build(polars): migrate to crates.io 0.54.4, drop git rev pin#378
Merged
Conversation
Replace the pinned pola-rs/polars git rev (0.53.0) with a plain crates.io SemVer dependency on polars 0.54.4. Polars 0.54.x ships the nightly std::simd / core::unicode patches the git rev carried and accepts modern chrono (workspace resolves 0.4.45), so the git source is no longer needed. Renames the 0.54 streaming feature (was new_streaming). Regenerates the lockfile to a registry source (zero git sources remain). Unblocks crates.io publishability (release-automation-plan deviation row 5).
Two breaking changes in polars 0.54: - LazyFrame::with_new_streaming was renamed to with_streaming. - &ChunkedArray no longer implements IntoIterator; replace .into_iter() with .iter() (same Option<T> item type) across uffs-core and uffs-diag iteration sites. No behavioural change; verified by cargo nextest (1998 passed).
Polars is now a crates.io SemVer dependency, so there is no main HEAD to chase. Remove update_polars_git() + the STEP_UPDATE_POLARS ship step (scripts/ci-pipeline) and replace the 'just polars' git-HEAD bump recipe with a crates.io SemVer bump. Ship-pipeline step numbering is preserved, matching the existing 08/09 retirement convention.
Polars no longer comes from git (crates.io now), so the allow-git allowlist is empty. The file is also reflowed to taplo's 2-space style (it predated taplo enforcement, which is staged-scoped and so never caught it until this edit re-touched the file).
Add an §8.1 deviations-log row documenting that crates.io polars 0.54.4 removes the git-rev / chrono blocker, unblocking R5/R6/R8 publishability. Mark the stale 'keep update_polars_git()' R5 note as removed and update the dependency-policy rationale to reference the crates.io dependency.
…ration Vet-Reviewed-Diff: polars@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-arrow@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-async@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-buffer@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-compute@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-config@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-core@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-dtype@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-error@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-expr@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-io@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-json@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-lazy@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-mem-engine@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-ooc@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-ops@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-parquet@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-plan@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-row@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-schema@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-sql@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-stream@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-time@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4 Vet-Reviewed-Diff: polars-utils@0.53.0@git:1e9a63b95923291cd8849d70d96b8cec4e96da6a->0.54.4
githubrobbi
force-pushed
the
chore/polars-cratesio-migration
branch
from
34deaab to
206c134
Compare
This was referenced Jun 9, 2026
Merged
githubrobbi
added a commit
that referenced
this pull request
Jun 12, 2026
…tted dep bumps (#394) * docs(engine): refresh 09-performance with v0.5.120 benchmark data - Promote the all-7-drives full-scan export capture to its own table (23,322,046 rows -> CSV in 11.98 s, ~1.95 M rec/s; per-drive peaks 2.29 M rec/s), with the 4-drive cross-tool combined cell as the consistency check. - Mark the v0.5.62 aggregate full-scan row (13.5 s) superseded; note COLD/WARM + aggregation are unchanged since v0.5.62. - Add the v0.5.120 interactive-latency capture: targeted patterns 17-96 ms CLI e2e, spawn+pipe floor now 17-18 ms (was ~28 ms); daemon-side 0-3 ms figures remain v0.5.66 (not re-instrumented). - Tag the v0.5.4 bulk-retrieval tier sweep historical and point to the current ~1.95 M rows/s export numbers (~6x the old 323k peak). - Fix warm-restart bullet to the latest 5.7 s (v0.5.62) measurement. - Bump doc footer to v3.1 / 2026-06-11 / UFFS 0.5.120. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * docs(engine): deep v0.5.120 verification pass on 09 + 11 performance docs 09-performance.md: add v0.5.120 targeted-latency data to Known Optimization Targets (17-96 ms e2e, 17-18 ms spawn floor; top-N shape explicitly not re-measured). 11-performance-deep-dive.md: - Caching-levels table + HOT paragraphs: add v0.5.120 targeted CLI figures and the 17-18 ms spawn floor (was ~28 ms on v0.5.66). - 'Real-World Benchmarks' callout: lead with v0.5.120 headlines (30/30 vs Everything at 0.36x median; 23.3 M rows -> CSV in 11.98 s at ~1.95 M rec/s); tag aggregations/WARM/RSS as v0.5.62-latest. - Fix internal contradiction: the HOT top-N shape WAS re-run on v0.5.66 (table in this doc); v0.5.120 measures file-sink cross-tool cells instead. - Bulk-retrieval tier sweep tagged v0.5.4 historical with pointer to current ~1.95 M rows/s (S: 25.6 s -> 3.83 s). - Daemon-HOT vs C++ section: add the v0.5.120 re-measurement summary (5/5 full-scan 4.9x-13.9x, 6.6x combined, 180x-3447x targeted, combined regex DNF >120 s) pointing at the canonical report; keep the v0.5.66 sub-phase walkthrough. - Footer bumped to 2026-06-11 / v0.5.120. All figures cross-checked against raw/2026-06-v0.5.120_*.csv; all relative links verified to resolve. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * docs(manual): deep v0.5.120 verification pass on performance page - §7 Bulk Retrieval: add the v0.5.120 all-7-drives full-scan export table (23.3 M rows -> CSV in 11.98 s, ~1.95 M rows/s; S: drive 25.6 s -> 3.83 s, ~6x); tag the v0.5.4 tier sweep + pipe-vs-file tables historical. - §5a: fix internal contradiction — the floor note claimed ~28 ms under a v0.5.120 table whose small-result cells read 19-20 ms; the v0.5.120 floor is 17-18 ms (G-drive empty-result cells), ~28 ms was the v0.5.66 figure. - §2 diagram, §3 blockquote, §6 export-throughput note, §1 version row: refresh to v0.5.120 (17 ms CLI floor; 1.95 M rec/s export, 1.72 was v0.5.66). All figures cross-checked against raw/2026-06-v0.5.120_*.csv; all relative links verified to resolve. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * build(deps): bump regex 1.12.4, memchr 2.8.2, which 8.0.3 — fully vetted Dependency patch bumps with real cargo-vet diff audits (no exemption rubber-stamps): - which 8.0.2 -> 8.0.3: only code change is a cfg-gated fallback is_valid_executable returning Ok(false) on non-unix/windows/wasi/ redox targets; no unsafe, no new capabilities. - regex 1.12.3 -> 1.12.4: no source changes in the regex crate itself (changelog + metadata + regex-syntax requirement 0.8.5 -> 0.8.11). - regex-syntax 0.8.10 -> 0.8.11: IntervalSet::push rewritten to binary_search + local union + splice (upstream PR rust-lang/regex #1308); all safe Rust, bounds-checked, invariants preserved. - memchr 2.8.1 -> 2.8.2: already covered by existing audit imports. Also repairs cargo-vet policy drift surfaced by the check: - Drop stale audit-as-crates-io entries for the 23 polars-* crates (needed for the old git pin; polars is plain crates.io 0.54.4 since #378). - Add audit-as-crates-io = false for the 19 first-party uffs-* workspace crates, whose versions now match crates.io after the 0.5.120 publishes — we develop them here, so they must not be audited as registry packages. cargo vet check: clean (149 fully audited, 21 partially, 345 exempted). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
build(polars): migrate to crates.io 0.54.4, drop git rev pin
Replaces the pinned pola-rs/polars git rev with a plain crates.io SemVer dependency on polars 0.54.4.
Changes
Verification
Impact
Unblocks R5/R6/R8 release automation.