Priority Level
Medium
Task Summary
The CLI correctly returns non-zero exit codes for most error paths (run, config, artifacts), but this behavior is largely untested. The only error-path exit code tests are in test_artifacts.py (added in the recent fix, PR #359 ) and one case in test_run.py (generate without --run-path).
We should add tests asserting non-zero exit codes for common failure modes to prevent regressions:
run --data-source missing.csv — nonexistent dataset filerun with no --data-source — missing required input for new runsconfig validate --config nonexistent.yaml — nonexistent config fileconfig validate --config malformed.yaml — invalid config (ValidationError)run generate --run-path /nonexistent — bad resume pathrun --data-source bad_url.txt — unsupported or unreachable data source
Each test should use CliRunner and assert result.exit_code != 0.
Technical Details & Implementation Plan
No response
Dependencies
No response
Priority Level
Medium
Task Summary
The CLI correctly returns non-zero exit codes for most error paths (
run,config,artifacts), but this behavior is largely untested. The only error-path exit code tests are intest_artifacts.py(added in the recent fix, PR #359 ) and one case intest_run.py(generatewithout--run-path).We should add tests asserting non-zero exit codes for common failure modes to prevent regressions:
run --data-source missing.csv— nonexistent dataset filerun with no --data-source— missing required input for new runsconfig validate --config nonexistent.yaml— nonexistent config fileconfig validate --config malformed.yaml— invalid config (ValidationError)run generate --run-path /nonexistent— bad resume pathrun --data-source bad_url.txt— unsupported or unreachable data sourceEach test should use
CliRunnerandassert result.exit_code != 0.Technical Details & Implementation Plan
No response
Dependencies
No response