Clojure - Recursively Semi-Flatten Nested Map - clojure

In clojure, how can I turn a nested map like this:
(def parent {:id "parent-1"
:value "Hi dude!"
:children [{:id "child-11"
:value "How is life?"
:children [{:id "child-111"
:value "Some value"
:children []}]}
{:id "child-12"
:value "Does it work?"
:children []}]})
Into this:
[
[{:id "parent-1", :value "Hi dude!"}]
[{:id "parent-1", :value "Hi dude!"} {:id "child-11", :value "How is life?"}]
[{:id "parent-1", :value "Hi dude!"} {:id "child-11", :value "How is life?"} {:id "child-111", :value "Some value"}]
[{:id "parent-1", :value "Hi dude!"} {:id "child-12", :value "Does it work?"}]
]
I'm stumbling through very hacky recursive attempts and now my brain is burnt out.
What I've got so far is below. It does get the data right, however it puts the data in some extra undesired nested vectors.
How can this be fixed?
Is there a nice idiomatic way to do this in Clojure?
Thanks.
(defn do-flatten [node parent-tree]
(let [node-res (conj parent-tree (dissoc node :children))
child-res (mapv #(do-flatten % node-res) (:children node))
end-res (if (empty? child-res) [node-res] [node-res child-res])]
end-res))
(do-flatten parent [])
Which produces:
[
[{:id "parent-1", :value "Hi dude!"}]
[[
[{:id "parent-1", :value "Hi dude!"} {:id "child-11", :value "How is life?"}]
[[
[{:id "parent-1", :value "Hi dude!"} {:id "child-11", :value "How is life?"} {:id "child-111", :value "Some value"}]
]]]
[
[{:id "parent-1", :value "Hi dude!"} {:id "child-12", :value "Does it work?"}]
]]
]

I don't know if this is idiomatic, but it seems to work.
(defn do-flatten
([node]
(do-flatten node []))
([node parents]
(let [path (conj parents (dissoc node :children))]
(vec (concat [path] (mapcat #(do-flatten % path)
(:children node)))))))
You can leave off the [] when you call it.

another option is to use zippers:
(require '[clojure.zip :as z])
(defn paths [p]
(loop [curr (z/zipper map? :children nil p)
res []]
(cond (z/end? curr) res
(z/branch? curr) (recur (z/next curr)
(conj res
(mapv #(select-keys % [:id :value])
(conj (z/path curr) (z/node curr)))))
:else (recur (z/next curr) res))))

I'd be inclined to use a bit of local state to simplify the logic:
(defn do-flatten
([node]
(let [acc (atom [])]
(do-flatten node [] acc)
#acc))
([node base acc]
(let [new-base (into base (self node))]
(swap! acc conj new-base)
(doall
(map #(do-flatten % new-base acc) (:children node))))))
Maybe some functional purists would dislike it, and of course you can do the whole thing in a purely functional way. My feeling is that it's a temporary and entirely local piece of state (and hence isn't going to cause the kinds of problems that state is notorious for), so if it makes for greater readability (which I think it does), I'm happy to use it.

Related

Clojure: set value as a key

May be, it is a stupid question, but it may help many of newbies. How do I add a key-value pair to the map?
I mean something like:
(defn init-item [v item]
(let [{:keys [id value]} item]
(-> v
(assoc :{ID_AS_A_KEY} value))))
And I get:
(init-item {} {:id "123456789" :value [:name "King" :surname "Leonid"]})
user=> {:123456789 [:name "King" :surname "Leonid"]}
Just don't do it. Use the string itself as your map key. There's no reason to make it a keyword. It's much easier to work with if you leave it alone.
(defn init-item [v item]
(assoc v (:id item) (:value item)))
I think this is what you meant to do:
(defn init-item
[dest-map item]
(let [item-id-str (:id item)
item-val (:value item)
item-id-kw (keyword item-id-str)]
(assoc dest-map item-id-kw item-val)))
(let [all-items {:a 1 :b 2 :c 3}
item-1 {:id "123456789"
:value [:name "King" :surname "Leonid"]}]
(init-item all-items item-1)
;=> {:a 1, :b 2, :c 3, :123456789 [:name "King" :surname "Leonid"]}
Clojure has functions name, symbol, and keyword to convert between strings and symbols/keywords. Since you already have the ID as a string, you just need to call keyword to convert it.
Be sure to always keep a browser tab open to The Clojure CheatSheet.

