def test_cheapest_gets_info(self):
user_request = 'audifonos inalambricos'
country = 'mx'
ebay_url = Ebay.adapt_url(Ebay, user_request, country)
ebay_soup = extract_soup(ebay_url, 1, just_soup=True)
ebay_boxes = search_boxes(ebay_soup, Ebay.boxes)
ebay_prices = get_price(country, ebay_boxes, Ebay.price)
ebay_cheapest_idx, ebay_cheapest_price = cheapest(
ebay_prices, position_and_price=True)
cheapest_ebay_product_1 = get_cheapest(ebay_cheapest_idx, ebay_boxes,
ebay_cheapest_price, country,
Ebay)
for value in cheapest_ebay_product_1:
self.assertIsNotNone(cheapest_ebay_product_1[value])
def test_cheapest_gets_info(self):
user_request = 'audifonos inalambricos'
country = 'mx'
ml_url = Mercado_Libre.adapt_url(Mercado_Libre, user_request, country)
ml_soup = extract_soup(ml_url, 1, just_soup=True)
ml_boxes = search_boxes(ml_soup, Mercado_Libre.boxes)
meli_prices = get_price(country, ml_boxes, Mercado_Libre.price)
meli_cheapest_idx, meli_cheapest_price = cheapest(
meli_prices, position_and_price=True)
cheapest_ml_product_1 = get_cheapest(meli_cheapest_idx, ml_boxes,
meli_cheapest_price, country,
Mercado_Libre)
for value in cheapest_ml_product_1:
self.assertIsNotNone(cheapest_ml_product_1[value])
def scraper(Page, user_request, country):
#Adapt the url
url = Page.adapt_url(Page, country, user_request)
#All the HTML of the page
soup = extract_soup(url, 1, just_soup=True)
# #HTML divided by products, and stored as elements of an array
boxes = search_boxes(soup, Page.boxes)
# From this part, could get better AFTER the 4 scrapers are made
#From the Boxes, obtain the prices
prices = get_price(country, boxes, Page.price)
#Obtain the cheapest from prices and then, you obtain the cheapest product as a dictionary
cheapest_idx, cheapest_price = cheapest(prices, position_and_price=True)
cheapest_product_dictionary = get_cheapest(cheapest_idx, boxes,
cheapest_price, country, Page)
return cheapest_product_dictionary
def scraper(user_request, country):
#Adapt the url
ml_url = Mercado_Libre.adapt_url(Mercado_Libre, country, user_request)
#All the HTML of the page
ml_soup = extract_soup(ml_url, 1, just_soup=True)
# #HTML divided by products, and stored as elements of an array
ml_boxes = search_boxes(ml_soup, Mercado_Libre.boxes)
# From this part, could get better AFTER the 4 scrapers are made
#From the Boxes, obtain the prices
meli_prices = get_price(country, ml_boxes, Mercado_Libre.price)
#Obtain the cheapest from prices and then, you obtain the cheapest product as a dictionary
meli_cheapest_idx, meli_cheapest_price = cheapest(meli_prices,
position_and_price=True)
cheapest_ml_product_dictionary = get_cheapest(meli_cheapest_idx, ml_boxes,
meli_cheapest_price, country,
Mercado_Libre)
return cheapest_ml_product_dictionary
def scraper(user_request, country):
#Adapt the url
amazon_url = Amazon.adapt_url(Amazon, user_request, country)
#All the HTML of the page
amazon_soup = extract_soup(amazon_url, 1, just_soup=True)
# #HTML divided by products, and stored as elements of an array
amazon_boxes = search_boxes(amazon_soup, Amazon.boxes)
# From this part, could get better AFTER the 4 scrapers are made
#From the Boxes, obtain the prices
amazon_prices = get_price(country, amazon_boxes, Amazon.price)
#Obtain the cheapest from prices and then, you obtain the cheapest product as a dictionary
amazon_cheapest_idx, amazon_cheapest_price = cheapest(
amazon_prices, position_and_price=True)
