My next 2 goals are to list Avatar Awards, and Gamer Pictures. Avatar Awards should be pretty simple, the same as the achievements. In fact I think I already have the data on my harddrive, just need to parse it. Done.
Gamer Pictures is gonna be harder, as far as I know there is no manifest of published Gamer Pictures. So I'm gonna have to scrape them :(, there is prior art of this, it sucks. I can Improve a little, but not much.
Internally Gamer Pictures are 3 uint32s.
- Title ID (eg 0x415607d1)
- Large Image ID (eg 0x0002 8000)
- Small Image ID (eg 0x0001 8000)
Now, these Image IDs are uints, 4 bytes, 32 bits. But the upper 2 bytes, seem to be flags, for the size of the image. And in all my checks that's the only thing that changes between the different Image IDs. So it's redundant, and makes the common Image ID a uint16.
It's still not great, and at the moment I can't get much better. I'll be grabbing the smaller image, 32px x 32px instead of 64px x 64px. I can possibly use the non TLS server, to reduce handshakes, but only if I control the client. If I do this through a web browser it would have to be TLS, or I'd get mixed content errors for using non TLS data on a TLS site.
But that's it, I think instead of me scrapings 65535 of images, for thousands of titles I'm going to do it in your browser, at least until it's cached. It's not ideal, I don't want to put JavaScript on your system. It's probably not even the best technological solution, but it's the best I can do.