Read the generated module as UTF-8 in three Python test drivers #8976

Task Summary

The Python drivers in MachineLearningScorerOpDescSpec, FilledAreaPlotOpDescSpec and TreePlotOpDescSpec read the generated module with open(sys.argv[1]), so Python decodes it with the locale's encoding, which is cp1252 on Windows. The spec writes the file as UTF-8. The modules these tests generate are ASCII today, so nothing fails yet, but a non-ASCII character in a generated module would make these tests fail on Windows only. The drivers that read Plotly HTML already pass encoding="utf-8"; these three should too.

Task Type

• Refactor / Cleanup
• DevOps / Deployment / CI
• Testing / QA
• Documentation
• Performance
• Other