date: support GNU-compatible out-of-range years and formatting parity

[abhishekpradhan]

Summary

This addresses #9774: date -d rejects GNU-valid out-of-range years (for example 18978-01-01) because the normal parser/runtime path stops at the in-range datetime boundary.

• accept parse_datetime::ParsedDateTime::{InRange, Extended} in the date -d flow
• add an extended-year formatting path for out-of-range years with GNU-compatible output behavior
• preserve timezone conversion for extended dates, including -u, local TZ output, explicit input offsets, and far-future IANA rule lookups
• support timezone abbreviations on extended dates, including preserved non--u output and -u conversion for both fixed-offset and mapped IANA cases
• support additional format specifiers on extended dates, including %Q, %q, %r, %c, %U, %W, %V, %G, and %g
• for format behavior that already differs from GNU in current uutils date (notably %Q and E/O modifiers), preserve current uutils behavior rather than expanding scope in this PR
• handle extended parse results in touch -d without panic
• add regression tests for large-year output, DST-aware IANA rules, formatter parity, timezone abbreviations, and touch -d

Dependency

• temporarily pin parse_datetime to the upstream 0.15.0 git tag
• update deny.toml and fuzz/Cargo.lock to keep CI green with the temporary git dependency
• this should switch back to the crates.io release once parse_datetime 0.15.0 is published there

Related issues

• uutils/coreutils#9774: the user-facing date bug this PR fixes
• uutils/parse_datetime#160: the parser/runtime limitation that caused the large-year failure upstream
• uutils/coreutils#10958: broader E/O modifier support, intentionally left out of scope here

Reviewer guide

• src/uu/date/src/date.rs: parse branching, extended formatting, timezone/output conversion
• src/uu/touch/src/touch.rs: extended touch -d path
• tests/by-util/test_date.rs and tests/by-util/test_touch.rs: regression coverage

Testing

• cargo fmt --all -- --check
• cargo check -p uu_date -p uu_touch
• cargo test -p uu_date --lib -q
• cargo test -p uu_touch --lib -q
• cargo clippy -p uu_date --all-features --all-targets -- -D warnings
• cd fuzz && cargo fetch --locked
• cargo test -q --test tests test_date::test_large_year_with_explicit_offset_converts_to_output_timezone -- --nocapture
• cargo test -q --test tests test_date::test_large_year_supports_r_and_c_formats -- --nocapture
• cargo test -q --test tests test_date::test_large_year_iana_timezone_rule_uses_dst_in_far_future -- --nocapture
• cargo test -q --test tests test_date::test_large_year_week_format_specifiers -- --nocapture
• cargo test -q --test tests test_date::test_large_year_supports_existing_q_format_in_local_timezone -- --nocapture
• cargo test -q --test tests test_date::test_large_year_preserves_existing_eo_literal_behavior -- --nocapture
• cargo test -q --test tests test_date::test_large_year_timezone_abbreviations -- --nocapture
• cargo test -q --test tests test_date::test_large_year_timezone_abbreviations_match_in_range_behavior -- --nocapture
• cargo test -q --test tests test_touch::test_touch_large_year_date_does_not_panic -- --nocapture

[codspeed-hq]

Merging this PR will not alter performance

✅ 305 untouched benchmarks
⏩ 46 skipped benchmarks¹

---

_{Comparing abhishekpradhan:codex/large-year-date-parity (435d316) with main (e510449)}

Footnotes

1. 46 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. ↩

[sylvestre]

some conflicts

[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/tty/tty-eof (fails in this run but passes in the 'main' branch)
Note: The gnu test tests/printf/printf-surprise is now being skipped but was previously passing.

[abhishekpradhan]

This branch is now rebased onto current main, and the parse_datetime dependency now points to the large-year support from uutils/parse_datetime#263.

I also fixed the remaining extended-format regressions for large years:

• support for %Q, matching current uutils date behavior
• preserve E/O-modified format sequences literally, instead of reinterpreting them
• locale-aware month/day names
• timezone abbreviations on large years, including preserved non--u output and -u conversion

Local fmt, check, clippy, uu_date, and uu_touch tests are clean. This is still pinned to the parse_datetime 0.15.0 git tag for now, since that version has been tagged upstream but has not been published on crates.io yet.

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/tty/tty-eof (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/cut/cut-huge-range is now passing!
Congrats! The gnu test tests/unexpand/bounded-memory is now passing!

[abhishekpradhan]

This is now pinned to the upstream 0.15.0 git tag rather than the previous commit SHA. I’ll switch it over to the crates.io release once parse_datetime 0.15.0 is published there.

[github-actions]

GNU testsuite comparison:

Skipping an intermittent issue tests/pr/bounded-memory (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/tty/tty-eof (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/cut/bounded-memory is now being skipped but was previously passing.
Congrats! The gnu test tests/printf/printf-surprise is now passing!

[github-actions]

GNU testsuite comparison:

Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/printf/printf-surprise is now passing!

[sylvestre]

This is too large to be reviewed
please split it into several PR:

• bump parse_datetime
• one to support large date in date
• formatting parity
• touch work
  etc

Having a quick look here, it isn't ready - too much duplication (Extended vs ExtendedInZone branches), some various issues, etc

Given it seems that you use codex, you should prompt it to do better work in term of modification sizes, code duplication, etc

thanks

[abhishekpradhan]

Of course, thanks for the feedback here. I’ll split into several PR’s and work on the quality here.

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/tail/symlink (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/tail/retry is no longer failing!