展平 json 个对象

Question

我有json个字符串

{"search_query": "51", "limit": 4, "groups": {"type": {"group_count": 1, "groups": [{"docs": [{"price_1": 2.99, "productCode": "000053", "price_3": 5.49, "priceOld_2": 2.99, "discount_2": 12.0, "discount_3": 3.0, "discount_1": 0.0, "priceOld_1": 0.0, "image": "777.jpg", "title": "Advanced/00", "priceDefault": 2.99, "priceLoyalty": 5.49, "addId": "141918", "url": "url", "price_2": 9.0921, "inStock": true, "measurementUnit": "vnt.", "priceOld_3": 2.99, "type": "product", "id": "product1436", "tags": ["price", "in", "out"]}], "value": "product", "doc_count": 1}], "doc_count": 1}}, "doc_count": 1}

import pandas
import json

def flatten_json(y):
    out = {}

    def flatten(x, name='Tags'):
        if type(x) is dict:
            for a in x:
                print(a)
                flatten(x[a], name + a + '_')
        elif type(x) is list:
            i = 0
            for a in x:
                flatten(a, name + str(i) + '_')
                i += 1
        else:
            out[str(name[:-1])] = str(x)

    flatten(y)
    return out

# load data using Python JSON module
with open('11.txt','r') as f:
    data = json.loads(f.read())
# Normalizing data
# format_All = pd.json_normalize(data['groups']['type']['groups'], 'docs') # this work normally but need to be flatten TAGS
format_Groups = data['groups']['type']['groups']
flat_Tags = flatten_json(format_Groups)
format_All = pd.json_normalize(flat_Tags)
# Saving to CSV format
format_All.to_csv('2.csv', index=False)

但是我得到了错误的 csv，那么如何展平 TAGS 列表？并将每一列命名为“价格”、“输入”、“输出”

，而不是 Tag1、Tag2

csv 输出

Tags0_docs_0_price_1,Tags0_docs_0_productCode,Tags0_docs_0_price_3,Tags0_docs_0_priceOld_2,Tags0_docs_0_discount_2,Tags0_docs_0_discount_3,Tags0_docs_0_discount_1,Tags0_docs_0_priceOld_1,Tags0_docs_0_image,Tags0_docs_0_title,Tags0_docs_0_priceDefault,Tags0_docs_0_priceLoyalty,Tags0_docs_0_addId,Tags0_docs_0_url,Tags0_docs_0_price_2,Tags0_docs_0_inStock,Tags0_docs_0_measurementUnit,Tags0_docs_0_priceOld_3,Tags0_docs_0_type,Tags0_docs_0_id,Tags0_docs_0_tags_0,Tags0_docs_0_tags_1,Tags0_docs_0_tags_2,Tags0_value,Tags0_doc_count
2.99,000053,5.49,2.99,12.0,3.0,0.0,0.0,777.jpg,Advanced/00,2.99,5.49,141918,url,9.0921,True,vnt.,2.99,product,product1436,price,in,out,product,1

预计

price_1,productCode,price_3,priceOld_2,discount_2,discount_3,discount_1,priceOld_1,image,title,priceDefault,priceLoyalty,addId,url,price_2,inStock,measurementUnit,priceOld_3,type,id,price,in,out
2.99,000053,5.49,2.99,12.0,3.0,0.0,0.0,777.jpg,Advanced/00,2.99,5.49,141918,url,9.0921,True,vnt.,2.99,product,product1436,price,in,out

Answer 1

import pandas as pd

# load data using Python JSON module
data = {
    "search_query": "51",
    "limit": 4,
    "groups": {
        "type":
            {"group_count": 1,
             "groups":
                 [
                     {
                         "docs": [
                             {"price_1": 2.99, "productCode": "000053", "price_3": 5.49, "priceOld_2": 2.99,
                              "discount_2": 12.0,
                              "discount_3": 3.0, "discount_1": 0.0, "priceOld_1": 0.0, "image": "777.jpg",
                              "title": "Advanced/00",
                              "priceDefault": 2.99, "priceLoyalty": 5.49, "addId": "141918", "url": "url",
                              "price_2": 9.0921,
                              "inStock": True,
                              "measurementUnit": "vnt.", "priceOld_3": 2.99, "type": "product", "id": "product1436",
                              "tags": ["price", "in", "out"]
                              }],
                         "value": "product",
                         "doc_count": 1
                     }
                 ],
             "doc_count": 1}},
    "doc_count": 1}

# select fields
fields_of_interest = data['groups']['type']['groups'][0]["docs"][0]

# flatten field "tags"
for value in fields_of_interest["tags"]:
    fields_of_interest[value] = value
fields_of_interest.pop("tags")

# export csv
df = pd.DataFrame.from_dict(fields_of_interest, orient='index').transpose()
df.to_csv("flatten_dataframe.csv", index=None)

输出（来自文件）：

price_1,productCode,price_3,priceOld_2,discount_2,discount_3,discount_1,priceOld_1,image,title,priceDefault,priceLoyalty,addId,url,price_2,inStock,measurementUnit,priceOld_3,type,id,price,in,out
2.99,000053,5.49,2.99,12.0,3.0,0.0,0.0,777.jpg,Advanced/00,2.99,5.49,141918,url,9.0921,True,vnt.,2.99,product,product1436,price,in,out

展平 json 个对象

Flatten json object

python

json

flatten

pandas