It can't be using the Amazon affiliate API, so it has to be a scraper. I'd like it to configure it to be able to pull from either A) an Amazon category or sub-category page, B) an Amazon search url, or C) any Amazon goldbox deals page, with categories selected, etc. (All of these options should be do-able by just scraping off a single URL leading to each one, but please verify thats the case)
The scraper needs to grab the url, title, company, and description of each item, along with the image urls. It should create or update an XML RSS file, with each item listed with the Amazon URL, but the Amazon url should have a user-provided affiliate tag added.
There should be some configuration variables that set:
- The maximum number of items the RSS feed will have before old items get removed
- The affiliate tag to add to the URLs
- A proxy or list of proxies to use (this can be pulled from a text file)
- SMTP information to send an email to me in case an error occurs and the script stops running