如何只为一个 Key 更新 Set in Dictionary？ （Python）

Question

I am having a difficult time seeing my mistake. I'm trying to just update the set of values for one specific key in a dictionary at a time, except every key is getting updated.

final_df = dict.fromkeys({'a','b','c'}, {'1'})
final_df['a'].update({'1','2'})
final_df

Actual Output: {'b': {'1', '2'}, 'a': {'1', '2'}, 'c': {'1', '2'}}

Desired Output: {'a': {'1', '2'}, 'b': {'1'}, , 'c': {'1'}}

Also tried:

final_df = dict.fromkeys({'a','b','c'}, {'1'})
final_df['a'] = final_df['a'].update({'1','2'})
final_df

Answer 1

This is a simple mistake. What you produce here:

final_df = dict.fromkeys({'a','b','c'}, {'1'})

is a dictionary that with the keys a , b , c all maps to the same set object {'1'} . Then, all subsequent mutation on the set like you did:

final_df['a'].update({'1','2'})

will mutate all mappings as they are all point to the same set object.

To solve this problem, you can consider dictionary comprehension, which will replicate the set object for each key:

>>> final_df = {k:{'1'} for k in {'a','b','c'}}
>>> final_df
{'c': {'1'}, 'a': {'1'}, 'b': {'1'}}
>>> final_df['a'].update({'2','3'})
>>> final_df
{'c': {'1'}, 'a': {'1', '3', '2'}, 'b': {'1'}}

Answer 2

Instead of .update just assign:

final_df = dict.fromkeys({'a','b','c'}, {'1'})
final_df['a'] = {'1','2'}
print(final_df)

Answer 3

It looks like I needed union instead of update.

final_df = dict.fromkeys({'a','b','c'}, {'1'})
final_df['a'] = final_df['a'].union({'2'})
final_df

Output:

{'b': {'1'}, 'a': {'1', '2'}, 'c': {'1'}}