I'm new to Clojure and I have a question regarding nested doseq loops.
I would like to iterate through a sequence and get a subsequence, and then get some keys to apply a function over all the sequence elements.
The given sequence has an structure more or less like this, but with hundreds of books, shelves and many libraries:
([:state/libraries {6 #:library {:name "MUNICIPAL LIBRARY OF X" :id 6
:shelves {3 #:shelf {:name "GREEN SHELF" :id 3 :books
{45 #:book {:id 45 :name "NECRONOMICON" :pages {...},
{89 #:book {:id 89 :name "HOLY BIBLE" :pages {...}}}}}}}}])
Here is my code:
(defn my-function [] (let [conn (d/connect (-> my-system :config :datomic-uri))]
(doseq [library-seq (read-string (slurp "given-sequence.edn"))]
(doseq [shelves-seq (val library-seq)]
(library/create-shelf conn {:id (:shelf/id (val shelves-seq))
:name (:shelf/name (val shelves-seq))})
(doseq [books-seq (:shelf/books (val shelves-seq))]
(library/create-book conn (:shelf/id (val shelves-seq)) {:id (:book/id (val books-seq))
:name (:book/name (val books-seq))})
)))))
The thing is that I want to get rid of that nested doseq mess but I don't know what would be the best approach, since in each iteration keys change. Using recur? reduce? Maybe I am thinking about this completely the wrong way?
library/create-*
functions act entirely through side effects? They don't return anything? - Carcigenicate(doseq [library-seq ..., shelves-seq ...] ...)
. You don't need a newdoseq
for every sequence. They can be grouped like withfor
s. - Carcigenicate