Question

How can I group nested collections based on column values, which are dynamically given? For example, suppose we have the following nested collections; how can I group it by the values in first and second columns?

[ ["A" 2011 "Dan"]
["A" 2011 "Jon"]
["A" 2010 "Tim"]
["B" 2009 "Tom"] ]

The desired resulting map is:

{ A { 
      2011 [['A', 2011, 'Dan'] ['A', 2011, 'Joe']]
      2010 [['A', 2010, 'Tim']] 
    }
  B { 2009 [['B', 2009, 'Tom']] } 
}

Following is my solution, which almost works:

(defn nest [data criteria]
  (if (empty? criteria)
    data
    (for [[k v] (group-by #(nth % (-> criteria vals first)) data)]
      (hash-map k (nest v (rest criteria))))))

Answer 1

Here’s the solution I came up with. It works, but I’m sure it can be improved.

(defn nest [data criteria]
  (if (empty? criteria)
    data
    (into {} (for [[k v] (group-by #(nth % (-> criteria vals first)) data)]
      (hash-map k (nest v (rest criteria)))))))