How can I convert JSON to CSV?

Question

I have a JSON file I want to convert to a CSV file. How can I do this with Python?

I tried:

import json
import c         


        

Answer 1

Try this

import csv, json, sys

input = open(sys.argv[1])
data = json.load(input)
input.close()

output = csv.writer(sys.stdout)

output.writerow(data[0].keys())  # header row

for item in data:
    output.writerow(item.values())

Answer 2

Modified Alec McGail's answer to support JSON with lists inside

    def flattenjson(self, mp, delim="|"):
            ret = []
            if isinstance(mp, dict):
                    for k in mp.keys():
                            csvs = self.flattenjson(mp[k], delim)
                            for csv in csvs:
                                    ret.append(k + delim + csv)
            elif isinstance(mp, list):
                    for k in mp:
                            csvs = self.flattenjson(k, delim)
                            for csv in csvs:
                                    ret.append(csv)
            else:
                    ret.append(mp)

            return ret

Thanks!

Answer 3

I might be late to the party, but I think, I have dealt with the similar problem. I had a json file which looked like this

I only wanted to extract few keys/values from these json file. So, I wrote the following code to extract the same.

    """json_to_csv.py
    This script reads n numbers of json files present in a folder and then extract certain data from each file and write in a csv file.
    The folder contains the python script i.e. json_to_csv.py, output.csv and another folder descriptions containing all the json files.
"""

import os
import json
import csv


def get_list_of_json_files():
    """Returns the list of filenames of all the Json files present in the folder
    Parameter
    ---------
    directory : str
        'descriptions' in this case
    Returns
    -------
    list_of_files: list
        List of the filenames of all the json files
    """

    list_of_files = os.listdir('descriptions')  # creates list of all the files in the folder

    return list_of_files


def create_list_from_json(jsonfile):
    """Returns a list of the extracted items from json file in the same order we need it.
    Parameter
    _________
    jsonfile : json
        The json file containing the data
    Returns
    -------
    one_sample_list : list
        The list of the extracted items needed for the final csv
    """

    with open(jsonfile) as f:
        data = json.load(f)

    data_list = []  # create an empty list

    # append the items to the list in the same order.
    data_list.append(data['_id'])
    data_list.append(data['_modelType'])
    data_list.append(data['creator']['_id'])
    data_list.append(data['creator']['name'])
    data_list.append(data['dataset']['_accessLevel'])
    data_list.append(data['dataset']['_id'])
    data_list.append(data['dataset']['description'])
    data_list.append(data['dataset']['name'])
    data_list.append(data['meta']['acquisition']['image_type'])
    data_list.append(data['meta']['acquisition']['pixelsX'])
    data_list.append(data['meta']['acquisition']['pixelsY'])
    data_list.append(data['meta']['clinical']['age_approx'])
    data_list.append(data['meta']['clinical']['benign_malignant'])
    data_list.append(data['meta']['clinical']['diagnosis'])
    data_list.append(data['meta']['clinical']['diagnosis_confirm_type'])
    data_list.append(data['meta']['clinical']['melanocytic'])
    data_list.append(data['meta']['clinical']['sex'])
    data_list.append(data['meta']['unstructured']['diagnosis'])
    # In few json files, the race was not there so using KeyError exception to add '' at the place
    try:
        data_list.append(data['meta']['unstructured']['race'])
    except KeyError:
        data_list.append("")  # will add an empty string in case race is not there.
    data_list.append(data['name'])

    return data_list


def write_csv():
    """Creates the desired csv file
    Parameters
    __________
    list_of_files : file
        The list created by get_list_of_json_files() method
    result.csv : csv
        The csv file containing the header only
    Returns
    _______
    result.csv : csv
        The desired csv file
    """

    list_of_files = get_list_of_json_files()
    for file in list_of_files:
        row = create_list_from_json(f'descriptions/{file}')  # create the row to be added to csv for each file (json-file)
        with open('output.csv', 'a') as c:
            writer = csv.writer(c)
            writer.writerow(row)
        c.close()


if __name__ == '__main__':
    write_csv()

I hope this will help. For details on how this code work you can check here

Answer 4

Surprisingly, I found that none of the answers posted here so far correctly deal with all possible scenarios (e.g., nested dicts, nested lists, None values, etc).

