In our last post we dug in to lists (a.k.a first and rest). Lists seemed rather limiting/constrainging in functionality, wouldn’t you agree? The vector fixes all these issues.
Data within brackets constitutes a vector. So [1 2 3] is a vector.
There are other ways to make vectors as well:
;; make a vector
(vector 1 2 3)
;; make a vector with brackets
[1 2 3]
;; make a vector from a list
(vec '(1 2 3))So easy, right?
While rest and first are available here as well, vectors support getting a variable at an index N. Complexity for getting is log32N.
;; rest and first work here as well
(first [1 2 3]) ;; returns 1
(rest [1 2 3]) ;; returns (2 3) -- a list!
;; get the item at index x
;; get complexity is log32n
(get [1 2 3] 0) ;; returns 1
;; this is a neat way of getting a value by index
([1 2 3] 0) ;; returns 1The last way to get a value by index is interesting– you can use a vector as a function.
You can add to the end of a vector, but not the front (without paying an O(n) cost). This would be the exact opposite of lists if not for the fact that you can replace elements at set indexes via assoc.
(cons 0 [1 2 3]) ;; returns (0 1 2 3) -- a list!
(conj [1 2 3] 4) ;; returns [1 2 3 4]
;; adding to the front of a vector is O(n) operation
;; so maybe consider a list for that?
(into [0] [1 2 3]) ;; returns [0 1 2 3]
;; replace an element at index n
(assoc [1 2 3] 1 9) ;; returns [1 9 3]
;; get new vector from index (inclusive) to index (exclusive)
(subvec [1 2 3] 1 3) ;; returns [2 3]
;; get new vector via plucking elements at indexes [indexes]
(replace [1 2 3] [0 1]) ;; returns [1 2]
;; reverse in con
(rseq [1 2 3]) ;; returns (3 2 1) -- a list!I like the distinct? function.
;; are all values distinct?
(distinct? [1 2 3]) ;; returns trueOne thing to note is the contains? function. You would think that this would tell you if a vector contains a specific value. Nope– it tells you if the vector has a value at index X. So just don’t use contains? for things other than maps/hash-based data structures.