use Mac OS, so was assuming that HOME would point to the parent of desktop. – Ramnath Apr 06 '11 at 20:16
On the mac HOME points to your /Users folder (usually this is your name), so yes, this is the parent folder of the Desktop – Benjamin Apr 06 '11 at 20:19

score 4 · Answer 6 · answered Apr 06 '11 at 17:42

4

Save as in excel will keep the file open and lock it so you can't open it. Close the excel file or you won't be able to use it in R.
Give the full path and escape backslashes read.csv("c:\\users\\JoeUser\\Desktop\\JoesData.csv")

answered Apr 06 '11 at 17:42

Svun

81
2

Hasnain · Answer 7 · 2019-11-21T07:53:50.077

I have experienced that this error occurs when you either move the excel file to the destination other than where your r file is located or when you move your r file to the destination other than where your excel file is located.

Good Practice:

Keep your .r and .csv files in the same directory.
open your .r file from getting into its directory instead of opening the r file from rstuio's open file option.

You also have import Dataset option at Environment Block, just click there and get your required packages installed & from next time use this option to read datasets. You will not get this error again. I also appreciate the above provided answers.

Yes. Even if the working directory is set to where the csv file is located; as long as the R/Rmd file is NOT in that directory, it will not find the csv. — FLonLon, Mar 25 '21 at 12:16

score 1 · Answer 8 · answered Feb 03 '18 at 18:27

My issue was very simple, the working directory was not the "Source" directory that was printed when the file ran. To fix this, you can use getwd() and setwd() to get your relative links working, or just use a full path when opening the csv.

print(getwd()) # Where does the code think it is?
setwd("~/Documents") # Where do I want my code to be?
dat = read.csv("~/Documents/Data Visualization/expDataAnalysis/one/ac1_survey.csv") #just make it work!

score 1 · Answer 9 · answered Apr 07 '11 at 09:19

Another way of reading Excel including the new format xlsx could be the package speedR (https://r-forge.r-project.org/projects/speedr/). It is an interactive and visual data importer. Besides importing you can filter(subset) the existing objects from the R workspace.

score 0 · Answer 10 · answered Mar 20 '14 at 01:32

MAC OS It happened to me as well. I simply chose from the R toolbar MISC and then chose Change Working Directory. I was able to choose the directory that the .csv file was saved in. When I went back to the command line and typed getwd() the full directory was updated and correct and the read.csv function finally worked.

score 0 · Answer 11 · answered Feb 11 '15 at 19:44

I had the same problem and when I checked the properties of the file on file explorer, it shows me the next message:

"Security: This file came from another computer and might be blocked to help protect this computer"

You click on the "Unblock" button and... you can access to the file from R without any problem, just using read.csv() function and from the directory specified as your working directory, even if is not the same as the file’s directory you are accessing to.

score 0 · Answer 12 · answered Oct 01 '15 at 15:09

0

I just had this problem and I first switched to another directory and then switched back and the problem was fixed.

answered Oct 01 '15 at 15:09

user5398097

1

score 0 · Answer 13 · answered Jun 26 '18 at 06:55

0

this work for me, accesing data from root. use double slash to access address.

dataset = read.csv('C:\\Users\\Desktop\\Machine Learning\\Data.csv')

answered Jun 26 '18 at 06:55

Andrey Cls

35
6

score 0 · Answer 14 · edited Dec 31 '19 at 10:23

0

Kindly check whether the file name has an extension for example:
```
abc.csv
```
if so remove the .csv extension.
set wd to the folder containing the file (~)
data<-read.csv("abc.csv")

Your data has been read the data object

edited Dec 31 '19 at 10:23

Peyman Mohamadpour

17,954
24
89
100

answered Dec 31 '19 at 09:40

kelvinkahuro

101
1
3

score 0 · Answer 15 · answered Feb 12 '20 at 16:56

0

In my case this very problem was raised by wrong spelling, lower case 'c:' instead of upper case 'C:' in the path. I corrected spelling and problem vanished.

answered Feb 12 '20 at 16:56

Przemyslaw Remin

6,276
25
113
191

score 0 · Answer 16 · answered Dec 02 '22 at 04:34

You can add absolute path to the file

heisenberg <- read.csv(file="C:/Users/tiago/Desktop/sample_100000.csv")

If really want to run something like

heisenberg <- read.csv(file="sample_100000.csv")

then you'll have to change the working directory to match the place the .CSV file is at. More about it here.

How to open CSV file in R when R says "no such file or directory"?

16 Answers16

Linked

Related