[mdatagen] Fix metric tests on boolean attributes

[jade-guiton-dd]

Description

Since commit 8f3ca8a, mdatagen improperly generates metric tests: boolean attributes are always expected to be true in the output, no matter what the injected test value was. For example, this line in collector-contrib is incorrectly changed to expect true instead of false after updating mdatagen and running it again.

The reason is this line, which attempts to generate assert.True or assert.False based on the expected test value. However, (attributeInfo $attr).TestValue does not return a boolean but the string true or false, both of which are considered "truthy" by the templating engine.

Testing

I manually rebuilt mdatagen, ran make generate in collector-contrib, and ran the tests, to confirm that the test related to the previously mentioned change was no longer failing.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.28%. Comparing base (5141578) to head (4d9f1d6).
Report is 6 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #11169      +/-   ##
==========================================
- Coverage   92.29%   92.28%   -0.01%     
==========================================
  Files         413      413              
  Lines       19772    19761      -11     
==========================================
- Hits        18248    18237      -11     
  Misses       1152     1152              
  Partials      372      372              

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

[codeboten]

Thanks for the fix @jade-guiton-dd, could you add a test with the value shown in the link you posted: https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/19b01db307d984e62f81f1880b63327ba092cd81/receiver/apachesparkreceiver/internal/metadata/generated_metrics_test.go#L1250

[jade-guiton-dd]

Thanks for the fix @jade-guiton-dd, could you add a test with the value shown in the link you posted

I added an attribute in the samplereceiver metrics which should trigger a test failure if this bug shows up again (as long as the computation of TestValues isn't changed). Is this what you had in mind @codeboten?

[codeboten]

Yup, thanks @jade-guiton-dd, looks great