You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'd like to wait for a debugger to attach, but not pause directly after.
Interesting. Can you describe a situation in which this capability would be useful? To put it another way, why is AwaitDebuggerAndPause problematic in your case?
Would adding a flag and skipping the identified line be a reasonable way to implement this?
Well, we wouldn't want to make the API too clunky – e.g., AwaitDebuggerAndPause | DontActuallyPause 😅. Perhaps we could deprecate AwaitDebuggerAndPause and replace it with a pair of new flags.
why is AwaitDebuggerAndPause problematic in your case?
Two issues, first if I Thread.Sleep(5000) before the execution, the debugger attaches during the sleep and the execution never pauses. Second, if I put a breakpoint deep in my code, it pauses at the beginning and I have to hit continue to reach the next breakpoint—this is just really annoying.
replace it with a pair of new flags.
That would be my preference. Seems the logical solution.
@BinarySpike Hey Nathan small world its Todd, what you using ClearScript for?
@wysisoft ! Mostly looking at it for a hobby project. However, I'm looking at exposing a partial API of our application layer so that we can implement configurable business logic. We have a system that manages tons of files & folders and each tenant uses the system differently. Being able to deploy a tenant, then build them a script where they can enter properties, then hit a button and the script creates files, creates folders, and applies permissions based on their specific needs would be chefs kiss 😗👌
For performance/quality-of-life reasons, I'd like to wait for a debugger to attach, but not pause directly after.
See: https://github.com/microsoft/ClearScript/blob/1ba26876fc9dc1dd290fac657914cc1a8cf4c487/ClearScriptV8/V8IsolateImpl.cpp#L737C1-L737C25
Would adding a flag and skipping the identified line be a reasonable way to implement this?
The text was updated successfully, but these errors were encountered: