Continued API Rewrite

[amaitland]

Trying to reduce the code complexity a little bit.

More changes to come.

[rassilon]

I was doing just that and I'm debugging a MCefRefPtr related garbage hitting the finalizer or indeed not at all and CEF assert during CEF.Shutdown.

Bill

[amaitland]

Merge this PR so you can remove DisposableResource?

[rassilon]

Sure.
On Jun 11, 2015 11:05 PM, "Alex Maitland" [email protected] wrote:

Merge this PR so you can remove DisposableResource?

—
Reply to this email directly or view it on GitHub
#1072 (comment).

[amaitland]

Done. Still quite a few cleanups to go. Will look at creating another PR later. Need to review all the callbacks, make sure weren't not disposing the wrappers too quickly for them to be used an an async fashion.

[rassilon]

The callbacks don't clean themselves up. However, the fix for them looks
simple.
On Jun 11, 2015 11:20 PM, "Alex Maitland" [email protected] wrote:

Done. Still quite a few cleanups to go. Will look at creating another PR
later. Need to review all the callbacks, make sure weren't not disposing
the wrappers too quickly for them to be used an an async fashion.

—
Reply to this email directly or view it on GitHub
#1072 (comment).

[amaitland]

I was thinking more along the lines of when you pass in a CefFrameWrapper instance, making sure that it can be accessed from within a Task when using a callback. I'm not familiar with the %frameWrapper syntax enough to know exactly when the wrapper will be disposed. So I wanted to double check.

[amaitland]

I also changed the callbacks so after you execute they set the underlying MCefRefPtr instance to NULL. If you've got a cleaner way then I'm all for it 👍