sort: report read failures with context

[arcusbuilds]

Fixes #13992.

sort /proc/self/mem reported the read failure as the bare io::Error string:

sort: Input/output error (os error 5)

That names neither the operation that failed nor anything the user can act on, and the (os error 5) suffix is noise. GNU prints sort: read failed: /proc/self/mem: Input/output error for the same input.

chunks.rs was calling e.to_string() directly. This wraps it with a read failed context and drops the errno suffix using the existing strip_errno helper, matching how the other SortError variants are rendered:

sort: read failed: Input/output error

It deliberately stops short of naming the file. read_to_buffer takes &mut T: Read, and ext_sort hands it an iterator of already-opened readers, so the path is gone by the time a read fails. Carrying it through would mean changing that iterator's item type and threading a PathBuf through chunks.rs and ext_sort/{mod,threaded,wasi}.rs. That is worth doing, but it is a much wider diff and I did not want to bundle the two. I am happy to follow up with it, or to fold it into this PR if you would rather have the whole fix at once.

The existing sort-cannot-read string is also worded cannot read: {$path}: {$error} rather than read failed:, so a path-carrying version would need either a reworded key or a new one. That is another reason to keep the change separate.

Tests: test_read_error_message in tests/by-util/test_sort.rs, Linux-only since it relies on /proc/self/mem returning EIO. It fails on main with the old message and passes here. The full test_sort suite is green (187 passed), and cargo fmt --check and cargo clippy -p uu_sort --all-targets -- -D warnings are both clean.

[Copilot]

Pull request overview

This PR improves sort’s user-facing error reporting for input read failures by adding a dedicated localized message that includes context (“read failed”) and strips Rust’s trailing "(os error N)" suffix, along with an integration test that locks in the new output.

Changes:

• Wrap read errors in chunks.rs with a read failed context message and format the underlying io::Error via strip_errno.
• Add a new Fluent localization key sort-read-failed for the new message.
• Add a Linux-only regression test using /proc/self/mem to verify the exact stderr output.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
tests/by-util/test_sort.rs	Adds a Linux-only regression test asserting the improved read-failure message formatting.
src/uu/sort/src/chunks.rs	Replaces raw io::Error::to_string() with a translated, contextualized error message using strip_errno.
src/uu/sort/locales/en-US.ftl	Introduces the new sort-read-failed localization string used by the new error path.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated no new comments.

[codspeed-hq]

Merging this PR will degrade performance by 3.72%

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

❌ 2 regressed benchmarks
✅ 355 untouched benchmarks
⏩ 50 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	cksum_crc	41.4 ms	43.1 ms	-3.86%
❌	Simulation	numfmt_stream_to_si_precision	346.4 ms	359.2 ms	-3.59%

Tip

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

---

_{Comparing arcusbuilds:fix/sort-read-error-message (f9dbbd3) with main (1a36740)}

Footnotes

1. 50 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. ↩

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/cut/bounded-memory (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/tail/inotify-dir-recreate (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/tail/tail-n0f (passes in this run but fails in the 'main' branch)