Cleanup: bench fixtures, FsCache module split, deslop 12.6%, doc refresh (+ main merge)#295
Merged
Conversation
… 100% Basilisk is now listed in the official python/typing conformance results (python/typing#2316). Surface that across the site and READMEs, and re-stamp the graded commit to the current upstream tip. - Comparison tables (EN + zh homepage) and the /docs/comparison leaderboard now show every graded checker's official score, including pycroscope (92.2%), the 7th tool previously omitted. Basilisk tops the board as the only tool at a perfect 100%; each score links to that tool's live results folder. - READMEs (EN + zh): restore the "only Python type checker with a perfect 100%" claim now that it is substantiated by the official results.html and link it there (Documentation Honesty: comparative claim carries its source). - Re-stamp conformance references c94dfce -> f051625 (still 141/141, 0 FP) in conformance_report.json, README, and CHECKER-ARCHITECTURE-SPEC to match the fixtures the gate now scores against. - Expand vscode-extension/NOTICES to the full third-party notice set. pycroscope facts (Python, Apache-2.0, no LSP/debugger/editor extension) verified against its repository; all conformance figures verified against the live python/typing results.html at f051625.
# Conflicts: # README.md # README.zh.md # docs/specs/CHECKER-ARCHITECTURE-SPEC.md # vscode-extension/README.md # website/src/_data/conformance_report.json
The main merge (PR #294) brought code below three crates' coverage thresholds. Add behavioral tests to restore them; thresholds unchanged: - basilisk-parser: unit tests for depth::check_nesting (bracket/indent/ operator-chain limits + boundaries, dedent, bracket-resume, chain-breakers) — depth.rs 45% -> 100%. - basilisk-resolver: function-body TypedDict tests exercising the recursive check_td_stmts (del of required/NotRequired keys, reads, subscript-assign, disallowed mutator methods) — 94% -> 95%. - basilisk-cli: stub-subcommand tests (find_package_source success, cache_stub, run_stubs* paths) — 87% -> 91%. Authoritative scripts/test-rust.sh: all coverage thresholds pass, conformance 141/141 (100%, 0 false positives).
![github-code-quality[bot]](https://avatars.githubusercontent.com/u/9919?s=60&v=4){kind=small}
| type GoodC1 = Callable[[int], str] | ||
| type BadA1 = {"a": "b"} | ||
| type BadB1 = var1 | ||
| type BadC1 = int if 1 < 3 else str |
| type GoodC5 = Callable[[int], str] | ||
| type BadA5 = {"a": "b"} | ||
| type BadB5 = var5 | ||
| type BadC5 = int if 1 < 3 else str |
| type GoodC9 = Callable[[int], str] | ||
| type BadA9 = {"a": "b"} | ||
| type BadB9 = var9 | ||
| type BadC9 = int if 1 < 3 else str |
| type GoodC13 = Callable[[int], str] | ||
| type BadA13 = {"a": "b"} | ||
| type BadB13 = var13 | ||
| type BadC13 = int if 1 < 3 else str |
| type GoodC17 = Callable[[int], str] | ||
| type BadA17 = {"a": "b"} | ||
| type BadB17 = var17 | ||
| type BadC17 = int if 1 < 3 else str |
| type GoodC21 = Callable[[int], str] | ||
| type BadA21 = {"a": "b"} | ||
| type BadB21 = var21 | ||
| type BadC21 = int if 1 < 3 else str |
| type GoodC25 = Callable[[int], str] | ||
| type BadA25 = {"a": "b"} | ||
| type BadB25 = var25 | ||
| type BadC25 = int if 1 < 3 else str |
| type GoodC29 = Callable[[int], str] | ||
| type BadA29 = {"a": "b"} | ||
| type BadB29 = var29 | ||
| type BadC29 = int if 1 < 3 else str |
| type GoodC33 = Callable[[int], str] | ||
| type BadA33 = {"a": "b"} | ||
| type BadB33 = var33 | ||
| type BadC33 = int if 1 < 3 else str |
| type GoodC37 = Callable[[int], str] | ||
| type BadA37 = {"a": "b"} | ||
| type BadB37 = var37 | ||
| type BadC37 = int if 1 < 3 else str |
…e shell GitHub Actions runs steps with `bash --noprofile --norc`, which skips the profile scripts that add Homebrew to PATH, so `command -v brew` failed on ubuntu-24.04 (brew is installed at /home/linuxbrew) and the Lint job aborted before the duplication gate ran. Load brew's shellenv from its known install locations first. Keeps the unpinned-latest deslop so CI matches local measure.
