DataDog / datadog-lambda-js

The Datadog AWS Lambda Library for Node
Apache License 2.0
105 stars 35 forks source link

feat: Remove extension hello check as it's redundant #516

Closed astuyve closed 3 months ago

astuyve commented 4 months ago

What does this PR do?

Checking the file presence is enough, start order is enforced by the lambda runtime.

Motivation

Testing Guidelines

Additional Notes

Types of Changes

Check all that apply

codecov-commenter commented 4 months ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 81.95%. Comparing base (9c44eab) to head (859d234).

:exclamation: Current head 859d234 differs from pull request most recent head d892ff3. Consider uploading reports for the commit d892ff3 to get more accurate results

Additional details and impacted files ```diff @@ Coverage Diff @@ ## main #516 +/- ## ========================================== - Coverage 82.09% 81.95% -0.14% ========================================== Files 54 54 Lines 2212 2206 -6 Branches 513 512 -1 ========================================== - Hits 1816 1808 -8 - Misses 332 334 +2 Partials 64 64 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

purple4reina commented 3 months ago

So. I know the extension specifically uses the call to Hello to determine if there's a tracing layer present. See https://github.com/DataDog/datadog-agent/blob/11e8ea8b5455530ba8afa8ef1831d179c5523479/pkg/serverless/daemon/routes.go#L29

astuyve commented 3 months ago

We do, but the library never calls start-invocation, so I assume the call to enrichInferredSpan() never happens or happens but the result is never used. source.

We can see with this build, there are no duplicate spans. And we have made this change in Python already.

purple4reina commented 3 months ago

We do, but the library never calls start-invocation, so I assume the call to enrichInferredSpan() never happens or happens but the result is never used. source.

We can see with this build, there are no duplicate spans. And we have made this change in Python already.

Ah, yes I see now. Thanks for the explanation. We can probably just go ahead and delete that code from the extension then yeah?