Intersecting a list of lists of maps overriding equality

I have a list of lists of maps:
(( {:id 1 :temp 1} {:id 2} )
( {:id 1 :temp 2} )
( {:id 1 :temp 3} {:id 2} ))
I want to get ids which are at intersection of these 3 sets only by :id key. So my result here will be 1
I came up with this solution but it's hurting my eyes:
(def coll '(( {:id 1 :temp 1} {:id 2} )
( {:id 1 :temp 2} )
( {:id 1 :temp 3} {:id 2} )))
(apply clojure.set/intersection
(map set (map (fn [m]
(map #(select-keys % '(:id)) m)) coll)))
returns
#{{:id 1}}
which is Ok, but any other suggestions?
If you are fine with getting #{1} (as you mention initially) instead of #{{:id 1}}, then it can be slightly improved:
(apply set/intersection (map (fn [c] (into #{} (map :id c))) coll))
(require '[clojure.set :refer [intersection]])
The select keys I guess you don't need, since you are only interested in the id. (map :id m) does the job for the inner-most map. By this you are getting rid of a function shorthand. You can use it in the next map:
(map #(map :id %) coll)
;; ((1 2) (1) (1 2))
The third map you introduce is not necessary. it can be merged in the above piece of code:
(map (comp set #(map :id %)) coll)
or:
(map #(set (map :id %)) coll)
both evaluating to: (#{1 2} #{1} #{1 2})
This is still pretty nested. Threading macros don't help here. But you can use a very powerful list comprehension macro called for:
(for [row coll]
(set (map :id row)))
This gives you the advantage of naming list items (rows) but keeping it concise at the same time.
So finally:
(apply intersection (for [row coll]
(set (map :id row))))
;; #{1}

How to group-by a collection that is already grouped by in Clojure?

I have a collection of maps
(def a '({:id 9345 :value 3 :type "orange"}
{:id 2945 :value 2 :type "orange"}
{:id 145 :value 3 :type "orange"}
{:id 2745 :value 6 :type "apple"}
{:id 2345 :value 6 :type "apple"}))
I want to group this first by value, followed by type.
My output should look like:
{
:orange [{
:value 3,
:id [9345, 145]
}, {
:value 2,
:id [2935]
}],
:apple [{
:value 6,
:id [2745, 2345]
}]
}
How would I do this in Clojure? Appreciate your answers.
Thanks!
Edit:
Here is what I had so far:
(defn by-type-key [data]
(group-by #(get % "type") data))
(reduce-kv
(fn [m k v] (assoc m k (reduce-kv
(fn [sm sk sv] (assoc sm sk (into [] (map #(:id %) sv))))
{}
(group-by :value (map #(dissoc % :type) v)))))
{}
(by-type-key a))
Output:
=> {"orange" {3 [9345 145], 2 [2945]}, "apple" {6 [2745 2345], 3 [125]}}
I just couldnt figure out how to proceed next...
Your requirements are a bit inconsistent (or rather irregular) - you use :type values as keywords in the result, but the rest of the keywords are carried through. Maybe that's what you must do to satisfy some external formats - otherwise you need to either use the same approach as with :type through, or add a new keyword to the result, like :group or :rows and keep the original keywords intact. I will assume the former approach for the moment (but see below, I will get to the shape as you want it,) so the final shape of data is like
{:orange
{:3 [9345 145],
:2 [2945]},
:apple
{:6 [2745 2345]}
}
There is more than one way of getting there, here's the gist of one:
(group-by (juxt :type :value) a)
The result:
{["orange" 3] [{:id 9345, :value 3, :type "orange"} {:id 145, :value 3, :type "orange"}],
["orange" 2] [{:id 2945, :value 2, :type "orange"}],
["apple" 6] [{:id 2745, :value 6, :type "apple"} {:id 2345, :value 6, :type "apple"}]}
Now all rows in your collection are grouped by the keys you need. From this, you can go and get the shape you want, say to get to the shape above you can do
(reduce
(fn [m [k v]]
(let [ks (map (comp keyword str) k)]
(assoc-in m ks
(map :id v))))
{}
(group-by (juxt :type :value) a))
The basic idea is to get the rows grouped by the key sequence (and that's what group-by and juxt do,) and then combine reduce and assoc-in or update-in to beat the result into place.
To get exactly the shape you described:
(reduce
(fn [m [k v]]
(let [type (keyword (first k))
value (second k)
ids (map :id v)]
(update-in m [type]
#(conj % {:value value :id ids}))))
{}
(group-by (juxt :type :value) a))
It's a bit noisy, and it might be harder to see the forest for the trees - that's why I simplified the shape, to highlight the main idea. The more regular your shapes are, the shorter and more regular your functions become - so if you have control over it, try to make it simpler for you.
I would do the transform in two stages (using reduce):
the first to collect the values
the second for formating
The following code solves your problem:
(def a '({:id 9345 :value 3 :type "orange"}
{:id 2945 :value 2 :type "orange"}
{:id 145 :value 3 :type "orange"}
{:id 2745 :value 6 :type "apple"}
{:id 2345 :value 6 :type "apple"}))
(defn standardise [m]
(->> m
;; first stage
(reduce (fn [out {:keys [type value id]}]
(update-in out [type value] (fnil #(conj % id) [])))
{})
;; second stage
(reduce-kv (fn [out k v]
(assoc out (keyword k)
(reduce-kv (fn [out value id]
(conj out {:value value
:id id}))
[]
v)))
{})))
(standardise a)
;; => {:orange [{:value 3, :id [9345 145]}
;; {:value 2, :id [2945]}],
;; :apple [{:value 6, :id [2745 2345]}]}
the output of the first stage is:
(reduce (fn [out {:keys [type value id]}]
(update-in out [type value] (fnil #(conj % id) [])))
{}
a)
;;=> {"orange" {3 [9345 145], 2 [2945]}, "apple" {6 [2745 2345]}}
You may wish to use the built-in function group-by. See http://clojuredocs.org/clojure.core/group-by

