Re: Deferred backtrace generation algorithm.
Hello Bob.
> So you effectively will have to manage a weakmap of next-stack-frame to exception objects.
Yes, that was exactly my first idea — to use a WeakMap from
exceptions. Now it seems that copying part of the frame information
will use less memory. But that's not certain :)
> Of note is also that $trace and $string are actually private properties.
Agree.
> If you're interested in providing a patch, you're definitely welcome.
Yes, I can give it a try. I'm just wondering if I'm missing something
important in the logic.
Thread (5 messages)