[naufalfx805-source]

Summary

• Implement last-flag-wins semantics for --check-order / --nocheck-order in join, matching GNU coreutils behavior
• Uses indices_of() to compare argument positions, following the existing pattern in ls, cp, and du
• Previously --check-order always won regardless of command-line order

Repro (from #16004)

printf 'b\na\n' > a.txt
join --check-order --nocheck-order a.txt a.txt
# GNU: prints b and a, exits 0
# uutils before fix: "join: a.txt:2: is not sorted: a" exit 1
# uutils after fix: prints b and a, exits 0

Test plan

• cargo test --test coreutils test_join::check_order_nocheck_order_last_wins
• Test covers both orderings: --check-order --nocheck-order (nocheck wins) and --nocheck-order --check-order (check wins)
• Existing join tests still pass

Fixes #16004

🤖 Generated with Claude Code

https://claude.ai/code/session_01Gc3pLfULRGQmfBnqr3oo3D

[sylvestre]

I think you can do it with clap

[codspeed]

Merging this PR will not alter performance

✅ 10 untouched benchmarks
⏩ 462 skipped benchmarks¹

---

_{Comparing naufalfx805-source:fix/join-check-order-last-wins (fcdb428) with main (ba501a7)²}

Footnotes

1. 462 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

2. No successful run was found on main (5dd33b8) during the generation of this report, so ba501a7 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[naufalfx805-source]

Good call — refactored to use overrides_with on both --check-order and --nocheck-order args. This lets clap handle the last-flag-wins semantics natively, so the parsing simplifies to plain get_flag() checks. Also fixed the rustfmt issue on the test file.

[cakebaker]

Duplicate of #16013

Anyway, thanks for your PR!