Part 4 looks at what changes when the same technique has to run every day in production. Infinite scroll replaces the familiar Next button with a page that keeps growing as you move down it, which means your script has to decide when there is nothing left to load.
Step by step
Log every request with its URL, status code and duration. When something goes wrong at two in the morning, those logs are the difference between a quick fix and a long investigation.
Character encodings still cause trouble. Decode responses using the charset the server declares, and normalise text before comparing or storing it.
Keep configuration such as base URLs, delays and selectors out of your code. Changing a delay should not require a redeploy.
Putting it into practice
Read the site's terms of service before you start. Many sites welcome automated access within limits and say so explicitly; others ask you to use an API instead.
Deduplicate as you go. The same item often appears on several listing pages, and a stable identifier such as a SKU or ID makes duplicates easy to spot.
Store raw responses alongside parsed output while you are developing. When a parser breaks, you can replay the saved HTML instead of hitting the site again.
Key takeaway
Scroll, wait for the item count to change, and stop when two consecutive scrolls add nothing new. Better still, find the JSON request the page makes and call it directly.