Filtering a map based on expected keys

In my Clojure webapp I have various model namespaces with functions that take a map as an agrument and somehow insert that map into a database. I would like to be able take out only the desired keys from the map before I do the insert.
A basic example of this is:
(let [msg-keys [:title :body]
msg {:title "Hello" :body "This is an example" :somekey "asdf" :someotherkey "asdf"}]
(select-keys msg msg-keys))
;; => {:title "Hello" :body "This is an example"}
select-keys is not an option when the map is somewhat complex and I would like to select a specific set of nested keys:
(let [person {:name {:first "john" :john "smith"} :age 40 :weight 155 :something {:a "a" :b "b" :c "c" :d "d"}}]
(some-select-key-fn person [:name [:first] :something [:a :b]]))
;; => {:name {:first "john"} :something {:a "a" :b "b"}}
Is there a way to do this with the core functions? Is there a way do this purely with destructuring?
This topic was discussed in the Clojure Google Group along with a few solutions.
Destructuring is probably the closest to a "core" capability, and may be a fine solution if your problem is rather static and the map has all of the expected keys (thus avoiding nil). It could look like:
(let [person {:name {:first "john" :john "smith"} :age 40 :weight 155 :something {:a "a" :b "b" :c "c" :d "d"}}
{{:keys [first]} :name {:keys [a b]} :something} person]
{:name {:first first} :something {:a a :b b}})
;; => {:name {:first "john"}, :something {:a "a", :b "b"}}
Below is a survey of the solutions in the Clojure Google Group thread, applied to your sample map. They each have a different take on how to specify the nested keys to be selected.
Here is Christophe Grand's solution:
(defprotocol Selector
(-select [s m]))
(defn select [m selectors-coll]
(reduce conj {} (map #(-select % m) selectors-coll)))
(extend-protocol Selector
clojure.lang.Keyword
(-select [k m]
(find m k))
clojure.lang.APersistentMap
(-select [sm m]
(into {}
(for [[k s] sm]
[k (select (get m k) s)]))))
Using it requires a slightly modified syntax:
(let [person {:name {:first "john" :john "smith"} :age 40 :weight 155 :something {:a "a" :b "b" :c "c" :d "d"}}]
(select person [{:name [:first] :something [:a :b]}]))
;; => {:something {:b "b", :a "a"}, :name {:first "john"}}
Here is Moritz Ulrich's solution (he cautions that it doesn't work on maps with seqs as keys):
(defn select-in [m keyseq]
(loop [acc {} [k & ks] (seq keyseq)]
(if k
(recur
(if (sequential? k)
(let [[k ks] k]
(assoc acc k
(select-in (get m k) ks)))
(assoc acc k (get m k)))
ks)
acc)))
Using it requires another slightly modified syntax:
(let [person {:name {:first "john" :john "smith"} :age 40 :weight 155 :something {:a "a" :b "b" :c "c" :d "d"}}]
(select-in person [[:name [:first]] [:something [:a :b]]]))
;; => {:something {:b "b", :a "a"}, :name {:first "john"}}
Here is Jay Fields's solution:
(defn select-nested-keys [m top-level-keys & {:as pairs}]
