How can I write a function that returns odd numbers only from a list of integers, using R language?

Question

How can I write a function that returns odd numbers only from a list of integers, using R language? So far I tried this

 function3 <- function(x){
      x<-as.integer(x)
      if (x %% 2 ==1) {
          return(x)
      }
    }

but it doesn't work for lists or vectors, and I don't really know how to change my code since the "if" will complain with the condition >"has length > 1" and only the first element will be used

Answer 1

justodd <- function(x) x[ x %% 2 == 1 ]
justodd(4:20)
# [1]  5  7  9 11 13 15 17 19

Explanation:

• Indexing is described in ?Extract (also in ?[ ), you'll see that it takes either a list of integer (or integerizable numeric ) or logical . If the former, the numbers need to be within the length of the vector; if the latter, then it should be the same length as the original vector. For example,

   x <- 4:20 x[c(3,5,1)] # [1] 6 8 4 x[c(F,F,T,F,T,F,F,F,T,F,F,F,F,F,F,F,F)] # [1] 6 8 12 

• so our [ -insides looks like

   x %% 2 == 1 # [1] FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE # [13] FALSE TRUE FALSE TRUE FALSE 

• then we index the original vector x based on that return value

---

Update

Your mention of it not working with a list as an argument suggests an alternative that is slower but works on both vectors and lists.

justodd2 <- function(x) Filter(function(a) a %% 2 == 1, x)
vec <- 4:20
lst <- as.list(4:20)
str(vec)
#  int [1:17] 4 5 6 7 8 9 10 11 12 13 ...
str(lst)
# List of 17
#  $ : int 4
#  $ : int 5
#  $ : int 6
#  $ : int 7
#  $ : int 8
#  $ : int 9
#  $ : int 10
#  $ : int 11
#  $ : int 12
#  $ : int 13
#  $ : int 14
#  $ : int 15
#  $ : int 16
#  $ : int 17
#  $ : int 18
#  $ : int 19
#  $ : int 20
justodd2(vec)
# [1]  5  7  9 11 13 15 17 19
justodd2(lst)
# [[1]]
# [1] 5
# [[2]]
# [1] 7
# [[3]]
# [1] 9
# [[4]]
# [1] 11
# [[5]]
# [1] 13
# [[6]]
# [1] 15
# [[7]]
# [1] 17
# [[8]]
# [1] 19

Performance comparison:

microbenchmark::microbenchmark(
  a = justodd(vec),
  b = justodd2(vec),
  c = justodd2(lst)
)
# Unit: nanoseconds
#  expr   min    lq  mean median    uq     max neval
#     a   800  1000 26642   1100  1200 2537700   100
#     b 12100 12500 24154  12700 13150 1055600   100
#     c 12100 12300 23777  12500 12800 1022900   100

(Ignore the high mean and max values from microbenchmark , they are often biased by garbage-collection in the R internals. If you don't know what that is ... just press the [I believe] button or google it.)

So ultimately, if you are always dealing with vectors, then I suggest the first justodd solution, otherwise the justodd2 is safer (as justodd(lst) fails).