pythonic way of avoiding duplicated dictionary keys

Question

I'm trying to make a dictionary of pairs of values, but I don't want it added if the pair is already there (even reversed). So let's say (0,1) is already there, I don't want (1,0). This is my code but I'm looking for a more pythonic way of doing it, I'm sure there is one. PS. I can't use itertools for this exercise

    test = {(0,1):1, (0,2):2}
    for a in range(3):
        for b in range(3):
            if a!= b:
                if (b,a) not in test:
                    test[a,b]=4

Does anything here https://stackoverflow.com/questions/8749158/removing-duplicates-from-dictionary help you? — Jason R Stevens CFA, Apr 02 '20 at 21:12
And if your values stored _are hashable_ I'd consider a simple `set` conversion on an entire list/array/column of your keys. This strategy is probably overkill. — Jason R Stevens CFA, Apr 02 '20 at 21:17
Lastly, and probably more overkill, but checkout `Numpy`'s `numpy.unique` function: https://github.com/numpy/numpy/blob/v1.17.0/numpy/lib/arraysetops.py#L151-L294 — Jason R Stevens CFA, Apr 02 '20 at 21:18
for a in range(3): for b in range(a,3): if (a,b) not in test: test[a,b]=4 — splash58, Apr 02 '20 at 21:19
@splash58 that works if I put if a!=b and (a,b) not in test cause they need to be different — user12195705, Apr 02 '20 at 21:25
for a in range(3): for b in range(a+1,3): if (a,b) not in test: test[a,b]=4 — splash58, Apr 02 '20 at 21:27

score 1 · Answer 1 · answered Apr 02 '20 at 21:27

If you want a way to be sure that (a,b) and (b,a) never both get into the dictionary, you can loop like this:

test = {(0,1):1, (0,2):2}
for a in range(3):
    for b in range(a+1, 3):
        if (b,a) not in test.keys(): #pretty sure it's better to ask for .keys()
            test[a,b]=4

If you want to treat your keys as the appropriate data type, then don't use a tuple. Your business problem indicates that order doesn't matter. Tuples are ordered. So use something that isn't.

test = {set((0,1)):1, set((0,2)):2}
for a in range(3):
    for b in range(3):
        if a!= b:
            if set((b,a)) not in test.keys():
                test[set(a,b)]=4

score -1 · Answer 2 · answered Apr 02 '20 at 21:11

-1

I'm not sure why this isn't Pythonic.

I would, however, make sure that (a,b) is in the dict only if a<=b, which means you don't have to check both pairs every time - just one of them.

answered Apr 02 '20 at 21:11

zmbq

38,013
14
101
171

"flat is better than nested" - PEP8 ? – Neil Apr 02 '20 at 21:12
@zmbq doesnt work cause it only gets (0,1) and (0,2) but not (1,2) – user12195705 Apr 02 '20 at 21:16

pythonic way of avoiding duplicated dictionary keys

2 Answers2