从 python 中的字典中删除多个键值对

Question

I generated a python dictionary for all the duplicate images in a folder. The python dictonary now contains values in the following format:

{
  "image_1.jpg": ['image_xyz.jpg', 'image_abc.jpg'],
  "image_xyz.jpg": ["image_1.jpg", "image_abc.jpg"],
  "image_abc.jpg": ["image_xyz.jpg","image_1.jpg"],
  "image_2.jpg": ["image_3.jpg"],
  "image_3.jpg": ["image_2.jpg"],
  "image_5.jpg": []
}

Each key, value pair thus appears atleast twice in the list. Empty list for keys are present which have no duplicates. Is there a way to delete all the duplicate key value pairs present? so that the dictionary looks like the following:

{
  "image_1.jpg": ['image_xyz.jpg', 'image_abc.jpg'],
  "image_2.jpg": ["image_3.jpg"],
  "image_5.jpg": []
}

I tried using list to first store all the values from the key value pair and then deleting them from the dictionary but it empties the whole dictionary.

Answer 1

source = {
  "image_1.jpg": ['image_xyz.jpg', 'image_abc.jpg'],
  "image_xyz.jpg": ["image_1.jpg", "image_abc.jpg"],
  "image_abc.jpg": ["image_xyz.jpg","image_1.jpg"],
  "image_2.jpg": ["image_3.jpg"],
  "image_3.jpg": ["image_2.jpg"],
  "image_5.jpg": []
}

dest = dict()

for k,v in source.items():
    ok = True
    for k1,v1 in dest.items():
        if k in v1: ok = False
    if ok: dest[k] = v

print(dest) # New filtered dict

Answer 2

I usually do this method when getting rid of duplicates in a list:

First put all the values in a matrix / 2 dimensional list including the key, so the first 3 values would be like this:

{
  "image_1.jpg": ['image_xyz.jpg', 'image_abc.jpg'],
  "image_xyz.jpg": ["image_1.jpg", "image_abc.jpg"],
  "image_abc.jpg": ["image_xyz.jpg","image_1.jpg"],
}

would turn into:

List=[
  ["image_1.jpg","image_xyz.jpg","image_abc.jpg"],
  ["image_xyz.jpg","image_1.jpg","image_abc"],
  ["image_abc.jpg","image_xyz.jpg","image_1.jpg"]
]

make sure the keys are all in the 0th position so that you can save them.

keys=[x[0] for x in List]

then sort the list:

sorted_list=[sorted(x) for x in List]

then simply compare them using an if statement in a nested for loop and if a list is equal to another then delete it:

for i in sorted_list:
    for j,k in enumerate(sorted_list):
        if i==k:
            del sorted_list[j] # deleting any equal lists

now that the duplicates are gone and you have the keys convert the list back into a dictionary if preferred

Overall code if needed:

List=[
  ["image_1.jpg","image_xyz.jpg","image_abc.jpg"],
  ["image_xyz.jpg","image_1.jpg","image_abc"],
  ["image_abc.jpg","image_xyz.jpg","image_1.jpg"]
]
keys=[x[0] for x in List]
sorted_list=[sorted(x) for x in List]
for i in sorted_list:
    for j,k in enumerate(sorted_list):
        if i==k:
            del sorted_list[j] # deleting any equal lists