I have a column with a string value:
"geo:lat:55.892994,lng:10.510947,address:content:NA"
And I want to have twonew columnn with:
column1: 55.892994
column2: 10.510947
Would you help me with R code?
I was trying to use str_extract and str_split, but I cannot seem to achieve the result
string <- "geo:lat:55.892994,lng:10.510947,address:content:NA"
library(tidyverse)
data.frame(string=string) %>%
extract(string, c("lat", "lng"), "lat:([0-9.]+),lng:([0-9.]+)",convert = TRUE)
#> lat lng
#> 1 55.89299 10.51095
Created on 2023-02-12 with reprex v2.0.2
Please check the below code
library(tidyverse)
library(stringr)
string <- "geo:lat:55.892994,lng:10.510947,address:content:NA"
data.frame(string=string) %>%
transmute(column1=str_remove_all(str_extract_all(string, '.*\\d+(?=\\d\\,lng)'),'[a-zA-Z:]'),
column2=str_remove_all(str_extract_all(string, '(?<=\\,).*\\d+(?=\\d\\,address)'),'[a-zA-Z:]')
)
Created on 2023-02-12 with reprex v2.0.2
column1 column2
1 55.89299 10.51094
You can use sub from base R.
data <- data.frame(xy="geo:lat:55.892994,lng:10.510947,address:content:NA")
data$lat <- sub('.*lat:(\\d+\\.\\d+).*', '\\1', data$xy)
data$lng <- sub('.*lng:(\\d+\\.\\d+).*', '\\1', data$xy)
See help(regex) for details.
Try the following regex.
library(stringr)
s="geo:lat:55.892994,lng:10.510947,address:content:NA"
str_match(s,"lat:([0-9]{2}.[0-9]{6}).*?lng:([0-9]{2}.[0-9]{6})")
Outputs
[,1] [,2] [,3]
[1,] "lat:55.892994,lng:10.510947" "55.892994" "10.510947"
You can use gregexpr with [0-9.]+ to find the numbers and regmatches to extract them.
cbind(x, do.call(rbind, regmatches(x$s, gregexpr("[0-9.]+", x$s))))
# s 1 2
#1 geo:lat:55.892994,lng:10.510947,address:content:NA 55.892994 10.510947
#2 geo:lat:57.892994,lng:14.510947,address:content:NA 57.892994 14.510947
Data
x <- data.frame(s = c("geo:lat:55.892994,lng:10.510947,address:content:NA",
"geo:lat:57.892994,lng:14.510947,address:content:NA"))
