Ryan C. Gordon
Ryan C. Gordon
And furnace is definitely trying ``, right?
This still doesn't reproduce for me, but it's possible https://github.com/libsdl-org/SDL/commit/4c035a1fd8d00a9d364d175f88af1fef0c6c5d66 will fix (or workaround) the problem.
Hmm...we _shouldn't_ be passing both on...preventing default handling should prevent the browser from generating a mouse event to match the touch event: https://github.com/libsdl-org/SDL/blob/9ed83e71f6cc445316e1b599aad98fa79019544d/src/video/emscripten/SDL_emscriptenevents.c#L523-L526 But I notice we're not preventing default...
My current thinking on this is that the event handlers are marked passive and emscripten_set_touchstart_callback and friends don't have an interface to specify this. Passive event handlers aren't allowed to...
Okay, I have a PR that fixes this, but it's way too large of a change for 3.2.18. This can land for 3.4.0, I think.
Go for it!
This is good feedback, and I agree with it. The README stuff probably just needs a little exposure from the front page. I'm not sure if I want to do...
I think AppEvent and AppIterate already serialize...? If it doesn't, it's a bug.
Also, if vsync==1 and the CPU usage is high, then a) the OS is busywaiting for the sync event or b) it didn't work at all and you're running frames...
I haven't tried this on Windows yet, but here's a test app: ```c #define SDL_MAIN_USE_CALLBACKS 1 #include #include // This is probably a deeply stupid way to do this. typedef...