从字典创建列表列表

Question

I have a dictionary as follows:

d = {'a':['b','c'],'d':['c'],'e':['f'],'i':['g'],'g':['e'],'b':[],'c':[],'h':[],'f':[],}

where the keys represent each node and the values of each key represent their children.

I need an output as follows:

[['a', 'd', 'i', 'h'], ['b', 'c', 'g'], ['e'], ['f']]

• A list of lists, where the list1 has in-degree 0(level-0 nodes) and list2 has level-1 nodes, list3 has level-2 nodes and so on.. (Order of elements in the internal lists does not matter).

• There won't be any changes in the input dictionary. But a parent can be declared later as another node's child as well, which has to be taken into consideration.(For example, here e is declared first as parent of f, but it later is declared as the child of g).

• There won't be any duplicate entries of parents or keys. Every node in the tree is declared only once as a single key. But there will be same values with different keys, depending on which parent they are linked to.

• No cases of cycles in the dict.

Could any of you please help me with the Python code for the same?

(I tried something similar to BFS algorithm but it's not giving me the desired output)

Answer 1

you can do this with two functions.
this algorithm is not optimized, but it's possible for you to improve it:)

def append_value_at(l, value, index):
    # placing value in the index-th list of l. (lists are added if necessary)
    if len(l) <= index:
        l.append([ value ])
    else:
        if value not in l[index]:
            l[index].append(value)

    if len(l) == index:
        l.append([ ])

def convert_to_lists(d):
    returning = [ ]
    for key, values in d.items(): # iterate through the dictionary
        # finding index of the key
        node = 0
        for i in range(len(returning)):
            l = returning[i]
            for k in l:
                if k == key:
                    node = i

        append_value_at(returning, key, node)
        for val in values:
            append_value_at(returning, val, node + 1)

    return returning

d = {'a':['d'],'b':['d','e'],'c':['e','f'],'d':[],'e':['f'],'f':[]}
convert_to_lists(d)
>>> [['a', 'b', 'c'], ['d', 'e', 'f'], ['f']]

**note: ** your dictionary has a problem, f is a lvl 1 and 2 node... but it works anyway:)
hope it helped !
kyros