Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.55%. Comparing base (7697af6) to head (a63511f).
⚠️ Report is 636 commits behind head on master.

@@           Coverage Diff           @@
##           master    #1687   +/-   ##
=======================================
  Coverage   98.55%   98.55%           
=======================================
  Files         136      136           
  Lines        7691     7695    +4     
=======================================
+ Hits         7580     7584    +4     
  Misses        111      111           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@sl0thentr0py Yeah I agree with you. I think it depends on whether we expect users to initialize events themselves and call capture_event manually.

If we do, then I think the current approach (2 levels of checks) is necessary.
If we don't, we need to prohibit such usages, which should have its own PR.

Event.new should be considered as internal api, users shouldn't really use it directly. Only methods on hub or those exposed directly on the top-level Sentry module are our 'contract'.