CODEWITHSUNDEEP
rsorting
bill.meiner
bill.meiner

Reputation: 335

Detect alphabetical order in character vector

I have several character vectors like this:

lastname <- c("Smith"  ,"Johnson" , "Williams" , "Moore" , "Taylor", "Jones"
 , "Brown" , "Davis" ,  "Miller" , "Wilson" )

lastname's last 4 elements are in alphabetical order. I want to split the vector where the alphabetical order starts. Thus, the result would look like:

lastname1 <- c("Smith"  ,"Johnson" , "Williams" , "Moore" , "Taylor", "Jones")
lastname2 <- c("Brown" , "Davis" ,  "Miller" , "Wilson" )

The part in alphabetical order is always located at the end but it's length may differ.

Any help greatly appreciated!

Upvotes: 2

Views: 199

Answers (2)

Sandipan Dey
Sandipan Dey

Reputation: 23101

This will also work:

pos <- tail(which(sapply(1:(length(lastname)-1), function(i) lastname[i] < lastname[i+1])==FALSE),1)
lastname[1:pos]
# [1] "Smith"    "Johnson"  "Williams" "Moore"    "Taylor"   "Jones"   
lastname[(pos+1):length(lastname)]
# [1] "Brown"  "Davis"  "Miller" "Wilson"

Upvotes: 0

akrun
akrun

Reputation: 887118

We can create a logical index and then split the vector by the index to create a list of vectors.

i1 <- rev(cumsum(c(TRUE, diff(rank(rev(lastname))) >0))==1)
split(lastname, i1)
#$`FALSE`
#[1] "Smith"    "Johnson"  "Williams" "Moore"    "Taylor"   "Jones"   

#$`TRUE`
#[1] "Brown"  "Davis"  "Miller" "Wilson"

Upvotes: 2

Related Questions