import csv to json/dict with additional keywords/organization - json

Question

I'm trying to convert a .csv to json/dict such that the data in its current form:
cat1,cat2,cat3,name
1,2,3,a
4,5,6,b
7,8,9,c
I'm currently using something like this(as well as importing using pandas.df bc it will be used for graphing from json file):
with open('Data.csv') as f:
reader = csv.DictReader(f)
rows = list(reader)
print (rows)
[{'cat1': '1', 'name': 'a', 'cat3': '3', 'cat2': '2'},
{'cat1': '4', 'name': 'b', 'cat3': '6', 'cat2': '5'},
{'cat1': '7', 'name': 'c', 'cat3': '9', 'cat2': '8'}]
and I want it to look like this in json/dict format:
{"data: [{"all_cats": {"cat1": 1}, {"cat2": 2}, {"cat3": 3}}, "name": a},
{"all_cats": {"cat1": 4}, {"cat2": 5}, {"cat3": 6}}, "name": b},
{"all_cats": {"cat1": 7}, {"cat2": 8}, {"cat3": 8}}, "name": c}]}
Importing directly doesn't allow me to include: 'cat1', 'cat2', 'cat3' under 'all_cats' and keep 'name' separate.
Any help would be appreciated.

Answer 1

Since it's space separated and not comma separated you have to add delimiter=" ". Additionally since some of your rows have whitespace beforehand, that means you also have to add skipinitialspace=True.
reader = csv.DictReader(f, delimiter=" ", skipinitialspace=True)
rows = list(dict(row) for row in reader)
Thus if you now do:
for row in rows:
print(row)
The output will be:
{'cat1': '1', 'cat2': '2', 'cat3': '3', 'name': 'a'}
{'cat1': '4', 'cat2': '5', 'cat3': '6', 'name': 'b'}
{'cat1': '7', 'cat2': '8', 'cat3': '9', 'name': 'c'}

Answer 2

As already mentioned in the other answer you don't specify valid JSON format for what you want to achieve. You can check if a string contains valid JSON format using json.loads(jsonDATAstring) function:
import json
jsonDATAstring_1 = """
{"data: [{"all_cats": {"cat1": 1}, {"cat2": 2}, {"cat3": 3}}, "name": a},
{"all_cats": {"cat1": 4}, {"cat2": 5}, {"cat3": 6}}, "name": b},
{"all_cats": {"cat1": 7}, {"cat2": 8}, {"cat3": 8}}, "name": c}]}
"""
json.loads(jsonDATAstring_1)
what in case of the by you specified expected JSON format results in:
json.decoder.JSONDecodeError: Expecting ':' delimiter: line 2 column 12 (char 12)
From what is known to me from your question I assume, that the JSON string you want to get is a following one:
jsonDATAstring_2 = """
{"data": [{"all_cats": {"cat1": 1, "cat2": 2, "cat3": 3}, "name": "a"},
{"all_cats": {"cat1": 4, "cat2": 5, "cat3": 6}, "name": "b"},
{"all_cats": {"cat1": 7, "cat2": 8, "cat3": 8}, "name": "c"}]}
"""
json.loads(jsonDATAstring_2)
This second string loads OK, so assuming:
rows = [{'cat1': '1', 'name': 'a', 'cat3': '3', 'cat2': '2'},
{'cat1': '4', 'name': 'b', 'cat3': '6', 'cat2': '5'},
{'cat1': '7', 'name': 'c', 'cat3': '9', 'cat2': '8'}]
you can get what you want as follows:
dctData = {"data": []}
lstCats = ['cat1', 'cat2', 'cat3']
for row in rows:
dctAllCats = {"all_cats":{}, "name":"?"}
for cat in lstCats:
dctAllCats["all_cats"][cat] = row[cat]
dctAllCats["name"] = row["name"]
dctData["data"].append(dctAllCats)
import pprint
pp = pprint.PrettyPrinter()
pp.pprint(dctData)
what gives:
{'data': [{'all_cats': {'cat1': '1', 'cat2': '2', 'cat3': '3'}, 'name': 'a'},
{'all_cats': {'cat1': '4', 'cat2': '5', 'cat3': '6'}, 'name': 'b'},
{'all_cats': {'cat1': '7', 'cat2': '8', 'cat3': '9'}, 'name': 'c'}]}
Now it is possible to serialize the Python dictionary object to JSON string (or file):
jsonString = json.dumps(dctData)
print(jsonString)
what gives:
{"data": [{"all_cats": {"cat1": "1", "cat2": "2", "cat3": "3"}, "name": "a"}, {"all_cats": {"cat1": "4", "cat2": "5", "cat3": "6"}, "name": "b"}, {"all_cats": {"cat1": "7", "cat2": "8", "cat3": "9"}, "name": "c"}]}