Trouble with reading CSV file in R

Question

I'm a newbie in R. I have a 24MB CSV file. Reads this into RStudio on my MacBook Air with OS Yoswmite, 4GB RAM. R version 3.1.1 (2014-07-10). Viewing the contents of View (df) is OK. Attempting to apply filter. Do not get any hits. Attempting to cast from character to number. R replacing all charaters with NA in the column where the casting is done! What happens here? It seems that R can not read the contents of the cells. Is there anything about encoding? This is what I have done so fare: First a summary:

R Code:

eiendommer <- read.csv("eiendommer.csv",  sep = ";", quote = "",  encoding="UTF-8", stringsAsFactors = FALSE)
View(eiendommer)# I can view the content of the csv file
filtereiendommer <- filter(eiendommer, kommune == "0101")# no match
filtereiendom <- eiendommer [eiendommer$kommune == "0101",]#no match
utvalg <- eiendommer[160567:161934,]#manual selection of rows do work             utvalgsortert <- arrange(utvalg, desc(jordbruksareal), desc(skogareal))# works
View(utvalgsortert)
##Try to transform columns from character to number. 
transformedEiendom <- transform(sortertEiendom, jordbruksareal = as.numeric(jordbruksareal),
                       skogareal = as.numeric(skogareal) )
#This result in NA where it earlier was characters with lengt 1-3:"646", "18", "2"

Summary:

kommune           X.gardsnr.         X.bruksnr.         X.festenr.         bruksnavn         jordbruksareal    
 Length:207554      Length:207554      Length:207554      Length:207554           Length:207554      Length:207554     
 Class :character   Class :character   Class :character   Class :character   Class :character   Class :character  
 Mode  :character   Mode  :character   Mode  :character   Mode  :character   Mode  :character   Mode  :character  
 X.annetareal.       skogareal         X.fulldyrket.        X.overflatedyrket. X.innmarksbeite.  
 Length:207554      Length:207554      Length:207554      Length:207554      Length:207554     
 Class :character   Class :character   Class :character   Class :character   Class :character  
 Mode  :character   Mode  :character   Mode  :character   Mode  :character   Mode  :character

Head:

head(eiendommer)
  kommune X.gardsnr. X.bruksnr. X.festenr.    bruksnavn jordbruksareal X.annetareal. skogareal X.fulldyrket.
1  "0101"        "1"        "1"        "0" "PRESTEGÅRD"            "0"           "5"       "0"           "0"
2  "0101"        "1"        "6"        "0"         "MO"            "8"           "4"       "7"           "8"
3  "0101"        "1"        "9"        "0"  "BERG GÅRD"          "415"          "16"      "39"         "415"
4  "0101"        "2"        "1"        "0"     "BOBERG"          "467"          "22"     "276"         "463"
5  "0101"        "4"        "1"        "0"  "LUNDESTAD"          "877"          "62"     "793"         "837"
6  "0101"        "4"        "5"        "0"     "LEIREN"           "74"          "14"     "165"          "74"

Hei CodeR, your question is not how to read a CSV file but how to transform objects from one class to another. — Kvasir EnDevenir, Mar 17 '15 at 00:47

score 2 · Answer 1 · answered Mar 17 '15 at 00:37

2

It would appear that you have specified quote = "", where in fact you should have, and perhaps quote='"' or simply the default value would work.

See the example below

d <- data.frame(x='a',y='"a"',stringsAsFactors=FALSE)
d 
#   x   y
# 1 a "a"

For a regular character vector, print.data.frame will not enclose it in "

answered Mar 17 '15 at 00:37

mnel

113,303
27
265
254

Ok, if I understand you right: all my data have gotten quotes (" ") and this is beacuse ot the quotes statement in read.csv? I did this is order to read the file. Earlier I got this message: Warning message: In scan(file, what, nmax, sep, dec, quote, skip, nlines, na.strings, : EOF within quoted string When including the quotes statement in read.csv I was able to read the file. Thanks. – CodeR Mar 17 '15 at 08:59
And here I found the tip about quotes: http://stackoverflow.com/questions/17414776/read-csv-warning-eof-within-quoted-string-prevents-complete-reading-of-file – CodeR Mar 17 '15 at 09:03

score 0 · Answer 2 · answered Mar 17 '15 at 02:25

One challenge you will have with Norwegian Kommune Nummer, is those starting with 0 like Halden "0101".

#Prepare Data
    kommune = rep("0101", 6) 
    jordbruksareal<- c("5","4","16","22","62","14")
    skogareal <- c("0","8","415","463","837", "74")
    eiendommer <- cbind(kommune, jordbruksareal, skogareal)
    eiendommer <- as.data.frame(cbind(kommune, jordbruksareal, skogareal), stringsAsFactors=FALSE)
    #Transform into numeric
    str(eiendommer) #All is Character
    eiendommer$skogareal<-as.numeric(eiendommer$skogareal)
    eiendommer$jordbruksareal<-as.numeric(eiendommer$jordbruksareal)
    eiendommer$kommune<-as.numeric(eiendommer$kommune)
    str(eiendommer) #All is numeric, but losing first zer0
#Make a filter
    require(dplyr)
    filterA <- filter(eiendommer, eiendommer$jordbruksareal == "4")
    filter <- subset(eiendommer, eiendommer$kommune == 101)
#Treat Kommune Numbers
    eiendommer$kommune <- formatC(eiendommer$kommune, digits = 0, format = "f", width = 4, flag = 0)
    eiendommer$kommune <- sprintf("%04d",eiendommer$kommune)
    str(eiendommer)
    filter2 <- subset(eiendommer, eiendommer$kommune == "0101")

Hope this helps a little, Ha det bra!

score 0 · Accepted Answer · answered Mar 17 '15 at 19:01

0

This caused my problem in the file:

;"BLOMSTERHAGEN\"";

Changing it to:

;"BLOMSTERHAGEN";

Fixed the problem. No I can read.csv like this:

eiendommer <- read.csv("eiendommer.csv",  sep = ";", encoding="UTF-8", stringsAsFactors = FALSE)

Thanks

answered Mar 17 '15 at 19:01

CodeR

13
3

Trouble with reading CSV file in R

3 Answers3