R package to geolocate IP addresses using http://freegeoip.net/ . The package only contains one functon freegeoip. The freegeoip function only takes data.frames (with a single IP column). And it returns a data.frame with the IP addresses geocoded to the maximum granularily available in the http://www.maxmind.com/ IPs database.
You will need the devtools package to install the freegeoip package.
install.packages('devtools')
library(devtools)
install_github('freegeoip', 'luiscape')Input a data.frame with a single IP column. The result will be a data.frame with the geocoded addresses.
x <- data.frame('34.23.143.243')
freegeoip(x) This is the data.frame in the output:
ip country_code country_name region_code region_name city zipcode latitude longitude metro_code area_code
1 34.23.143.243 US United States TX Texas Houston 77072 29.6997 -95.5858 618 281The results are neatly geocoded:
This package was written when trying to geocode IP addresses from SurveyMonkey. SurveyMonkey provides an export feature that allows you to download your list of subscribers in a CSV format. They collect IP addresses from subscribers by default. I needed a function that geocoded the IP column in the CSV file. That need was the main reason why I created this package.
With that said, the package is really slow. It sends a single IP address at a time. As a result, if you have a large list of IP addresses, you end up making many -- many -- queries. I did it this way in order to properly catch errors -- I was geocoding a very international IP address list with lots of errors. Errors are catched and properly categorized as NAs.
- Allow bulk query of IP addresses using lists.
- Create a simple mapping function using rMaps.