Purpose of TryCatch?

[mayfield]

Hi, I'm trying to understand the purpose of the default integration TryCatch. The docs indicate the wrappers installed by this integration are required to capture errors of asynchronous events, but I'm looking for a little more context as to what is actually gained or lost with this integration enabled or disabled.

From my own testing, I couldn't see much difference with setTimeout related exceptions with this integration turned off as the window.onerror mechanism still works fine. The only thing I noticed is that the mechanism goes from instrument to onerror and handled from true to false. The latter of these changes seems a bit odd to me, because the setTimeout error was in fact never "handled."

I see that there are a few other wrappers for things like XHR, RAF and DOM event listeners but I believe the Window error Event will handle all of these in similar fashion.

Thanks in advance,
Justin

[mayfield]

@AbhiPrasad resolved with no comment?

[AbhiPrasad]

Hey @mayfield - just cleaning up our older discussions so was closing a ton of them. Will re-open!

Perhaps @kamilogorek can help give some more historical context why we needed the TryCatch integration.

We might need to set handled: false for these, but there might be a reason for that as well that I'm not familiar with.

[mayfield]

Okay thank you. I'm still curious to know what it provides as I'm actually disabling it currently.

[kamilogorek]

It was all about extracting correct function name for the event handler and as Abhi mentioned, about getting the information where the event originated from, as error thrown within event handler was very hard to trace back, back in the day.

It originates from raven-js, loong long time ago - https://github.com/getsentry/sentry-javascript/blob/3.x/packages/raven-js/test/integration/test.js#L616-L737