Discrepancy between Node v10.15 and v10.16

• Version: 5.0.1
• Platform: Mac OS Mojave 10.14.6 Node Versions: 10.15 and 10.16 mocha: v6.2.0 ts-mocha: v6.0.0

Greetings, I’ve recently started to write unit tests for one of our projects, using mocha, and I’ve noticed that the reports generated by c8 show some discrepancies, when switching between Node v10.15 and v10.16.

Using Node v10.15:

Using Node v10.16:

I noticed that the Uncovered lines section in particular shows very different results. In both cases I used c8@5.0.1.

I could probably live with the 0.05% difference in the statements and lines sections (for now, at least), but an almost 18% difference for branch is rather huge.

So the question is: How can I tell which one of these reports is more accurate?

Here are the full coverage logs: v10.15_coverage.txt v10.16_coverage.txt

Thanks in advance for any help!

[bcoe]

Hey @ElRaptorus, I’ve tested on newer versions of Node.js, and now that we support source-maps behavior seems correct:

Node 12+ and c8 6+

👇 the correct parts of the codebase are labeled as covered/uncovered.

Node <12 c8 < 6

👇 random parts of the codebase are highlighted, because of a lack of source-map information.

---

The missing source-map information would have been what was creating your highly variable coverage results.

Going forwards, I recommend doing something like we’ve done with yargs, where you only run coverage for the newest version of Node.js (I think for most folks, this is a sufficient window into the state of the project).

[bcoe]

@ElRaptorus thank you, I’ll take a look as soon as possible.