Benjamin Byholm
Are there threads involved somehow?
I tried this and did not see a crash. Could this problem be specific to Electron or is it exposing a deeper flaw? ```c++ #include #include NAN_METHOD(CalculateSync) { Nan::HandleScope scope;...
This is for sure a previously unknown issue and should be noted among the known issues. NAN, being a header-only library, has never promised a stable ABI. node-addon-api was created...
After closer inspection, it seems the problem stems from `env->current_async_id()` always returning `0`.
> Not sure if this is the root cause of the above assert but in nodejs/node#14040 there was a significant change in async hooks API but it was not applied...
Nothing concrete comes to mind. Maybe because of using `v8::Persistent`? Why don't you use `Nan::Persistent`? Does it work if you only use V8?
Actually, since it returns `0`, it seems that your `v8::Persistent` has been reset. This happens when it goes out of scope, since `v8::Persistent` has the `kResetInDestructor` flag set to `true`.
It ought to mean that, yes. What you would need to do instead is keep a reference alive on tge javascript side until it is no longer needed, then set...
The objectwrapped object will not be collected until its js correspondent has been collected. On March 27, 2016 5:43:59 PM GMT+03:00, Enerccio [email protected] wrote: > I thought that it is...
That would seem appropriate, given that it creates a new, separate context. However, there ought to be some way of creating a reference between contexts, but I'm not familiar with...