Use the main script from the CLI. It expects a URL as the first param,
either via -u url
or --url url
or --url=url
, for example:
php stadium_goods.php https://www.stadiumgoods.com/air-jordan
php stadium_goods.php -u https://www.stadiumgoods.com/air-jordan
php stadium_goods.php --url https://www.stadiumgoods.com/air-jordan
php stadium_goods.php --url=https://www.stadiumgoods.com/air-jordan
It will download all the products (name and price) into a CSV (stadium_goods.csv), accounting for any pagination. Additionally, it will output the CSV to stdout.
Additionally, it is currently configured to bypass CloudFlare DDoS protection.
https://github.com/samacs/simple_html_dom https://github.com/KyranRana/cloudflare-bypass