[ozenyunus35]

Part of #15254 (the mkdir -m 1,2 and mkdir --mo=1,2 rows).

The mode parser skipped empty comma-separated clauses and accepted a numeric mode without an operator anywhere in a list, so mkdir -m 1,2, mkdir -m 'u+x,' or chmod 644,u+x f succeeded. GNU coreutils 9.12 rejects all of these with "invalid mode":

mode	GNU	before
u+x, / ,u+x / u+r,,g+w / ,	invalid	accepted (chmod: invalid mode ())
644,u+x / u+x,644 / 1,2	invalid	accepted
=644,u+x / u+x,+644 / =,u+x / +,-	accepted	accepted

A new uucore::mode::check_clause checks each clause: it can't be empty, and a numeric mode without an operator has to be the whole mode. parse_chmod (used by mkdir, mkfifo, mknod and, through parse, install) and the clause loop in chmod both call it. For chmod, the message now names the whole mode, e.g. invalid mode (u+x,) instead of invalid mode ().

A few mode.rs unit tests asserted the old behavior (parse("644,u+x"), parse("u+w,755"), parse(",,") and parse("") succeeding). GNU rejects those, so they now use =644,... / ...,=755, and the empty cases moved to test_parse_empty_clauses, which expects an error.

Tests: test_mkdir_mode_rejects_empty_and_bare_numeric_clauses, test_chmod_rejects_empty_and_bare_numeric_clauses (both fail without the change) and test_chmod_numeric_mode_with_operator_in_a_list. The chmod, mkdir, mkfifo, mknod and install tests and the uucore mode unit tests pass.

Not changed here: GNU also rejects spaces around clauses (' u+x', 'u+rw, g+rw'), which parse_chmod still trims.

I used an AI assistant to help draft parts of the code and tests; I reviewed the changes and ran the tests myself.

[codspeed]

Merging this PR will degrade performance by 0.42%

⚠️ 119 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• cksum_default
• du_wide_tree[(5000, 500)]
• df_deep_directory
• cksum_multiple_files
• cksum_crc32b
• dd_copy_64k_blocks
• dd_copy_4k_blocks
• cksum_sysv
• sort_long_line[10000]
• du_all_wide_tree[(5000, 500)]
• …and 109 more

ℹ️ 1 benchmark has no flame graph

The flamegraph could not be generated because too much time was spent in runtime internals, making the visualization not useful for performance analysis.

• sort_dictionary_order[500000]

⚡ 1 improved benchmark
❌ 2 regressed benchmarks
✅ 398 untouched benchmarks
⏩ 70 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	factor_five_38_bit_primes	1.8 s	1.9 s	-8.17%
❌	Simulation	factor_thirteen_39_bit_primes	8.7 s	9 s	-3.63%
⚡	Simulation	factor_three_39_bit_primes	800.7 ms	717.6 ms	+11.58%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing ozenyunus35:mkdir-invalid-mode-message (cdd1a1a) with main (2fb77e8)}

Footnotes

1. 70 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. ↩

[ozenyunus35]

Closing this, as I can't follow up on it further. Thanks for the reviews.