(reduce #(update-in %1 (first %2) select-keys (last %2)) (select-keys m top-level-keys) pairs))
It uses a different syntax:
(let [person {:name {:first "john" :john "smith"} :age 40 :weight 155 :something {:a "a" :b "b" :c "c" :d "d"}}]
(select-nested-keys person [:name :something] [:name] [:first] [:something] [:a :b]))
;; => {:something {:b "b", :a "a"}, :name {:first "john"}}
Here is Baishampayan Ghose's solution:
(defprotocol ^:private IExpandable
(^:private expand [this]))
(extend-protocol IExpandable
clojure.lang.Keyword
(expand [k] {k ::all})
clojure.lang.IPersistentVector
(expand [v] (if (empty? v)
{}
(apply merge (map expand v))))
clojure.lang.IPersistentMap
(expand [m]
(assert (= (count (keys m)) 1) "Number of keys in a selector map can't be more than 1.")
(let [[k v] (-> m first ((juxt key val)))]
{k (expand v)}))
nil
(expand [_] {}))
(defn ^:private extract* [m selectors expand?]
(let [sels (if expand? (expand selectors) selectors)]
(reduce-kv (fn [res k v]
(if (= v ::all)
(assoc res k (m k))
(assoc res k (extract* (m k) v false))))
{} sels)))
(defn extract
"Like select-keys, but can select nested keys.
Examples -
(extract [{:b {:c [:d]}} :g] {:a 1 :b {:c {:d 1 :e 2}} :g 42 :xxx 11})
;=> {:g 42, :b {:c {:d 1}}}
(extract [:g] {:a 1 :b {:c {:d 1 :e 2}} :g 42 :xxx 11})
;=> {:g 42}
(extract [{:b [:c]} :xxx] {:a 1 :b {:c {:d 1 :e 2}} :g 42 :xxx 11})
;=> {:b {:c {:d 1, :e 2}}, :xxx 11}
Also see - exclude"
[selectors m]
(extract* m selectors true))
It uses another syntax (and the parameters are reversed):
(let [person {:name {:first "john" :john "smith"} :age 40 :weight 155 :something {:a "a" :b "b" :c "c" :d "d"}}]
(extract [{:name [:first]} {:something [:a :b]}] person))
;; => {:name {:first "john"}, :something {:a "a", :b "b"}}
Your best bet is probably to use select keys on each nested portion of the structure.
(-> person
(select-keys [:name :something])
(update-in [:name] select-keys [:first])
(update-in [:something] select-keys [:a :b]))
You could of course use a generalized version of the above to implement the syntax you suggest in a function (with a reduce rather than a -> form, most likely, and recursive calls for each nested selection of keys). Destructuring would not help much, it makes binding nested data convenient, but isn't really that useful for constructing values.
Here is how I would do it with reduce and recursion:
(defn simplify
[m skel]
(if-let [kvs (not-empty (partition 2 skel))]
(reduce (fn [m [k nested]]
(if nested
(update-in m [k] simplify nested)
m))
(select-keys m (map first kvs))
kvs)
m))
note that your proposed argument format is inconvenient, so I changed it slightly
user=> (simplify {:name {:first "john" :john "smith"}
:age 40
:weight 155
:something {:a "a" :b "b" :c "c" :d "d"}}
[:name [:first nil] :something [:a nil :b nil]])
{:something {:b "b", :a "a"}, :name {:first "john"}}
the syntax you propose will require a more complex implementation

