The map data structure in clojure is entirely similar to ones in other languages– key/value based data storage/access. In most other languages that i use, map access/setting is O(1), which is part of their advantage. In clojure access is log32n but at least we’re getting the benefits of immutable, persistent storage.
Pretty easy to make a map:
;; make a map
(hash-map :key "value" )
;; make a map with curly brackets
{:key "value"}
;; nested map
{:key {:sub-key "value"}}Clojure has a few types of maps. We’ll only be investigating hash maps in this article, though.
Its super easy to get data out of a map.
;; get a key's value
(get {:key "val"} :key) ;; returns "val"
;; get a non-existing key/value
(get {:key "val"} :missing-key) ;; returns nil
;; another means to get a key's value
({:key "val"} :key) ;; returns "val"
;; get the keys of the map
(keys {:key1 "val1" :key2 "val2"}) ;; returns (:key1 :key2)
;; get the values of the map
(vals {:key1 "val1" :key2 "val2"}) ;; returns ("val1" "val2")Note that like vectors, you can use maps as the functions to access their values.
Again, maps are immutable and persistent– so functions that modify them return new objects with the old object left unchanged.
assoc is (from what I observe) the most used method to modify a map.
;; use assoc to "upsert" values in a map
(assoc {:key "val1"} :key "val2") ;; returns {:key "val2"}
;; use dissoc to remove entries in a map
(dissoc {:key1 "val1" :key2 "val2"} :key2) ;; returns {:key1 "val1"}
;; use assoc-in to update a map/sub-section of a map
;; this returns {:key1 {:sub-key1 "val2"}}
(assoc-in {:key1 {:sub-key1 "sval1"}} [:key1 :sub-key1] "val2")
;; use update-in to update a map/sub-section of a map via a function application
;; this returns {:key1 {:sub-key1 "SVAL1"}}
(update-in {:key1 {:sub-key1 "sval1"}} [:key1 :sub-key1] clojure.string/upper-case)Interesting how assoc-in takes a map and a vector and updates “within” a structure. Seems really useful for nested scenarios.