R package for zip code & other geodata from IP addresses?

Question

I have a list of about 200,000 IP addresses. I would like to link these to geographic location and get any other data that an IP address can give as well.

The best I've found so far is a service provided by infochimps: http://www.infochimps.com/datasets/digital-element-ip-intelligence-demographics There's also an R package for infochimps. But infochimps requires you to pay and for 200,000 IP addresses this could get expensive.

Is there any R package that can do something like this?

Thank you

Answer 1

Try using the RDSTK package, which provides an R interface to the Data Science Toolkit API. Here is a presentation by the author of the package, that should help you get started.

From Xu Wang's comments (moved here to increase future findability): For reference purposes: To install that package, one must install RCurl and rjson. Before installing RCurl, on Ubuntu I had to install two packages: sudo apt-get install curl libcurl4-gnutls-dev The function that I needed was ip2coordinates, which accepts an IP address as input

Answer 2

The function IPtoXY (http://thebiobucket.blogspot.com/2011/12/function-to-collect-geographic.html) uses the same API but does not need additional packages..

Edit, 26th of Sept: Thanks to @Peter M I became aware that my function mentioned above was not working anymore - here is the edited version which should work (also the link above was updated..):

# Purpose: Get geographic coordinates for a given IP-address
# Author: Kay Cichini
# Date: 2011-12-18
# Output: A string holding longitude and latitude with format "X;Y"

IPtoXY <- function(x) {
   URL_IP <- paste("http://www.datasciencetoolkit.org//ip2coordinates/",
                   x, sep = "")

   api_return <- readLines(URL_IP, warn = F)
   lon1 <- api_return[grep("longitude", api_return)]
   lon <- gsub("[^[:digit:].]", "", lon1)
   lat1 <- api_return[grep("latitude", api_return)]
   lat <- gsub("[^[:digit:].]", "", lat1)
   return(paste(lat, lon, sep = ";"))
}

# Example:
> IPtoXY("74.88.200.52")
[1] "40.951301574707;73.78759765625"

Answer 3

The function from: http://thebiobucket.blogspot.com/2011/12/function-to-collect-geographic.html , does not work.

But the idea still does, so this should do:

iplocation <- function(ip=""){
  response    <- readLines(paste("http://www.datasciencetoolkit.org//ip2coordinates/",ip,sep=""))
  success     <- !any(grepl("null",response))

  ip <- grep("[[:digit:]]*\\.[[:digit:]]*\\.[[:digit:]]*\\.[[:digit:]]*",response,value=T)
    match <- regexpr("[[:digit:]]*\\.[[:digit:]]*\\.[[:digit:]]*\\.[[:digit:]]*",ip)
    ip <- substr(ip,match,as.integer(attributes(match)[1])+match-1)
  if(success==T){
    extract <- function(label,response){
            text <- grep(label,response,value=T)
            match <- regexpr(paste('"',label,'"',": ",sep=""),text)
            text <- substr(text,match+as.integer(attributes(match)[1]),nchar(text))
            if(grepl("[[:digit:]]",text)){
                    text <- substr(text,1,nchar(text)-2)
            }else{
                    text <- substr(text,2,nchar(text)-2)
                }
            if( regexpr('"',text)!= -1){
                text<-substr(text,2,nchar(text))
            }
            print(text)
            text
        }
  }
  RESULT <- list()
  RESULT$success     <- success
  RESULT$ip          <- ip
  if(success==T){
    RESULT$latitude    <- as.numeric(extract("latitude",response))
    RESULT$longitude   <- as.numeric(extract("longitude",response))
    RESULT$country     <- extract("country_name",response)
    RESULT$locality    <- extract("locality",response)
    RESULT$postalcode  <- extract("postal_code",response)
    RESULT$region      <- extract("region",response)
    RESULT$countrycode <- extract("country_code3",response)
  }
  RESULT
}

iplocation()