…most likely NOT be how it's done in 1.5.0
…ady in Master and 1.4.2, so more like a hand-me-down)
…rizations on old API.
…isabled_ by passing oauth:false as creation arg in createLoginButton(), but at the moment the old, non-oauth way is not working.
…g an infinite loop in repositionZIndex.
…liberately disallowed but tests with it enabled don't show any problems, and with it enabled we can again have the click event.
… (instead of fbconnect://cancel), which matches Facebook's own setting per http://github.com/facebook/facebook-android-sdk/blob/ef53183c59dac75cda7b6787a7fcb6f4bcd35b7b/facebook/src/com/facebook/android/Facebook.java#L42
… This will definitely void the NPE from this ticket, but as we don't have a reproducible failcase we can't really see a before/after
…t application startup, in the default uncaught handler and the analytics event for exceptions
…tside of constructor.
…d do not report back as part of the root view's window hierarchy.
…hen not needed. Also, some leaks of movie were squished.
…ed - page load doesn't seem to work for subsequent pages past the first one (i.e., linked pages)
… POST-- now the activity accurately sets the result.
…e other views are in our new memory-saving scheme. Some native (JNI) code seems to freak out and crash the app if the webview isn't still around at a certain point in the GC process
…ified class names as keys to avoid collisions between Ti.Android and Ti.UI.Android
…e the value.
…tive control gets in a bad state if the images are changed dynamically. Will have to be addressed later to see where the problem lies.