I am merging two json in python
I'm doing
import json
json_obj = json.dumps({"a": [1,2]})
json_obj1 = json.dumps({"a": [3,4]})
json_obj += json_obj1
print(json_obj)
I am expecting the output as
{"a": [1, 2,3,4]}
but i got
{"a": [1, 2]}{"a": [3, 4]}
How to get the earlier one?
In json module, dumps convert python object to a string, and loads convert a string into python object. So in your original codes, you just try to concat two json-string. Try to code like this:
import json
from collections import defaultdict
def merge_dict(d1, d2):
dd = defaultdict(list)
for d in (d1, d2):
for key, value in d.items():
if isinstance(value, list):
dd[key].extend(value)
else:
dd[key].append(value)
return dict(dd)
if __name__ == '__main__':
json_str1 = json.dumps({"a": [1, 2]})
json_str2 = json.dumps({"a": [3, 4]})
dct1 = json.loads(json_str1)
dct2 = json.loads(json_str2)
combined_dct = merge_dict(dct1, dct2)
json_str3 = json.dumps(combined_dct)
# {"a": [1, 2, 3, 4]}
print(json_str3)
json.dumps() converts a dictionary to str object, not a json(dict) object.
So, adding some dumps statement in your code shows that the type is changed to str after using json.dumps() and with + you are effectively concatenating the two string and hence you get the concatenated output.
Further, to merge the two dictionaries for your simple case, you can just use the append:
import json
json_obj = json.dumps({"a": [1,2]})
json_obj1 = json.dumps({"a": [3,4]})
print(type(json_obj1)) # the type is `str`
json_obj += json_obj1 # this concatenates the two str objects
json_obj = {"a": [1,2]}
json_obj1 = {"a": [3,4]}
json_obj["a"].extend(json_obj1["a"])
print(json_obj)
I suggest you to study basic fundamental of Python for your own sake as you don't seem to understand why your code wouldn't work.
import json
# We have two dictionaries to combine
json_obj_1 = {"a": [1,2], "b":[2,3], 'c': [1,2,3]}
json_obj_2 = {"a": [3,4], 'd':[4,2], 'e': [4,2,2]}
Merged dictionary will be stored here
hold_json_obj = {}
Don't worry, it's not actually that complicated. Read the code line by line with comments attached and you'll understand.
# We'll loop through every item in the json_obj_1 dictionary
for item_1 in json_obj_1:
# We'll also loop through every item in the json_obj_2 dictionary
for item_2 in json_obj_2:
# Now let's compare whether they are the same KEYS (not values)
if item_1 == item_2:
# if they match, we create a list to store the array
hold_array = []
hold_array.extend(json_obj_1[item_1])
hold_array.extend(json_obj_2[item_1])
# finally putting the array to our hold_json_obj
hold_json_obj[item_1] = hold_array
else:
# if they don't match, check if the key already exists in the
# hold_json_obj because we might be iterating json_obj_2 for the second time.
if item_2 not in hold_json_obj:
#add the ummatched array to hold_json_obj
hold_json_obj[item_2] = json_obj_2[item_2]
Now simply update json_obj_1 with the update method. The update function is required because if json_obj_1 has keys that json_obj_2 doesn't then we may have missed them out in the above loops.
json_obj_1.update(hold_json_obj)
print(json_obj_1)
This is what the print displays.
{'a': [1, 2, 3, 4], 'b': [2, 3], 'c': [1, 2, 3], 'd': [4, 2], 'e': [4, 2, 2]}
It's important to differentiate between JSON Strings and JSON objects. I use this function to merge 2 complex JSON objects.
def merge_json_objects(obj1, obj2):
if isinstance(obj1, dict) and isinstance(obj2, dict):
merged = obj1.copy()
for key, value in obj2.items():
if key in merged:
merged[key] = merge_json_objects(merged[key], value)
else:
merged[key] = value
return merged
elif isinstance(obj1, list) and isinstance(obj2, list):
# For lists, extend obj1 with items from obj2 that are not already in obj1
merged = obj1.copy()
for item in obj2:
if item not in merged: #avoid duplicates
merged.append(item)
return merged
elif isinstance(obj1, str) and isinstance(obj2, dict) or isinstance(obj2, str) and isinstance(obj1, dict) : # If one is a dict and the other a string, they belong to different places, so we ignore one or the other.
return obj1
else:
# Incompatible types, return obj1 by default
return obj1
Here is how you can use it:
obj1 = {
"world": {
"Europe": {
"France": {
"1960": {
"song1": "path1",
"song2": "path2"
},
"_unknown":{
"song_unknown": "path_unknown"
}
}
}
}
}
obj2 = {
"world": {
"Europe": {
"France": {
"1970": {
"song3": "path3"
},
"1960": {
"song4": "path4"
}
},
"Spain": {
"1980": {
"song5": "path5"
}
}
}
}
}
merged_object = merge_json_objects(obj1, obj2)
Then, with the resultant JSON Object, you can get the JSON String that represents that object and save that JSON String as a JSON File:
import json
# Get the string with:
json_string = json.dumps(merged_object, indent=4)
# or Save the JSON Object as a JSON String in a JSON File:
with open("myfile.json", "w", encoding="utf-8") as f:
json.dump(merged_object, f, indent=4)
