The logic in the main function, json_to_dataframe is then the same as before. This pretty much mimics the ncat behaviour in this case. I added the flatten_list generator as I wanted to make sure that the JSON arrays are all nice and flattened, then provided as a single list of dictionaries comprising of the previous key from one iteration before assigned to each of the list's values.

OUTPUT: id type name ppu topping.id topping.typeĠ 0001 donut Cake 0.55 1001 Regular 5001 Noneġ 0001 donut Cake 0.55 1001 Regular 5002 GlazedĢ 0001 donut Cake 0.55 1001 Regular 5005 Sugarģ 0001 donut Cake 0.55 1001 Regular 5007 Powdered SugarĤ 0001 donut Cake 0.55 1001 Regular 5006 Chocolate with Sprinklesĥ 0001 donut Cake 0.55 1001 Regular 5003 ChocolateĦ 0001 donut Cake 0.55 1001 Regular 5004 Mapleħ 0001 donut Cake 0.55 1002 Chocolate 5001 NoneĨ 0001 donut Cake 0.55 1002 Chocolate 5002 Glazedĩ 0001 donut Cake 0.55 1002 Chocolate 5005 Sugarġ0 0001 donut Cake 0.55 1002 Chocolate 5007 Powdered Sugarġ1 0001 donut Cake 0.55 1002 Chocolate 5006 Chocolate with Sprinklesġ2 0001 donut Cake 0.55 1002 Chocolate 5003 Chocolateġ3 0001 donut Cake 0.55 1002 Chocolate 5004 Mapleġ4 0001 donut Cake 0.55 1003 Blueberry 5001 Noneġ5 0001 donut Cake 0.55 1003 Blueberry 5002 Glazedġ6 0001 donut Cake 0.55 1003 Blueberry 5005 Sugarġ7 0001 donut Cake 0.55 1003 Blueberry 5007 Powdered Sugarġ8 0001 donut Cake 0.55 1003 Blueberry 5006 Chocolate with Sprinklesġ9 0001 donut Cake 0.55 1003 Blueberry 5003 ChocolateĢ0 0001 donut Cake 0.55 1003 Blueberry 5004 MapleĢ1 0001 donut Cake 0.55 1004 Devil's Food 5001 NoneĢ2 0001 donut Cake 0.55 1004 Devil's Food 5002 GlazedĢ3 0001 donut Cake 0.55 1004 Devil's Food 5005 SugarĢ4 0001 donut Cake 0.55 1004 Devil's Food 5007 Powdered SugarĢ5 0001 donut Cake 0.55 1004 Devil's Food 5006 Chocolate with SprinklesĢ6 0001 donut Cake 0.55 1004 Devil's Food 5003 ChocolateĢ7 0001 donut Cake 0.55 1004 Devil's Food 5004 MapleĪs per what the above does, well, the cross_join function does pretty much the same thing as in the dataframe solution, but without dataframes, thus being faster. Return left.assign(key=1).merge(right.assign(key=1), on='key', how='outer').drop('key', 1)ĭf = cross_join(df, to_frame(data, prev_key + '.' + key))ĭf = ncat(, prev_key)])ĭf = pandas.DataFrame(,

This is an older question, but I struggled the entire night to get a satisfactory result for a similar situation, and I came up with this: import json Please scroll down for the newer, faster solution

Nested json to csv python