fix: ls_remote.py docstring claimed a --format flag and JSON-default output that don't exist
Discovered while fixing musehub's 'muse plumbing <verb>' doc drift and checking muse ls-remote -h for real: the module docstring claimed 'Output (JSON, default)' and documented an '--format text' flag. Neither is true -- the real, only flags are --json/-j (boolean); default output is tab-delimited text, JSON only with --json. The 'unknown --format' exit code documentation was equally stale.
Also found the same fiction baked into tests/test_cmd_ls_remote.py: test_format_error_to_stderr and test_no_traceback_on_bad_format passed '--format xml'/'--format bad' and asserted nonzero exit + stderr output -- which is true, but only because argparse rejects '--format' as an unrecognized argument before the command's own logic ever runs. Deleted both -- there's no format-value validation left to test since --json is a boolean, not a choice flag. TestSecurity's two ANSI-stripping tests had the same bug: '--format text' made argparse reject the invocation before ever reaching the branch-printing code they claim to exercise, so they were passing without testing anything. Fixed by invoking with no flag at all (text is the real default) -- confirmed still passing against the actual sanitize_display() call path this time.
Tests: full tests/test_cmd_ls_remote.py + tests/test_ls_remote_supercharge.py sweep: 70 passed (72 - 2 deleted).
Co-Authored-By: Claude Sonnet 5 <[email protected]>
Files Changed
~2
1185 in snapshot
0 comments
muse hub commit comment sha256:8470fce45bed2f87c008e5dd4be8c21cdb39a57ad327511a182b5c4013f18b69 --body "your comment"
No comments yet. Be the first to start the discussion.