将解压元组列表写入 CSV

Question

I am trying to write a list of tuples (which is not flat) into a CSV. So for example the input would be:

[(('temperature', 42), ('date', datetime.date(2017, 1, 22)), ('locations', ('Berlin', 'Paris')), ('weather', 'sunny')), 
(('temperature', -42), ('date', datetime.date(2017, 1, 22)), ('locations', ('Marseille', 'Moscow')), ('weather', 'cloudy'))]

and output should be

temperature,date,locations,weather
42,01/22/2017,"Berlin,Paris",sunny
-42,01/22/2017,"Marseille,Moscow",cloudy

However what I am getting is

temperature,date,locations,weather
42
2017-01-22
"('Berlin', 'Paris')"
sunny
-42
2017-01-22
"('Marseille', 'Moscow')"
cloudy

My code looks like this:

import csv
import datetime

def generate_csv(a_list):
    with open('results.csv', 'w', newline='') as csvfile:
        writer = csv.writer(csvfile, delimiter=',', quotechar='"')
        #writer.writerow(['temperature','date','locations','weather'])
        for i in a_list:
            for j in i:
                print(j[1:])
                writer.writerow(j[1:])
            
test = [(('temperature', 42), ('date', datetime.date(2017, 1, 22)), ('locations', ('Berlin', 'Paris')),  ('weather', 'sunny')),
(('temperature', -42), ('date', datetime.date(2017, 1, 22)), ('locations', ('Marseille', 'Moscow')), ('weather', 'cloudy'))]

generate_csv(test)

The instructions for this example are:

Create a function generate_csv(a_list) that will create a csv file called results.csv.

Your function will be called with a list of tuples (like in the following examples). 
The tuples will contains tuples which contains as a first a key and as a second value the value
associated with the key. The keys will always be the same. You must show this keys in 
the first line of your csv. After this line you need to add the values formatted like this :

a list or a tuple must be string: "a,b,c"
a date must follow the US Standard: "month/day/year"
You don't need to format the other values.

Your csv must use ',' as separator and '"' as quotechar.

How can I get rid of the newline between each line I write? What is the best way to format the tuples and the datetime so it matches the requested output? If I convert the output to str like this:

writer.writerow(str(j[1:]))

The output is a mess and becomes this:

(,4,2,",",)
(,d,a,t,e,t,i,m,e,.,d,a,t,e,(,2,0,1,7,",", ,1,",", ,2,2,),",",)
(,(,',B,e,r,l,i,n,',",", ,',P,a,r,i,s,',),",",)
(,',s,u,n,n,y,',",",)
(,-,4,2,",",)
(,d,a,t,e,t,i,m,e,.,d,a,t,e,(,2,0,1,7,",", ,1,",", ,2,2,),",",)
(,(,',M,a,r,s,e,i,l,l,e,',",", ,',M,o,s,c,o,w,',),",",)
(,',c,l,o,u,d,y,',",",)

Answer 1

You need to create a dictionary out of the nested tuples by using map

import csv

test = list(map(dict, test))
keys = test[0].keys()
with open('result.csv', 'w', newline='') as op:
    dict_writer = csv.DictWriter(op, keys)
    dict_writer.writeheader()
    dict_writer.writerows(test)

An alternative using pandas package if you have a lot of data and multiple manipulations on the columns.

import pandas as pd

df = pd.DataFrame(map(dict, test))
df.to_csv('result.csv')