Repository navigation
exit codes: export-preset-axis uses 9, outside the 0 / 2+ convention #136
Description
What
templates/headless-batch-script-template/script.py (and its README) uses 0 for success and 2+ for distinct failures. Argparse usage also exits 2.
examples/export-preset-axis/ does not. --same-axis when orientations did not differ returns 9. A second path in the same function also returns 9 when orientations did not differ on the happy path. Neighboring failures in that file already use 7, 8, and 11.
Why it matters
Phase 3 recorded that there is no repo-wide exit-code table, and that exit 9 is example-local. If two more examples copy that numbering, the template convention is dead and every consumer has to read each script.
Proposal (not this issue's implementation)
Normalize export-preset-axis onto 0 / 2+ (or document an example-local table in that README and stop implying a shared scheme). Do not do that in a CI-workflow PR; it is a shipped example and needs its own falsification pass.
Out of scope
Do not change the smoke matrix, templates, or other examples in the same PR as this cleanup.
Activity
TMHSDigital commentedon Sep 12, 2026
Invalid premise: exit 9 is already within
2+. 26 example scripts use 9 for a sequential check;export-preset-axis --same-axisexiting 9 is the documented falsifier, not an outlier.No exit code will change. The real gap was that none of this was written down. That lands in #139 (
CONTRIBUTING.mdExit codes + smoke coverage). This issue closes on merge of that PR (Closes #136).Follow-up for a report-mode per-README check: #138.
docs: write down exit-code roles and the smoke coverage model (#139)