MelbourneDeveloper
added a commit
that referenced
this pull request
Jul 8, 2026
## What The `main`→`cleanup` merge in #295 duplicated the `## 致谢` (Acknowledgments) block verbatim in the Chinese README, so it now renders **two identical Acknowledgments sections** back-to-back (shipped in v0.32.0). This removes the duplicate, restoring parity with the single `## Acknowledgments` in the English `README.md`. ## Why Surfaced by a post-merge doc-integrity audit of the v0.32.0 release. Cosmetic, non-blocking — a clean follow-up. No code, no metric, no sentinel change; the conformance/bench markers are untouched. ## Verification `grep -c '^## 致谢' README.zh.md` → `1` (was `2`); English README unchanged at `1`.
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.
TLDR
Consolidation of the
cleanupbranch — expands the benchmark-fixture suite, splitsFsCacheinto its own module (500-LOC cap), refreshes the READMEs/website, ratchets the deslop duplication gate to 12.6% and simplifies its installer — with currentmainmerged in and every per-crate coverage threshold restored.What Was Added?
benchmarks/fixtures/*.py(~24k lines) expanding the perf suite (aliases, callables, constructors, dataclasses, enums, generics defaults, literals, narrowing, overloads, protocols, returns, tuples).crates/basilisk-checker/src/imports/fs_cache.rs—FsCacheextracted into its own module soresolve.rsstays under the 500-LOC cap; its tests move with it.mainmerge dropped below threshold:basilisk-parser: unit tests fordepth::check_nesting(bracket/indent/operator-chain limits + boundaries, dedent, bracket-resume, chain-breakers).basilisk-resolver: function-bodyTypedDicttests exercising the recursivecheck_td_stmts(required/NotRequiredkey deletes, reads, subscript-assign, disallowed mutator methods).basilisk-cli: stub-subcommand tests (find_package_source,cache_stub,run_stubs*).examples/cpu_demo_loop.pydemo.What Was Changed or Deleted?
resolve.rsimportsFsCachefromfs_cache;apply.rsimports it fromsuper::fs_cache.generics_type_erasure,callables_kwargs,specialtypes_never_2,generics_basic_3/helpersprecision changes.debug_spawn.rsa_missing_interpreter_fails_fast_without_port_retriesuses 3 candidate ports so the documented free-port TOCTOU can't surfacePortTakenbefore the spawn (assertion unchanged)..deslop.tomlmax_duplication_percent→ 12.6 with non-product excludes;scripts/install-deslop.shreduced to an unpinnedbrew install nimblesite/tap/deslop.vscode-extension/README.md; rootREADME/README.zh; website conformance/benchmark data.main(PR Six bug fixes: hover inference, glyph tallies, safe-only Fix All, uv import isolation, Modules loading state, deep-expression stack overflow #294: hover inference, glyph tallies, safe-only Fix All, uv import isolation, Modules loading state, deep-expression stack-overflow guard).How Do The Automated Tests Prove It Works?
conformance/score.py --gate): 141/141 (100%), 0 false positives, 0 missed required errors againstpython/typing@main(f4f2952).scripts/test-rust.sh, thresholds incoverage-thresholds.jsonunchanged): all pass —basilisk-parser100%,basilisk-resolver95%,basilisk-cli91%,basilisk-checker94%, plus db/lsp/mojo/parser/plugin/stubs/config.cargo fmt --all --check,cargo clippy --workspace --all-targets -- -D warnings, and the deslop gate (repo-wide duplication under the 12.6% ceiling) all clean.debug_spawnsuite passes across repeated runs that previously flaked under port contention.gen_conformance_reference.py --checkandgen_rules_reference.py --datadrift checks clean; 24 Playwright navigation/screenshot e2e tests pass.Spec / Doc Changes
docs/specs/CHECKER-ARCHITECTURE-SPEC.mdconformance stamp updated to the graded commit.README.md,README.zh.md,vscode-extension/README.md) and website docs refreshed..deslop.tomlduplication budget documented (ratchets down only).Breaking Changes