Hi! I am a java developer who would love to have this program support the Swedish Ebay tradera.com interface, and would like to join in on the project. Is this possible?
Please send me an e-mail if you are interested!

Cheers

Greetings,
So…I have way too much on my plate to be able to support it directly. There’s some interesting things you can learn from their site, though. For instance, a lot of the URL is ‘noise’, i.e. unnecessary.

and

are the same thing. The URL format (for auctions) appears to be {dash-separated-title}-auktion-{category id}-{auction id}.

However, when displaying the page it doesn’t use any of it except the auction id.

Normalizing URLs is one important key to determining how to parse the pages, so in the future you can just use the unique id of the page to load the item.

Beyond that, there’s a ton of stuff around what messages show up when you bid on an item and are in the lead, are below the existing hidden bid, tie the existing hidden bid, bid after it’s over, bid against yourself, underbid your existing bid, underbid the current bid, bid less than a bid increment.

You also have to get a feeling for the HTML layout of the page; what fields precede each of the data values you’re trying to extract, generally: current price, buy-it-now price (if applicable), high bidder, shipping, insurance, # of bids, title, end date & time, seller info, location, thumbnail…

It’s a lot less about Java programming than the grungy work of figuring out how the site behaves, what it shows, and how to identify it’s gotten into various states.

I hope this brain-dump helps!

— Morgan Schweers, CyberFOX!

OK I understand the problem here. However, what I ment by the tradera API was the following:
http://api.tradera.com/
which appears to allow real simple communication with the site without having to manually parse information from html-pages. Have you looked into this? What do you think?

Cheers

UPDATE: It appears that the buyer part of the API is available for test purposes only atm.

Greetings,
Right, and section 2.2b of the Terms of Use says that you can’t scrape for the functionality that isn’t available in their API, meaning you have to either scrape 100% and not use the API, or use the API and not snipe.

Hobson’s choice, essentially.

This is the same reason that JBidwatcher doesn’t use the eBay API.

— Morgan Schweers, CyberFOX!