Fix #13 by including the standard output as well. Previously we just included the standard error. Philosophically, you could argue that should be sufficient for a failed process. Anyway, as icing on the cake, we also change the outputs and errors from byte strings into UTF-8 character stirngs.
There's no way to test this without a massive test harness (which would include: GitHub Actions ecosystem, actions runners, virtual machines, public key infrastructure, OSSRH stub, PyPI stub, etc.).
📜 Summary
Fix #13 by including the standard output as well. Previously we just included the standard error. Philosophically, you could argue that should be sufficient for a failed process. Anyway, as icing on the cake, we also change the outputs and errors from byte strings into UTF-8 character stirngs.
🩺 Test Data and/or Report
There's no way to test this without a massive test harness (which would include: GitHub Actions ecosystem, actions runners, virtual machines, public key infrastructure, OSSRH stub, PyPI stub, etc.).
🧩 Related Issues
13