Dict 1 :

{'result2': [{'2': '22'}, {'22': '222'}], 'result1': [{'1': '11'}, {'11': '111'}]}

Dict 2:

{'result2': [{'two': 'twentytwo'}, {'22': '222'}], 'result1': [{'one': 'eleven'}, {'11': '111'}]}

I want to merge them based on keys result1 and result 2 and preserve the same format.

Expected Output

{"result1":[{"1":"11"},{"one":"eleven"},{"11":"111"}]},{"result2":[{"2":"22"},{"two":"twentytwo"},{"22":"222"}]}

And the keys of both dict may not necessarily have same index.

CodePudding user response：

It's always good to add desired format for output. It likes you want to list of dict which contains data from both with respect to result1 & result 2.

Given -

a = [{"result1":[{"1":"11"},{"11":"111"}]},{"result2":[{"2":"22"},{"22":"222"}]}]
b = [{"result1":[{"one":"eleven"},{"11":"111"}]},{"result2":[{"two":"twentytwo"},{"22":"222"}]}]

Here is code-

desired_op = []
for index, item in enumerate(a):
   inner_item = {}
   for key, value in item.items():
     #print (key, value   b[index][key])
     inner_item[key] = value   b[index][key]
   desired_op.append(inner_item)
 
print (desired_op)
[{'result1': [{'1': '11'}, {'11': '111'}, {'one': 'eleven'}, {'11': '111'}]}, {'result2': [{'2': '22'}, {'22': '222'}, {'two': 'twentytwo'}, {'22': '222'}]}]

If you need unique items, it will be bit complex-

desired_op = []
 for index, item in enumerate(a):
   inner_item = {}
   for key, value in item.items():
     #print (key, value   b[index][key])
     inner_item[key] = list({list(litem.keys())[0]:litem for litem in value   b[index][key]}.values())
   desired_op.append(inner_item)
 
print (desired_op)
[{'result1': [{'1': '11'}, {'11': '111'}, {'one': 'eleven'}]}, {'result2': [{'2': '22'}, {'22': '222'}, {'two': 'twentytwo'}]}]

follow up on comment

simple append can work, but for this you must be sure that first index from second list should contain result1 & second index should contain result2 as dict keys.

ex. (but this won't remove duplicates)

a[0]['result1'].extend(b[0]['result1'])
a[1]['result2'].extend(b[1]['result2'])
print(a)
[{'result1': [{'1': '11'}, {'11': '111'}, {'one': 'eleven'}, {'11': '111'}]}, {'result2': [{'2': '22'}, {'22': '222'}, {'two': 'twentytwo'}, {'22': '222'}]}]

if you aren't sure about index position in second list, this will work-

keys_mapping =  {list(item.keys())[0]:index for index, item in enumerate(a)}
print (keys_mapping)
{'result1': 0, 'result2': 1}

for item in b:
  key = list(item.keys())[0]
  a[keys_mapping[key]][key].extend(item[key])
print (a)
[{'result1': [{'1': '11'}, {'11': '111'}, [{'one': 'eleven'}, {'11': '111'}]]}, {'result2': [{'2': '22'}, {'22': '222'}, [{'two': 'twentytwo'}, {'22': '222'}]]}]

CodePudding user response：

You could separate this in two phases:

1. merge the dictionaries by appending the lists
2. remove duplicates from the resulting merged dictionary

...

dict1={'result2': [{'2': '22'}, {'22': '222'}], 
       'result1': [{'1': '11'}, {'11': '111'}]}
dict2={'result2': [{'two': 'twentytwo'}, {'22': '222'}], 
       'result1': [{'one': 'eleven'}, {'11': '111'}]}

# merge phase
dict3 = {k:dict1.get(k,[]) dict2.get(k,[]) for k in {*dict1,*dict2}}
        
# uniqueness phase
dict3 = {k:[d for i,d in enumerate(v) if v.index(d)==i] 
         for k,v in dict3.items()}
    
print(dict3)

{'result2': [{'2': '22'}, {'22': '222'}, {'two': 'twentytwo'}],
 'result1': [{'1': '11'}, {'11': '111'}, {'one': 'eleven'}]}

If dict1 and dict2 are guaranteed to have the same keys, then the merge phase can be performed more concisely:

dict3 = {k:v dict2[k] for k,v in dict1.items()}

Note that you could combine the two phases in one large dictionary comprehension:

dict3 = {k: [d for i,d in enumerate(v) if v.index(d)==i]
         for k in {*dict1,*dict2}
         for v in [dict1.get(k,[]) dict2.get(k,[])] }