clojure way to update a map inside a vector

What is the clojure way to update a map inside a vector e.g. if I have something like this, assuming each map has unique :name
(def some-vec
[{:name "foo"
....}
{:name "bar"
....}
{:name "baz"
....}])
I want to update the map in someway if it has :name equal to foo. Currently I'm using map, like this
(map (fn [{:keys [name] :as value}]
(if-not (= name "foo")
value
(do-something .....))) some-vec)
But this will loop through the entire vector even though I only update one item.
Keep the data as a map instead of a vector of map-records, keyed by :name.
(def some-data
{"foo" {:name "foo" :other :stuff}
"bar" {:name "bar" :other :stuff}
"baz" {:name "baz" :other :stuff}})
Then
(assoc-in some-data ["bar" :other] :things)
produces
{"foo" {:other :stuff, :name "foo"},
"bar" {:other :things, :name "bar"},
"baz" {:other :stuff, :name "baz"}}
in one go.
You can capture the basic manipulation in
(defn assoc-by-fn [data keyfn datum]
(assoc data (keyfn datum) datum))
When, for example,
(assoc-by-fn some-data :name {:name "zip" :other :fassner})
produces
{"zip" {:other :fassner, :name "zip"},
"foo" {:other :stuff, :name "foo"},
"bar" {:other :stuff, :name "bar"},
"baz" {:other :stuff, :name "baz"}}
Given that you have a vector of maps, your code looks fine to me. Your concern about "looping through the entire vector" is a natural consequence of the fact that you're doing a linear search for the :name and the fact that vectors are immutable.
I wonder whether what you really want is a vector of maps? Why not a map of maps?
(def some-map
{"foo" {...}
"bar" (...}
"baz" {...}}
Which you could then update with update-in?
Given this shape of the input data and unless you have an index that can tell you which indices the maps with a given value of :name reside at, you will have to loop over the entire vector. You can, however, minimize the amount of work involved in producing the updated vector by only "updating" the matching maps, rather than rebuilding the entire vector:
(defn update-values-if
"Assumes xs is a vector. Will update the values for which
pred returns true."
[xs pred f]
(let [lim (count xs)]
(loop [xs xs i 0]
(if (< i lim)
(let [x (nth xs i)]
(recur (if (pred x)
(assoc xs i (f x))
xs)
(inc i)))
xs))))
This will perform as many assoc operations as there are values in xs for which pred returns a truthy value.
Example:
(def some-vec [{:name "foo" :x 0} {:name "bar" :x 0} {:name "baz" :x 0}])
(update-values-if some-vec #(= "foo" (:name %)) #(update-in % [:x] inc))
;= [{:name "foo", :x 1} {:name "bar", :x 0} {:name "baz", :x 0}]
Of course if you're planning to transform the vector in this way with some regularity, then Thumbnail's and Paul's suggestion to use a map of maps will be a much more significant improvement. That remains the case if :name doesn't uniquely identify the maps – in that case, you could simply transform your original vector using frequencies and deal with a map of vectors (of maps with a given :name).
If you're working with vector, you should know index of element that you want to change, otherwise you have to traverse it in some way.
I can propose this solution:
(defn my-update [coll val fnc & args]
(let [index (->> (map-indexed vector coll)
(filter (fn [[_ {x :name}]] (= x val)))
ffirst)]
(when index
(apply update-in coll [index] fnc args))))
Where:
coll - given collection of maps;
val - value of field :name;
fnc - updating function;
args - arguments of the updating function.
Let's try it:
user> (def some-vec
[{:name "foo"}
{:name "bar"}
{:name "baz"}])
;; => #'user/some-vec
user> (my-update some-vec "foo" assoc :boo 12)
;; => [{:name "foo", :boo 12} {:name "bar"} {:name "baz"}]
user> (my-update some-vec "bar" assoc :wow "wow!")
;; => [{:name "foo"} {:name "bar", :wow "wow!"} {:name "baz"}]
I think that Thumbnail's answer may be quite useful for you. If you can keep your data as a map, these manipulations become much easier. Here is how you can transform your vector into a map:
user> (apply hash-map (interleave (map :name some-vec) some-vec))
;; => {"foo" {:name "foo"}, "bar" {:name "bar"}, "baz" {:name "baz"}}