astral-sh/ruff - CodSpeed

ruff

Blog Docs Changelog

Performance History

Latest Results

[ty] Projected narrowing node_constraints side table

codex/ty-projected-narrowing-node-constraints

2 minutes ago

[ty] Exact-runtime-class TypeIs negative-narrowing skip

codex/ty-exact-typeis-negative-narrowing-skip

2 minutes ago

[ty] Terminal/no-op narrowing fast paths

codex/ty-terminal-noop-narrowing-fastpaths

2 minutes ago

[ty] Cache upper-bound satisfiability

charlie/cache-upper-bound-satisfiability

4 minutes ago

[ty] Compact retained use-def bindings (#25682) ## Summary `UseDefMap` currently retains every interned binding state as a `SmallVec`, including builder-only capacity and an optional unbound constraint that are no longer needed after semantic indexing finishes. This change packs the shadowing policy into the scope-local definition ID and stores retained binding states as ranges into one contiguous `LiveBinding` array. Binding construction and interning behavior remain unchanged, while each `LiveBinding` shrinks from 16 to 12 bytes and retained states no longer carry per-vector allocation metadata. On a large codebase, this reduced retained memory by 2.96% without a wall-time regression.

main

9 minutes ago

[ty] Preserve Unknown expression presence

charlie/reduce-openai-memory-fresh

17 minutes ago

[ty] Index retained bindings by interned ID

charlie/reduce-retained-semantic-index-memory

20 minutes ago

[ty] Use same-file expression inference in reachability (#25694) ## Summary Reachability currently calls `infer_expression_type` for each expression that it evaluates. This creates and retains a separate Salsa query for every expression, even though reachability only evaluates predicates from the same file as the enclosing query. This uses `infer_same_file_expression_type` instead. The underlying `infer_expression_types` query continues to provide caching and cycle recovery, while cross-file callers continue through `infer_expression_type`; reachability avoids retaining the redundant single-expression query layer. Across the memory-report projects, retained `infer_expression_type_impl` memory fell by 88% to 95%, reducing total retained memory by 0.15% to 0.68% without diagnostic changes.

main

28 minutes ago

Latest Branches

+19%

[ty] Cache upper-bound satisfiability#25710

13 minutes ago

0f78a1b

charlie/cache-upper-bound-satisfiability

0%

[ty] Omit retained Unknown expression types#25709

25 minutes ago

38e18d3

charlie/reduce-openai-memory-fresh

+2%

[ty] Compact retained use-def bindings#25682

31 minutes ago

7148fb8

charlie/reduce-retained-semantic-index-memory

© 2026 CodSpeed Technology

Home Terms Privacy Docs