transpose a map collecting keys along the way

Question

I'm trying to transpose a map so that:

[x: y, w: y, a: b]

becomes

[y: [x, w], b: a]

(all variables are strings) Doing something like

["x": "y", "w": "y", "a": "b"].collectEntries { [it.value, it.key] }

gets me part way, but stomps on the first new value for "y". I only get: [y:w, b:a]

What is the best way to slurp up the new values into an array for their common new key? Thanks for any help or suggestions.

Answer 1

I hope this helps :

def map = ["x": "y", "w": "y", "a": "b"]

map.groupBy{ it.value }.collectEntries{ k, v -> [k, v.keySet()] }

Answer 2

I had a similar requirement, but needed individual keys (kind of a pivot), so my solution is a little different:

def map = ["x": "y", "w": "y", "a": "b"]
def newKeys = map.collect{ it.value }.flatten() as Set
def transpose = newKeys.collectEntries{ [
    (it): map.findAll{ k, v -> v.contains(it) }.collect{ it.key } 
]}
println transpose
// [y:[x, w], b:[a]]