Exercism: "Nucleotide count in Clojure"

This is my solution to the Nucleotide count problem in Clojure:

	(ns dna)
	(def nucleotides #{\A, \T, \C, \G, \U})
	(defn nucleotide-counts [strand]
	(let
	[count
	(fn [counted-nucleotides nucleotide]
	(assoc
	counted-nucleotides
	nucleotide
	(+ (get counted-nucleotides nucleotide) 1)))]
	(reduce count {\A 0, \T 0, \C 0, \G 0} strand)))
	(defn count [nucleotide strand]
	(if (contains? nucleotides nucleotide)
	(get (nucleotide-counts strand) nucleotide 0)
	(throw (Exception. "invalid nucleotide"))))

view raw nucleotide-count1.clj hosted with ❤ by GitHub

The code is very similar to the one for the Word count exercise. The main difference is that here I used a set to validate that the given nucleotide is valid.

I eliminated the duplication between the set and the map contents in this second version:

	(ns dna)
	(def dna-nucleotides #{\A, \T, \C, \G})
	(def nucleotides (conj dna-nucleotides \U))
	(defn nucleotide-counts [strand]
	(let
	[counted-nucleotides
	(zipmap dna-nucleotides (repeat (count dna-nucleotides) 0))
	count
	(fn [counted-nucleotides nucleotide]
	(assoc
	counted-nucleotides
	nucleotide
	(+ (get counted-nucleotides nucleotide) 1)))]
	(reduce count counted-nucleotides strand)))
	(defn count [nucleotide strand]
	(if (contains? nucleotides nucleotide)
	(get (nucleotide-counts strand) nucleotide 0)
	(throw (Exception. "invalid nucleotide"))))

view raw nucleotide-count2.clj hosted with ❤ by GitHub

where I defined a set containing only DNA nucleotides, dna-nucleotides, that I used to define the nucletides set using conj. This dna-nucleotides set served to generate the counted-nucleotides map using the zipmap function.

Trying to avoid duplication I discovered several new things about Clojure.

You can nitpick my solution here or see all the exercises I've done so far in this repository.

--------------------------------

Update:

After learning some new stuff, I've been able to simplify the code a bit more:

	(ns dna)
	(def ^:private dna-nucleotides #{\A, \T, \C, \G})
	(def ^:private nucleotides (conj dna-nucleotides \U))
	(defn nucleotide-counts [strand]
	(merge {\A 0, \T 0, \C 0, \G 0} (frequencies strand)))
	(defn count [nucleotide strand]
	(if (contains? nucleotides nucleotide)
	(get (nucleotide-counts strand) nucleotide 0)
	(throw (Exception. "invalid nucleotide"))))

view raw nucleotide-count3.clj hosted with ❤ by GitHub

It turned out that the frequencies function already did the counting out of the box, I jut needed to merge it with the result for an empty strand to make frequencies output conform with what the tests were expecting.

I also made the dna-nucleotides and nucleotides sets private.

You can nitpick this new version here.