def test_products_info_getters(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)
getters = {
'ebay_names': len(get_names(ebay_boxes, Ebay.name_and_images)),
'ebay_images': len(get_images(ebay_boxes, Ebay)),
'ebay_urls': len(get_products_urls(ebay_boxes, Ebay)),
# 'ebay_price' : len(get_price(country, ebay_boxes, Ebay.price)),
}
for value in getters:
self.assertEqual(len(ebay_boxes), getters[value])
def test_products_info_getters(self):
user_request = 'audifonos inalambricos'
country = 'mx'
for Page in self.Pages:
page_url = Page.adapt_url(Page, user_request, country)
page_soup = extract_soup(page_url, 1, just_soup=True)
page_boxes = search_boxes(page_soup, Page.boxes)
#New test
page_names = len(get_names(page_boxes, Page))
page_images = len(get_images(page_boxes, Page))
page_urls = len(get_products_urls(page_boxes, Page))
page_price = len(get_price(country, page_boxes, Page))
trials = [page_names, page_images, page_urls, page_price]
for test in trials:
self.assertEqual(len(page_boxes), test)
def test_products_info_getters(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)
getters = {
'ml_names': len(get_names(ml_boxes,
Mercado_Libre.name_and_images)),
'ml_images': len(get_images(ml_boxes, Mercado_Libre)),
'ml_urls': len(get_products_urls(ml_boxes, Mercado_Libre)),
'ml_price': len(get_price(country, ml_boxes, Mercado_Libre.price)),
}
for value in getters:
self.assertEqual(len(ml_boxes), getters[value])
def get_cheapest(cheapest_idx,
products,
cheapest_price=None,
country=None,
Page=None):
products_type = str(type(products))
if products_type == "<class 'bs4.element.ResultSet'>":
if country and Page:
cheapest_dict = {}
cheapest_dict['name'] = get_names(products,
Page.name_and_images,
position=cheapest_idx)
cheapest_dict['image'] = get_images(products,
Page,
position=cheapest_idx)
cheapest_dict['url'] = get_products_urls(products,
Page.product_urls,
position=cheapest_idx)
cheapest_dict['price'] = cheapest_price
else:
if not country:
raise ValueError("Missing country value")
elif not Page:
raise ValueError("Missing Page object")
elif not cheapest_price:
raise ValueError("Missing cheapest_price value (int)")
else:
raise ValueError(f'''Missing Page and country values.
You can use this function with a dictionary with all the data preloaded as well'''
)
elif products_type == "<class 'dict'>":
cheapest_dict = {}
products_dictionary = products
for key in products_dictionary:
cheapest_dict[key] = products_dictionary[key][cheapest_idx]
else:
error_in_product_type = f"Products type must be a dict or a bs4.element.ResultSet with the boxes. Recived {type(products)}"
raise ValueError(error_in_product_type)
return cheapest_dict
def test_products_info_getters(self):
user_request = 'audifonos inalambricos'
country = 'mx'
amazon_url = Amazon.adapt_url(Amazon, user_request, country)
amazon_soup = extract_soup(amazon_url, 1, just_soup=True)
amazon_boxes = search_boxes(amazon_soup, Amazon.boxes)
amazon_names = len(get_names(amazon_boxes, Amazon.name_and_images))
amazon_images = len(get_images(amazon_boxes, Amazon))
amazon_urls = len(get_products_urls(amazon_boxes, Amazon))
amazon_price = len(get_price(country, amazon_boxes, Amazon.price))
amazon_reviews = len(get_reviews(country, amazon_boxes,
Amazon.reviews))
amazon_stars = len(get_stars(country, amazon_boxes, Amazon.stars))
trials = [
amazon_names, amazon_images, amazon_urls, amazon_price,
amazon_reviews, amazon_stars
]
for test in trials:
self.assertEqual(len(amazon_boxes), test)
def request_products(user_request, Page, header, home=False, country='mx'):
page_url = Page.adapt_url(Page, user_request, country)
# All the HTML of the page
page_soup, status = extract_soup(page_url, header)
# Wait until receive the info or been denied
if status == 503:
while status == 503:
time.sleep(1)
page_soup, status = extract_soup(page_url)
elif status == 200:
# HTML divided by products, and stored as elements of an array
page_boxes = search_boxes(page_soup, Page.boxes)
page_products = {}
# Obtain the info of the product
page_products['names'] = get_names(page_boxes, Page)
page_products['images'] = get_images(page_boxes, Page)
page_products['urls'] = get_products_urls(page_boxes, Page)
page_products['prices'] = get_price(country, page_boxes, Page, home)
page_products['status'] = status
return page_products
else:
page_products = {}
# With the empty values, not None, the script knows that this won't be
# uploaded. In case of one 'None', it thinks that there was a product box
# without info. Somethings that occurs in Amazon
page_products['store'] = Page.name
page_products['idx'] = Page.index
page_products['product'] = user_request
page_products['names'] = []
page_products['images'] = []
page_products['urls'] = []
page_products['prices'] = []
page_products['status'] = status
return page_products
if __name__ == "__main__":
user_request = 'audifonos inalambricos'
country = 'mx'
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)
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])
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])
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'])
cheapest_ml_product2 = get_cheapest(cheapest_idx, ml_products)
print(f'\n\nTest TWO:')
for key in cheapest_ml_product2:
print(key, ':', cheapest_ml_product2[key])