cheapest_amazon_product_dictionary = get_cheapest(amazon_cheapest_idx,
amazon_boxes,
amazon_cheapest_price,
country, Amazon)
return cheapest_amazon_product_dictionary
amazon_soup = extract_soup(amazon_url, 1, just_soup=True)
#HTML divided by products, and stored as elements of an array
amazon_boxes = search_boxes(amazon_soup, Amazon.boxes)
amazon_products = {}
amazon_products['name'] = get_names(amazon_boxes, Amazon.name_and_images)
'''Amazon's images source (link)'''
amazon_products['image'] = get_images(amazon_boxes, Amazon)
amazon_products['url'] = get_products_urls(amazon_boxes, Amazon)
'''Just Amazon's products id. Is used as a url generator:
amazon's url + domain + "/dp/" + product_id'''
# amazon_products['id']= amazon_products_id(amazon_boxes)
'''Just stars as float'''
amazon_products['star'] = get_stars(country, amazon_boxes, Amazon.stars)
'''Just number of reviews as int'''
amazon_products['review'] = get_reviews(country, amazon_boxes,
Amazon.reviews)
amazon_products['price'] = get_price(country, amazon_boxes, Amazon.price)
# print(len(amazon_reviews))
# for key in amazon_products:
# print(key, ':', amazon_products[key])
cheapest = cheapest(amazon_products['price'])
cheapest_amazon_product = get_cheapest(cheapest, amazon_products)
for key in cheapest_amazon_product:
print(key, ':', cheapest_amazon_product[key])
# for highlight in amazon_highlightners:
# print(highlight)
return cheapest_product_dictionary
if __name__ == "__main__":
user_request = 'audifonos inalambricos'
country = 'mx'
ebay_url = Ebay.adapt_url(Ebay, user_request, country)
#All the HTML of the page
ebay_soup = extract_soup(ebay_url, 1, just_soup=True)
# #HTML divided by products, and stored as elements of an array
ebay_boxes = search_boxes(ebay_soup, Ebay.boxes)
# print(ebay_boxes)
ebay_products = {}
ebay_products['names'] = get_names(ebay_boxes, Ebay.name_and_images)
# #Ebay's images source (link)
ebay_products['images'] = get_images(ebay_boxes, Ebay)
ebay_products['urls'] = get_products_urls(ebay_boxes, Ebay)
ebay_products['prices'] = get_price(country, ebay_boxes, Ebay.price)
cheapest_idx = cheapest(ebay_products['prices'])
cheapest_ebay_product2 = get_cheapest(cheapest_idx, ebay_products)
print(f'\nTest ONE:')
for key in cheapest_ebay_product2:
print(key, ':', cheapest_ebay_product2[key])
user_request = 'audifonos inalambricos'
country = 'mx'
ml_url = Mercado_Libre.adapt_url(Mercado_Libre, user_request, country)
#All the HTML of the page
ml_soup = extract_soup(ml_url, 0, just_soup=True)
# #HTML divided by products, and stored as elements of an array
ml_boxes = search_boxes(ml_soup, Mercado_Libre.boxes)
print(f'Test ONE:')
meli_prices = get_price(country, ml_boxes, Mercado_Libre.price)
meli_cheapest_idx, meli_cheapest_price = cheapest(meli_prices,
position_and_price=True)
cheapest_ml_product_1 = get_cheapest(meli_cheapest_idx, ml_boxes,
meli_cheapest_price, country,
Mercado_Libre)
for key in cheapest_ml_product_1:
print(key, ':', cheapest_ml_product_1[key])
# # print('boxes:', len(ml_boxes))
ml_products = {}
ml_products['names'] = get_names(ml_boxes, Mercado_Libre.name_and_images)
#Mercado_Libre's images source (link)
ml_products['images'] = get_images(ml_boxes, Mercado_Libre)
ml_products['urls'] = get_products_urls(ml_boxes, Mercado_Libre)
ml_products['prices'] = get_price(country, ml_boxes, Mercado_Libre.price)
cheapest_idx = cheapest(ml_products['prices'])