Substring (variable length) values in entire column of dataframe - r

Question

I have looked for this tirelessly with no luck. I am coming from a Java background and new to R. (On a side note, I am loving R, but disliking string operations in it as well as the documentation - maybe that's just a Java bias.)
Anyhow, I have a dataframe with a single column, it is composed of a latitude and longitude numbers seperated by a colon e.g. ROAD:_:-87.4968190989999:38.7414455360001
I would like to create 2 new data frames where each will have the separate lat and long numbers.
I have successfully written a piece of code where I use for loops (but I know this is inefficient - and that there has to be another way)
Here is a snippet of the inefficient code:
length <- length(fromLatLong)
for (i in 1:length){
fromLat[i] <- strsplit(fromLatLong[i] ,":")[[1]][4]
}
for (i in 1:length){
fromLong[i] <- strsplit(fromLatLong[i] ,":")[[1]][3]
}
for (i in 1:length){
toLat[i] <- strsplit(toLatLong[i] ,":")[[1]][4]
}
for (i in 1:length){
toLong[i] <- strsplit(toLatLong[i] ,":")[[1]][3]
}
Here is how I tried to optimize it using mutate, but I only get the first value copied over to all rows as such:
fromLat = mutate(fromLatLong, FROM_NODE_ID = (strsplit(as.character(fromLatLong$FROM_NODE_ID),":")[[1]][4]))
fromLong = mutate(fromLatLong, FROM_NODE_ID = (strsplit(fromLatLong$FROM_NODE_ID,":")[[1]][3]))
toLat = mutate(toLatLong, TO_NODE_ID = (strsplit(toLatLong$TO_NODE_ID,":")[[1]][4]))
toLong = mutate(toLatLong, TO_NODE_ID = (strsplit(toLatLong$TO_NODE_ID,":")[[1]][3]))
And here is the result:
FROM_NODE_ID
1
38.7414455360001
2
38.7414455360001
3
38.7414455360001
4
38.7414455360001
5
38.7414455360001
6
38.7414455360001
7
38.7414455360001
8
38.7414455360001
9
38.7414455360001
I would appriciete your help on this. Thanks

Answer 1

You can use the map_chr function of the purrr package. For instance:
fromLat = mutate(fromLatLong, FROM_NODE_ID = map_chr(FROM_NODE_ID, ~ strsplit(as.character(.x),":")[[1]][4]))

Answer 2

The following expression will produce a data frame with each of the colon-delimited components as a separate column. You can then break this up into separate data frames or do whatever else you want with it.
as.data.frame(t(matrix(unlist(strsplit(fromLatLong$coords, ":", fixed=TRUE), recursive=FALSE), nrow=4)),stringsAsFactors=FALSE)
(Assuming the column name of your values in the data frame is coords.)