Eventual-Inc/Daft - CodSpeed

Daft

Blog Docs Changelog

Performance History

Latest Results

refactor(functions): unwrap impossible-failure paths in UUIDv7 extraction Per review: the `None`-returns guarded "can't happen" cases, so make them panic loudly instead of silently producing nulls. - bucket_uuid7 now returns i64; the Month arm `.expect()`s the chrono conversion (a 48-bit ms timestamp is always representable). - extract_uuid7 maps over the null mask with `.map` (nulls still propagate) and `.expect()`s the &[u8] -> &[u8; 16] conversion, which FixedSizeBinary(16) guarantees. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

jay/uuidv7-partition-builtins

5 hours ago

test: verify strict-superset group-by over clustered source is correct Adds an execution-level test (not just an explain check) that a group-by whose keys are a strict superset of the source clustering takes the single-stage local aggregation path and still computes complete, correct groups. Guards the soundness of clustering_is_covered_by for single-input aggregation. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

jay/clustering-spec-datasource

5 hours ago

better docs for matching logic

jay/fix-clustering-spec-bound-translation

5 hours ago

use a typed BoundClusteringSpec

jay/fix-clustering-spec-bound-translation

6 hours ago

test: lock in clustering propagation soundness for non-injective projections Adds tests proving translate_clustering_spec matches by expression identity, not derivation, so a non-injective projection cannot fabricate a clustering guarantee: - Hash([a]) + (mod(a) AS b), a dropped => Unknown (not Hash([b])) - Hash([a]) + (a, mod(a) AS b) => Hash([a]) (follows the key, not the derivative) - Hash([mod(a)]) + (mod(a) AS b) => Hash([b]) (exact materialization is sound) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

jay/fix-clustering-spec-bound-translation

7 hours ago

test: verify strict-superset group-by over clustered source is correct Adds an execution-level test (not just an explain check) that a group-by whose keys are a strict superset of the source clustering takes the single-stage local aggregation path and still computes complete, correct groups. Guards the soundness of clustering_is_covered_by for single-input aggregation. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

jay/clustering-spec-datasource

8 hours ago

test+docs: clustering-aware shuffle elision over custom DataSource - Add tests/io/test_data_source_clustering.py exercising the distributed planner (ray runner): no-declaration baseline shuffles; exact-match and subset (refinement) for groupby/window/distinct skip the shuffle; the unsound inverse still shuffles; expression-valued clustering follows a projection; and the shuffle-free plan computes the correct result. - Document get_clustering_spec() in the custom connectors guide, with column and expression-valued examples and the distributed-only / soundness caveats. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

jay/clustering-spec-datasource

8 hours ago

Merge branch 'main' into patch-1

ARDA7787:patch-1

19 hours ago

Latest Branches

-1%

feat(functions): add UUIDv7 timestamp-extraction partition transforms#7032

5 hours ago

17d344a

jay/uuidv7-partition-builtins

0%

feat: allow custom datasources to specify clustering spec#7031

8 hours ago

42e948b

jay/clustering-spec-datasource

-10%

fix(planner): propagate clustering through bound projections#7033

6 hours ago

37b983c

jay/fix-clustering-spec-bound-translation

© 2026 CodSpeed Technology

Home Terms Privacy Docs