This solution should work across all scenarios:

def flatten_json(json):
    def process_value(keys, value, flattened):
        if isinstance(value, dict):
            for key in value.keys():
                process_value(keys + [key], value[key], flattened)
        elif isinstance(value, list):
            for idx, v in enumerate(value):
                process_value(keys + [str(idx)], v, flattened)
        else:
            flattened['__'.join(keys)] = value

    flattened = {}
    for key in json.keys():
        process_value([key], json[key], flattened)
    return flattened

Answer 5

JSON can represent a wide variety of data structures -- a JS "object" is roughly like a Python dict (with string keys), a JS "array" roughly like a Python list, and you can nest them as long as the final "leaf" elements are numbers or strings.

CSV can essentially represent only a 2-D table -- optionally with a first row of "headers", i.e., "column names", which can make the table interpretable as a list of dicts, instead of the normal interpretation, a list of lists (again, "leaf" elements can be numbers or strings).

So, in the general case, you can't translate an arbitrary JSON structure to a CSV. In a few special cases you can (array of arrays with no further nesting; arrays of objects which all have exactly the same keys). Which special case, if any, applies to your problem? The details of the solution depend on which special case you do have. Given the astonishing fact that you don't even mention which one applies, I suspect you may not have considered the constraint, neither usable case in fact applies, and your problem is impossible to solve. But please do clarify!

Answer 6

As mentioned in the previous answers the difficulty in converting json to csv is because a json file can contain nested dictionaries and therefore be a multidimensional data structure verses a csv which is a 2D data structure. However, a good way to turn a multidimensional structure to a csv is to have multiple csvs that tie together with primary keys.

In your example, the first csv output has the columns "pk","model","fields" as your columns. Values for "pk", and "model" are easy to get but because the "fields" column contains a dictionary, it should be its own csv and because "codename" appears to the be the primary key, you can use as the input for "fields" to complete the first csv. The second csv contains the dictionary from the "fields" column with codename as the the primary key that can be used to tie the 2 csvs together.

Here is a solution for your json file which converts a nested dictionaries to 2 csvs.

import csv
import json

def readAndWrite(inputFileName, primaryKey=""):
    input = open(inputFileName+".json")
    data = json.load(input)
    input.close()

    header = set()

    if primaryKey != "":
        outputFileName = inputFileName+"-"+primaryKey
        if inputFileName == "data":
            for i in data:
                for j in i["fields"].keys():
                    if j not in header:
                        header.add(j)
    else:
        outputFileName = inputFileName
        for i in data:
            for j in i.keys():
                if j not in header:
                    header.add(j)

    with open(outputFileName+".csv", 'wb') as output_file:
        fieldnames = list(header)
        writer = csv.DictWriter(output_file, fieldnames, delimiter=',', quotechar='"')
        writer.writeheader()
        for x in data:
            row_value = {}
            if primaryKey == "":
                for y in x.keys():
                    yValue = x.get(y)
                    if type(yValue) == int or type(yValue) == bool or type(yValue) == float or type(yValue) == list:
                        row_value[y] = str(yValue).encode('utf8')
                    elif type(yValue) != dict:
                        row_value[y] = yValue.encode('utf8')
                    else:
                        if inputFileName == "data":
                            row_value[y] = yValue["codename"].encode('utf8')
                            readAndWrite(inputFileName, primaryKey="codename")
                writer.writerow(row_value)
            elif primaryKey == "codename":
                for y in x["fields"].keys():
                    yValue = x["fields"].get(y)
                    if type(yValue) == int or type(yValue) == bool or type(yValue) == float or type(yValue) == list:
                        row_value[y] = str(yValue).encode('utf8')
                    elif type(yValue) != dict:
                        row_value[y] = yValue.encode('utf8')
                writer.writerow(row_value)

readAndWrite("data")