test(instrumentation-mongodb): Include all versions in coverage

[onurtemizkan]

Which problem is this PR solving?

• Test coverage of the MongoDB instrumentation is currently 49.74% - Link

This is because tests for MongoDB versions v3 and v4 are not included in the coverage report.

Short description of the changes

• Runs v3 and v4 tests in the test command to make sure they are included in the coverage report

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.41%. Comparing base (92106ff) to head (d71d793).
Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2681      +/-   ##
==========================================
+ Coverage   90.96%   92.41%   +1.45%     
==========================================
  Files         171      171              
  Lines        8133     8133              
  Branches     1649     1649              
==========================================
+ Hits         7398     7516     +118     
+ Misses        735      617     -118     

see 1 file with indirect coverage changes

[github-actions]

This package does not have an assigned component owner and is considered unmaintained. As such this package is in feature-freeze and this PR will be closed with 14 days unless a new owner or a sponsor (a member of @open-telemetry/javascript-approvers) for the feature is found. It is the responsibility of the author to find a sponsor for this feature.
Are you familiar with this package? Consider becoming a component owner.

[david-luna]

do you think is necessary to pin here the version? pacakge.json file already has the version fixed and if we want to update to a newer version we need to do it in both places.

[onurtemizkan]

I just double checked, for the pinned version, it's not needed. 👍

[david-luna]

I have my doubts about this PR. IMHO the wider the support range is the more tests with different versions we would need to do in order to aim for high coverage. We may end up doing a subset of test-all-versions 🤔

I'll bring up this into today's SIG

[onurtemizkan]

Thanks for the review @david-luna!

The alternative I thought was generating coverage on test-all-versions and merging reports in the end. But that seems a bit wider scale update if that's going to be applied to every package (there are just a few packages excluding specific versions as far as I can see).

[david-luna]

Thanks for the review @david-luna!

The alternative I thought was generating coverage on test-all-versions and merging reports in the end. But that seems a bit wider scale update if that's going to be applied to every package (there are just a few packages excluding specific versions as far as I can see).

@onurtemizkan we discussed this approach for coverage in the SIG. We accept the tradeoff of running 3 different tests to get the coverage report. Also

• locally you can get a different version of the package under test if npm run test fails in one of the steps.
• we will keep an eye on the CI to be sure is not impacting it significantly

Thanks for your contribution :) I'm going to approve, make sure your branch is up to date in order to be merged.

[david-luna]

@onurtemizkan sorry for the delay on this PR. I was checking if it's possible to have the coverage in TAV tests. Instead of stalling this PR I'll try it myself in another instrumentation.