Clojure: Create or update Map values in a vector - clojure

Question

I want to update some values in my vector.
To change this :
[{:key 0, :values {:value1 1, :value2 100}} {:key 1, :values {:value1 5, :value2 300}}]
In this:
[{:key 0, :values {:value1 1, :value2 100}} {:key 1, :values {:value1 6, :value2 500}}]
I'm trying to do something like that, without success:
(if (contains? myvectors myId)
(do
(assoc ((myvectors myId) :values) :value2 500));not working
)
(do
(def myvectors (merge myvectors {:key myId :values {:value1 1 :value2 300}}))
)
)

Answer 1

This would work:
(def v
[{:key 0 :values {:value1 1 :value2 100}} {:key 1 :values {:value1 5, :value2 300}}])
(defn update-values [v k value1 value2]
(mapv (fn [{:keys [key] :as m}]
(cond-> m
(= key k) (assoc :values {:value1 value1 :value2 value2})))
v))
(update-values v 1 6 500)
;; => [{:key 0 :values {:value1 1 :value2 100}}
;; {:key 1 :values {:value1 6 :value2 500}}]
If possible I would change your data structure so you can make better use of core functions, for example a map from key to values:
(def m
{0 {:value1 1 :value2 100}
1 {:value1 5 :value2 300}})
(assoc m 1 {:value1 6 :value2 500})
;; => {0 {:value1 1 :value2 100}
;; 1 {:value1 6 :value2 500}}

Answer 2

If all you want to do is change the second element (map) in your vector as per your before/after examples, you can do something like this:
(let [m [{:key 0, :values {:value1 1, :value2 100}}
{:key 1, :values {:value1 5, :value2 300}}]]
(update-in m [1 :values] merge {:value1 6 :value2 500}))
=> [{:key 0, :values {:value1 1, :value2 100}} {:key 1, :values {:value1 6, :value2 500}}]
where update-in updates the second element in the vector (index 1 in the [1 ...] expression - 0 based indexing) and on that element updates the :values key. For that value ({:value1 5, :value2 300}) we call the merge function. update-in will send in the old value (again, the old value is at this point {:value1 5, :value2 300}) as the first argument to merge and then we send in {:value1 6 :value2 500} as the second argument. Since later arguments to merge "win", the map values we send in will overwrite the ones already in your data.

Answer 3

It's not clear what's your goal, but I've tried to come up with something general enough:
(def data [{:key 0, :values {:value1 1, :value2 100}}
{:key 1, :values {:value1 5, :value2 300}}])
(reduce
(fn [v {:keys [key] :as row}]
(conj v
(cond-> row
(= key 1)
(assoc :values {:value1 6, :value2 500}))))
[]
data)
=> [{:key 0, :values {:value1 1, :value2 100}} {:key 1, :values {:value1 6, :value2 500}}]
This will reduce over the input sequence by changing the given row only based on some condition. Since it looks like you want to change based on the key I've added the condition (= key 1) but this can be easily adapted to any condition for anything that's contained in the row.