Paddy/Betterzon
1
0
Fork 0
mirror of https://github.com/Mueller-Patrick/Betterzon.git synced 2024-12-23 12:15:11 +00:00
Code Issues Projects Releases Wiki Activity

Added another selector for price on amazon (does not work for books)

Browse Source
This commit is contained in:
henningxtro 2021-05-18 23:18:15 +02:00
parent 63cbac5490
commit 91a9a47f00
1 changed files with 10 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
Crawler/crawler.py
View File

@ -3,8 +3,8 @@ import requests
from bs4 import BeautifulSoup from bs4 import BeautifulSoup
HEADERS = ({'User-Agent': HEADERS = ({'User-Agent':
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.138 ' 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.138 '
'Safari/537.36'}) 'Safari/537.36'})
def crawl(product_ids: [int]) -> dict: def crawl(product_ids: [int]) -> dict:
@ -62,16 +62,21 @@ def crawl(product_ids: [int]) -> dict:
'products_with_problems': products_with_problems 'products_with_problems': products_with_problems
} }
def __crawl_amazon__(product_info: dict) -> tuple: def __crawl_amazon__(product_info: dict) -> tuple:
""" """
Crawls the price for the given product from amazon Crawls the price for the given product from amazon
:param product_info: A dict with product info containing product_id, vendor_id, url :param product_info: A dict with product info containing product_id, vendor_id, url
:return: A tuple with the crawled data, containing (product_id, vendor_id, price_in_cents) :return: A tuple with the crawled data, containing (product_id, vendor_id, price_in_cents)
""" """
page = requests.get(product_info['url'], headers= HEADERS) page = requests.get(product_info['url'], headers=HEADERS)
soup = BeautifulSoup(page.content, features="lxml") soup = BeautifulSoup(page.content, features="lxml")
try: try:
price = int(soup.find(id='priceblock_ourprice').get_text().replace(".", "").replace(",", "").replace("€", "").strip()) price = int(
soup.find(id='priceblock_ourprice').get_text().replace(".", "").replace(",", "").replace("€", "").strip())
if not price:
price = int(soup.find(id='price_inside_buybox').get_text().replace(".", "").replace(",", "").replace("€", "").strip())
except RuntimeError: except RuntimeError:
price = -1 price = -1
except AttributeError: except AttributeError:
@ -89,7 +94,7 @@ def __crawl_apple__(product_info: dict) -> tuple:
:param product_info: A dict with product info containing product_id, vendor_id, url :param product_info: A dict with product info containing product_id, vendor_id, url
:return: A tuple with the crawled data, containing (product_id, vendor_id, price_in_cents) :return: A tuple with the crawled data, containing (product_id, vendor_id, price_in_cents)
""" """
#return (product_info['product_id'], product_info['vendor_id'], 123) # return (product_info['product_id'], product_info['vendor_id'], 123)
pass pass