test_data_builder.py

from faker import Faker

import datetime
import random
import sys
import csv

fake = Faker()

outfile = 'data_test.csv'
outsize = 1024 * 3024 

data = []

#Create people
for i in range(15):
    name = fake.name()
    age = fake.random_int(min=18, max=99)
    address = fake.address().replace('\n',' ')
    salary = fake.random_int(min=50000, max=250000)
    
    data.append((name, age, address, salary))
    
with open(outfile, 'w', newline='') as csvfile:
    
    #Creates lists for combination for random data
    amount_list = [random.randint(1,50) for i in range(30)]
    qty_list = [i for i in range(15)]
    purchase_date_list = [datetime.date(random.randint(2005,2015), 
                                      random.randint(1,12), 
                                      random.randint(1,28)).strftime('%Y-%m-%d')
                     for i in range(10000)]
    product_list = [
            'water', 'beer', 'chips', 'chocolate', 
            'pretzels', 'wine', 'fish', 'steak'
        ]
    
    size = 0
    csvwriter = csv.writer(csvfile)

    while csvfile.tell() < outsize:
        
        profile = random.choice(data)
        

        product = random.choice(product_list)
        amount = random.choice(amount_list)
        qty = random.choice(qty_list)
        purchase_date = random.choice(purchase_date_list)
        
        row = [
            profile[0], profile[1],
            profile[2], profile[3], 
            product, amount, qty,
            purchase_date
        ]
        
        csvwriter.writerow(row)
        
        p = float(csvfile.tell()) / outsize
        status = r"{0}  [{1:.2%}]".format(csvfile.tell(), p)
        status = status + chr(8)*(len(status)+1)
        sys.stdout.write(status)