Repository navigation
[uplift] chore: Vitest benchmarks vs. last release and fast-deep-equal - #153
Merged
Merged
Conversation
- `pnpm bench`: `bench/compare.bench.mjs` (Vitest 5, `node (bench)` project) compares this tree, 3.2.2 (`react-fast-compare-published` alias) and fast-deep-equal per fixture group; candidates that throw on a group are skipped - `bench/summary-reporter.mjs` + report-only `bench` CI job: tables with % vs. published in the job summary - Remove `benchmark/`, `assets/benchmarking.png`, README chart, Google Sheet step; drop `benchmark`, `lodash`, `nano-equal`, `shallow-equal-fuzzy` - README, CONTRIBUTING (Benchmarks section, noise rule), PR template
|
Merged
ryan-roemer
added a commit
that referenced
this pull request
Nov 12, 2026
Final merge of the 2026 uplift (#140) from the integration branch `chore/2026-uplift`. Each change was reviewed in its own PR into that branch; this PR brings them to `master`. **No published change.** The package still ships the same `index.js` (CommonJS, ES5, `module.exports = isEqual`) and the same `index.d.ts` as 3.2.2. `pack:check` checks this in CI: same file list and manifest fields, identical `index.d.ts`, and `index.js` byte-identical to 3.2.2's once both are minified, so only formatting and comments differ. No changeset, so nothing to release. ## What changed **Package manager and release** (Phase 1) - #141: pnpm 12 (yarn 1 removed), Node 26 (`.nvmrc`), 7-day `minimumReleaseAge`; fast-deep-equal's test cases copied into `test/fixtures/` (git dependency dropped); Codecov and AppVeyor removed. - 65c9439: changesets v3; `release.yml` publishes through npm trusted publishing (OIDC) with provenance, using per-job permissions and the `npm-publish` environment; canary releases to `next` via `workflow_dispatch`. Tested with `3.2.3-canary-20261105232110`. npm tokens are now disallowed and the `NPM_TOKEN` secret is deleted. **Tooling** (Phase 2) - #142–#144: oxlint + oxfmt replace ESLint 8 (the format and lint-fix commits are listed in `.git-blame-ignore-revs`). - #145: reverts #130's unreleased type guard to 3.2.2's `boolean` signature. - #146: `pack:check` compares the packed tarball with 3.2.2 and runs publint, attw, and `require`/`import` in a clean install. - #147, #148: TypeScript 7 + JSDoc types on `index.js` (`types:check` fails if they drift from `index.d.ts`); type probe compiled against the packed tarball. - #149: `pnpm size` report in the CI job summary (replaces the hand-edited badge). - #154: `pnpm dep:check` (npm-check-updates, 7-day cooldown) and a manual update routine; knip audit. - 219a679: `AGENTS.md` + `CLAUDE.md` pointer. **Tests and CI** (Phase 3) - #151: Vitest 5 runs the specs in Node and in Chromium, Firefox and WebKit (browser mode). This replaces mocha, nyc, Karma, webpack and Babel. Tests use React 19 + Preact 11, and an extra CI job runs React 18.0.0 + Preact 10. v8 coverage goes in the CI job summary. - #152: zizmor audits the workflows; third-party actions are pinned to SHAs (pinact). **Benchmarks** (Phase 4) - #153: `pnpm bench` compares the working tree, the last release (3.2.2) and fast-deep-equal for each fixture group, in a report-only CI job. The old `benchmark/` setup and its dependencies are gone. **Docs** (Phase 5) - #155: README, CONTRIBUTING and PR template updated for the new tooling, with Nearform naming (the repo stays `FormidableLabs/react-fast-compare`). - b882172: `AGENTS.md` and CI now point at `master`. Scripts are now `check` (everything CI gates on), `format`, `test`, plus `bench`, `size`, `dep:check` and `changeset`. Closes #140.
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
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.
Phase 4 of #140: self-regression benchmarks.
pnpm bench(Vitest 5 benchmarking, Node) compares this tree, the last release (react-fast-compare-published@npm:react-fast-compare@3.2.2) andfast-deep-equal@3.1.3for each fixture group (generic,es6,react,custom).custom, theObject.create(null)cases).benchCI job: a small custom reporter writes per-group tables, with % vs.published, to the job summary. No threshold; the job fails only if the benchmark breaks.benchmark/index.js, thebenchmark/lodash/nano-equal/shallow-equal-fuzzydevDeps,assets/benchmarking.png, the README chart and the Google Sheet step.publishedafter a release, and treating differences under ~3–5% as noise); PR template item.Locally (Node 26), this tree matches 3.2.2 within noise. fast-deep-equal is ~25% faster on
es6, mostly because of our ES5 Map/Set loops; that waits for 4.0.No published change:
pack:checkpasses.