Closed carlosalberto closed 5 years ago
The operating mode this classes proposes is in contrast with the main philosophy of 0.32 where we deprecated autoFinish flag in activate (), because it too easily leads to programming mistakes of using try-with-resource on the scope and not having the right span to handle errors.
I think we should stick with one recommended instrumentation pattern, shown in 0.32 readme, instead of having the instrumentation be dependent on which scope manager is in use.
Also, the historical context is that we wanted to preserve this for people migrating from 0.30.
That being said, I will add this and @yurishkuro 's note and then merge.
Hey hey guys - if you are ok with this, I will merge later today, and issue the 0.32 release right after ;)
Let me know.
The last fix before releasing 0.32 ;)
@yurishkuro