@Tye1980 - The answer is that it would scrape your wishlist off the site, and store it on a server. The ‘specials’ list is provided by DAZ in a convenient format for parsing, and once a day (or maybe twice) it would parse the list of specials, compare it to the list of things from your wishlist (and the list of things it’s already told you about), and let you know of any new changes.
The easiest way to get your wishlist into the system (i.e. I’ve already implemented it for myself) to let it log in as you, and the web app would scrape it for you. Another way, which I’ve toyed with but is a lot more complex, is to use a browser plugin for Chrome and FF (which I have to write anyhow) to add an ‘Import’ button on the Wishlist page, and it would use the browser’s own HTML parsing to scrape the Wishlist page and submit it to my web app.
@mrothello - I’ve actually started down the path of building this, as I realized there are a dozen useful features in there, many of which DAZ will probably not ever build. For example, ‘This item just went on sale for less than you paid for it <30 days ago.’ With appropriate caveats, because sometimes it matters that you paid for it at full price (e.g. $10 off $50 sales) of course.
The previous purchase thing would behave a bit like Amazon’s ‘Item Alert’ feature, where it’s not on the page when you load it, but it’d ping my server with the URL and if you’d imported your purchase history into my webapp, it could let you know if it’s in the history somewhere with a little warning message right by the ‘Add to Cart’ button a few seconds after page-load.
Other useful features would be downloading your purchase history as a CSV (which is why I started this process in the first place!) and if I can figure out how to automatically parse the bundle data someone else posted, automatically recognizing and splitting out bundle components. This might even make it more powerful than something DAZ can do, because it would recognize that you’d bought a component, even if you’d bought the bundle.
This would be hella useful for me, and I’ve only got around 500 items in my order history….
That said, the obvious drawback is that you’d have to trust me, and that’s hard, and many people won’t. That’s why I’m building it for myself first, and if other folks think it’s something they’d use, I’ll open it up. I’ve provided an HTML-scraping program to eBay users for about 12 years that isn’t terribly different than this, although it runs on your local computer.