Skip to content

reading multiple key=value pairs per line

7 messages · Christophe Pallier, PIKAL Petr, arun +2 more

Frank Singleton
# 
Hi,

Thanks for a great environmentfor statistical  computing :-)

I have some input data in a file ("input_kvpairs.csv") of the form

key1=23, key2=67, key3="hello there"
key1=7, key2=22, key3="how are you"
key1=2, key2=77, key3="nice day, thanks"

Now in my head I wish it was of the form ("input.csv")

#key1, key2, key3
23,    67,   "hello there"
7,     22,   "how are you"
2,     77,   "nice day, thanks"

so I could do

data <- read.csv("input.csv", header=TRUE)

where the header column names are derived from the key names dynamically,
and I could access the data using normal data$key1 or data$key2 mechanism.

I guess I could just pre process the file first  using python etc to create
a CSV file with column header derived from key names, and values derived 
from
key values, but I am interested to see how experienced R folks would 
handle this
inside R.

Thanks,

Frank
PIKAL Petr
# 
Hi

One option are regular expressions but you can also read data with "=" as separator.

test<-read.table("input_kvpairs.csv", sep=c("="), header=F, stringsAsFactors=F)

#use this function to split and extract numeric parts
extract<-function(x) as.numeric(sapply(strsplit(x,","),"[",1))

# and apply the function to appropriate columns
test1 <- data.frame(sapply(test[,2:3], extract), test[,4])

Now you can put names to resulting data frame see
?names

Regards
Petr
arun
# 
HI,

May be this helps:
Lines1<-readLines(textConnection('key1=23, key2=67, key3="hello there"
key1=7, key2=22, key3="how are you"
key1=2, key2=77, key3="nice day, thanks"'))
res<-read.table(text=gsub("key{0,1}\\d","",gsub("[\",]","",Lines1)),sep="=",header=FALSE,stringsAsFactors=F)[-1]


?names(res)<- paste(substr(Lines1,1,3),1:3,sep="")
?res
#? key1 key2??????????? key3
#1?? 23?? 67???? hello there
#2??? 7?? 22???? how are you
#3??? 2?? 77 nice day thanks
A.K.



----- Original Message -----
From: Frank Singleton <b17flyboy at gmail.com>
To: r-help at r-project.org
Cc: 
Sent: Thursday, January 17, 2013 11:21 PM
Subject: [R] reading multiple key=value pairs per line

Hi,

Thanks for a great environmentfor statistical? computing :-)

I have some input data in a file ("input_kvpairs.csv") of the form

key1=23, key2=67, key3="hello there"
key1=7, key2=22, key3="how are you"
key1=2, key2=77, key3="nice day, thanks"

Now in my head I wish it was of the form ("input.csv")

#key1, key2, key3
23,? ? 67,?  "hello there"
7,? ?  22,?  "how are you"
2,? ?  77,?  "nice day, thanks"

so I could do

data <- read.csv("input.csv", header=TRUE)

where the header column names are derived from the key names dynamically,
and I could access the data using normal data$key1 or data$key2 mechanism.

I guess I could just pre process the file first? using python etc to create
a CSV file with column header derived from key names, and values derived from
key values, but I am interested to see how experienced R folks would handle this
inside R.

Thanks,

Frank

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
arun
# 
Hi,

Sorry, there was a mistake.? I didn't notice comma in key3
.Lines1<-readLines(textConnection('key1=23, key2=67, key3="hello there"
key1=7, key2=22, key3="how are you"
key1=2, key2=77, key3="nice day, thanks"'))
res1<-read.table(text=gsub("key{0,1}\\d","",gsub("[\"]","",Lines1)),sep="=",header=FALSE,stringsAsFactors=F)[-1]
?res1[,1:2]<-sapply(res1[,1:2],function(x) as.numeric(gsub("\\, $","",x)))
names(res1)<- paste(substr(Lines1,1,3),1:3,sep="")
res1
#? key1 key2???????????? key3
#1?? 23?? 67????? hello there
#2??? 7?? 22????? how are you
#3??? 2?? 77 nice day, thanks
?str(res1)
#'data.frame':??? 3 obs. of? 3 variables:
# $ key1: num? 23 7 2
# $ key2: num? 67 22 77
# $ key3: chr? "hello there" "how are you" "nice day, thanks"
A.K.



----- Original Message -----
From: Frank Singleton <b17flyboy at gmail.com>
To: r-help at r-project.org
Cc: 
Sent: Thursday, January 17, 2013 11:21 PM
Subject: [R] reading multiple key=value pairs per line

Hi,

Thanks for a great environmentfor statistical? computing :-)

I have some input data in a file ("input_kvpairs.csv") of the form

key1=23, key2=67, key3="hello there"
key1=7, key2=22, key3="how are you"
key1=2, key2=77, key3="nice day, thanks"

Now in my head I wish it was of the form ("input.csv")

#key1, key2, key3
23,? ? 67,?  "hello there"
7,? ?  22,?  "how are you"
2,? ?  77,?  "nice day, thanks"

so I could do

data <- read.csv("input.csv", header=TRUE)

where the header column names are derived from the key names dynamically,
and I could access the data using normal data$key1 or data$key2 mechanism.

I guess I could just pre process the file first? using python etc to create
a CSV file with column header derived from key names, and values derived from
key values, but I am interested to see how experienced R folks would handle this
inside R.

Thanks,

Frank

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Frank Singleton
# 
Thanks Arun,Petr,Greg and Christophe,

I have some possibilities to consider now :-)

Cheers / Frank
On 01/18/2013 08:37 AM, arun wrote: