The customer needed to write a crawler to collect ratings and reviews from two types of cooking pages.

• Cloudflare bypass protection has been implemented;
• The project uses the Selenium library as well as Undetected_chromedriver;
• The project is dockerized;
• Two types of site layout are crawled;
• Project execution time is 2 days;
• From each page is parsed: the total rating, the number of comments. The author's name, the date of the comment and the text of the comment are parsed from each comment;
• The result is collected in jsonl format (modified json format for line-by-line writing).

To run it is enough to run one command.

docker-compose up --build

The links used for parsing are located in the file main.py.