Adding to set() if part of tuple not exists-CodePudding

I would like to use something similar to the add() function to add tuples to a set. However, I want them not to be added if a part of that tuple already exists in the set.

For example:

mySet = set()
myList = [['A','B',1],['B','C',2],['C','D',3],['A','B',2]]

for i in myList:
    mySet.add(((i[0],i[1]),i[2]))

This gives me a result as:

mySet = (('A', 'B'), 1), (('A', 'B'), 2), (('B', 'C'), 2), (('C', 'D'), 3)

But I would like the ('A','B')-combination to only appear once (the first time it is added), so that my result is:

mySet = {(('A', 'B'), 1), (('B', 'C'), 2), (('C', 'D'), 3)}

CodePudding user response：

Why not use a dictionary in that case?

d = {}
for i,j,k in myList:
    if (i,j) not in d:
        d[(i,j)] = k
mySet = set(d.items())

As @Matthias suggests, reversed makes it even better:

mySet = set({(i,j): k for i,j,k in reversed(myList)}.items())

Output:

{(('A', 'B'), 1), (('B', 'C'), 2), (('C', 'D'), 3)}

CodePudding user response：

Transform the data to a dictionary first.

my_list = [['A','B',1],['B','C',2],['C','D',3],['A','B',2]]
my_dict = {tuple([*x[:2]]):x[2] for x in reversed(my_list)}
# alternative for the previous line: my_dict = {(x, y): z for x, y, z in reversed(my_list)}
print(my_dict)  # debug output
my_set = {*my_dict.items()}  # or: my_set = set(my_dict.items())
print(my_set)

Output for my_dict is {('A', 'B'): 1, ('C', 'D'): 3, ('B', 'C'): 2}.

The final set is {(('A', 'B'), 1), (('B', 'C'), 2), (('C', 'D'), 3)}.

According to your wanted output you want the first element from the list that matches the criteria. So we have to loop over the list backwards with reversed. Otherwise a later value would overwrite the first matching entry.

CodePudding user response：

before you append it to the set, try using set(list) so that all elements will be unique so there should be no repeating elements