feat: add logging on ChooChooTrain

[IgnacioBecerra]

Closes #

This PR adds logs that are sent out to the collector regarding the duration of the whole ChooChooTrain process. To get the information needed to be more specific in these logs, I had to use the getGitInfo() function that was called within the ibmTelemetry() function, meaning it would be called twice per instrumented package. While this probably wouldn't add a lot of time overhead, this whole logging exercise is mainly being done to troubleshoot packages not reporting data, and since our current working theory is that the process is taking longer than it should, I wouldn't want to add anything that would make it take even longer.

Thus, the part of the code gathering all this git information was refactored to only be ran once at the beginning of the whole ChooChooTrain. It was also properly hashed in order to be passed down to the ibmTelemetry() function, where all needed to be done is add it to the otelContext.

Changelog

New

• adds function to send logs to collector

Changed

• refactored way git information is retrieved to be ran only once

Testing / reviewing

{{ Add descriptions, steps or a checklist for how reviewers can verify this PR works or not }}

[mattrosno]

@IgnacioBecerra to make sure I understand, if 20 instrumented packages run during a scan, currently, these commands would run 20 times?

• git remote get-url origin
• git rev-parse HEAD
• git branch --points-at=${commitHash} --format='%(refname:short)'
• git tag --points-at=${commitHash} --format='%(refname:short)'

With this refactor, these commands would only run once? And the hypothesis is this change would speed things up?

[IgnacioBecerra]

@mattrosno Yeah so currently every instance of the ChooChooTrain calls its own ibmTelemetry() instance, which runs those commands. If 20 instrumented packages are running, then those commands get ran 20 times.

Since we need identifiers for our logs within the ChooChooTrain, we're gonna need to run those commands before calling ibmTelemetry(), so at the moment (in this draft PR) we're calling it twice, so they're getting ran 40 times. My goal is to at least get those commands out of ibmTelemetry and only be ran in the ChooChooTrain.

I'm not totally sure if it'll be indeed possible to run it only once, but at least for the time being this slight refactor is mostly to avoid running those commands in two different functions each. The commands run very fast, so this isn't the bottleneck we think it might be, but it's always good to squeeze out any extra speed if possible, or at least to avoid repetitive code.

[mattrosno]

We need to remove new lines in the template literals so the new line characters and spaces (tabs characters?) aren't included in the logging.

[IgnacioBecerra]

@mattrosno I'm now using concatenation instead since we have an eslint rule that a single line can't be over 150 characters.

I'll be working on getting the SonarCloud number up.

[ariellalgilmore]

LGTM just one small thing commented

[sonarqubecloud]

Quality Gate passed

Issues
2 New issues
0 Accepted issues

Measures
0 Security Hotspots
88.6% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud