On Mar 6, 2012, at 11:01 AM, Danny Yoo wrote: > On Tue, Mar 6, 2012 at 1:22 PM, John Clements <[email protected]> > wrote: >> >> On Mar 5, 2012, at 1:17 PM, Danny Yoo wrote: >> >>> Here's another example that's starting to work, though it isn't flashy yet. >>> >>> http://hashcollision.org/tmp/js-get-message/js-get-message-parent.html >> >> Can you be more explicit about what's cool here? I have a sense that it has >> to do with JS interacting with compiled racket, and I agree that that's very >> cool, but I could use a bit more hand-holding :). > > > Yes; of course. I'm sorry, I get so caught up sometimes! > > The example is demonstrating JS interacting with Racket, particularly > to interact between JS and World programs. I will be making the world > APIs in Whalesong extensible, so that the set of handlers like on-key, > on-tick, ... etc can be extended by library developers. The key point > is that I don't (and can't) know all event types out there in the > browser. With the appropriate FFI, I should allow library developers > to add in `on-tilt`, `on-touch`, or any of the other events that the > browser is providing, and without needing to touch the core of the > world implementation. > > This means I can (and probably should) factor out behavior that used > to be hardcoded in my world implementation (like on-key and on-mouse) > and push these into libraries. As a demonstration, I'm doing this in > the context of my web-world library, with an example that listens on > the `postMessage` event.
Very cool! I'm currently developing a tiny "dragon tamer" game with one of my three kids, and this is exactly the kind of thing we're looking for. John
smime.p7s
Description: S/MIME cryptographic signature
____________________ Racket Users list: http://lists.racket-lang.org/users
