aboutsummaryrefslogtreecommitdiffhomepage
path: root/crates/shirabe/tests/dependency_resolver/pool_builder_test.rs
AgeCommit message (Collapse)Author
2026-07-20test(ignore): document root causes for unannotated #[ignore] testsnsfisis
74 tests carried a bare #[ignore] with no explanation. Re-ran each: 25 now pass and had the attribute removed; the remaining 49 got a concise reason (todo!() stubs, regex-crate PCRE gaps, PhpMixed type mismatches, config bool-coercion bugs, missing skipped_load wiring in PoolBuilder, etc.) so future work can find and fix them by grep. No production code or test logic/assertions were changed. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-07-18perf(regex): eliminate per-call clone overhead in preg_* dispatchnsfisis
regex::Regex::clone() does not share the underlying meta engine's search-cache pool, so every fresh clone pays a ~10us warmup cost on its first use. Two changes together eliminate this across nearly all preg_* call sites: - A php_regex! macro resolves PHP-style patterns to a per-call-site &'static regex::Regex (via regex-macro's LazyLock), applied at the majority of call sites throughout the codebase. - Call sites still passing dynamic pattern strings go through PATTERN_CACHE, which now stores Arc<(Regex, bool)> and hands out Arc::clone()s instead of cloning the Regex itself. PregPattern::resolve() returns a ResolvedPattern enum (Arc or 'static reference) rather than an owned Regex, so neither path ever clones the Regex proper. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-07-11chore: use fully-qualified name for Rc/RefCellnsfisis
2026-06-28refactor: add linternsfisis
2026-06-27refactor: fix compiler warnings and clippy warningsnsfisis
2026-06-22test: port previously-ignored Composer tests via __ test hatchesnsfisis
Re-evaluate the reason'd #[ignore] tests under the Phase D criterion: a test is unportable ONLY if the APIs/types needed to WRITE it do not exist. A test that compiles but panics at runtime (todo!() body, a regex the regex crate cannot compile) or fails at runtime (incomplete or incorrect impl behavior) is portable -- it is written in full and marked with a reason-less #[ignore]. About 120 test functions move from reason'd #[ignore] to reason-less #[ignore] (the ported-but-not-yet-passing signal). Impl crates gain only additive __ test hatches (init_command, pool, file_downloader, package handle link setters, artifact/path repository, repository manager, svn); no existing logic changes. Tests whose required APIs genuinely do not exist (mock/reflection harness, ApplicationTester, solve() discarding SolverProblemsException, a script::Event that cannot be passed as an originating event) keep their reason'd #[ignore]. cargo check -p shirabe --tests passes. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-22test: port more test casesnsfisis
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-21test(dependency-resolver): port ↵nsfisis
DefaultPolicy/PoolBuilder/PoolOptimizer/Solver as stubs All ignored: they build Pools/run the Solver over packages whose constraint parsing uses a look-around regex the regex crate cannot compile, with fixture-heavy setup. setUp/tearDown not ported. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-21test: add empty test filesnsfisis