Skip to content

build: Update swc_core to v72.0.0#643

Merged
kdy1 merged 7 commits into
mainfrom
kdy1/bump
Jul 6, 2026
Merged

build: Update swc_core to v72.0.0#643
kdy1 merged 7 commits into
mainfrom
kdy1/bump

Conversation

@kdy1

@kdy1 kdy1 commented Jul 6, 2026

Copy link
Copy Markdown
Member

No description provided.

kdy1 added 5 commits July 6, 2026 18:25
Releases:
  @swc/plugin-react-remove-properties@12.15.0
  @swc/plugin-styled-components@12.15.0
  @swc/plugin-transform-imports@12.15.0
  @swc/plugin-swc-confidential@11.15.0
  @swc-contrib/mut-cjs-exports@14.15.0
  @swc/plugin-remove-console@12.15.0
  @swc/plugin-experimental-feature-flags@0.14.0
  @swc/plugin-styled-jsx@13.15.0
  @swc/plugin-swc-magic@12.15.0
  @swc/plugin-formatjs@9.15.0
  @swc/plugin-prefresh@12.15.0
  @swc/plugin-emotion@14.15.0
  @swc/plugin-swc-sdk@12.15.0
  @swc/plugin-relay@12.15.0
  @swc-contrib/plugin-graphql-codegen-client-preset@0.27.0
  @swc/plugin-jest@12.15.0
  @swc/plugin-loadable-components@11.15.0
  @swc/plugin-noop@12.15.0

[skip ci]
@changeset-bot

changeset-bot Bot commented Jul 6, 2026

Copy link
Copy Markdown

⚠️ No Changeset found

Latest commit: 648a55e

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

@socket-security

socket-security Bot commented Jul 6, 2026

Copy link
Copy Markdown

@claude

claude Bot commented Jul 6, 2026

Copy link
Copy Markdown

Review: build: Update swc_core to v72.0.0

Reviewed the full diff. This is a mechanical dependency bump + release PR (changeset-generated), not a functional change — so I focused on consistency rather than logic.

Summary of changes

  • Root Cargo.toml: swc_core 71.0.5 → 72.0.0, plus the related swc_ecma_* / swc_ecma_minifier majors bumped in lockstep.
  • Internal crate swc_icu_messageformat_parser 30.0.0 → 31.0.0; packages/formatjs/transform references it as 31.0.0 (matches ✅) and its own version 34.0.0 → 35.0.0.
  • Cargo.lock regenerated: net removal of transitive deps (darling, derive_builder, getset, proc-macro-error*, fnv, ident_case) — a nice cleanup that drops the deprecated proc-macro-error 1.0.4 chain.
  • 18 package version bumps with matching CHANGELOG/README entries.

Consistency checks (all pass)

  • Workspace version pins, per-package Cargo.toml, and package.json bumps are aligned.
  • The path+version dep in formatjs/transform/Cargo.toml matches the crate's new version.
  • No source (.rs), test, or config logic touched — no behavioral risk beyond what swc_core v72 itself introduces.

Notes / suggestions

  • Test coverage: No test changes accompany the bump, which is expected for a version-only PR. The real safety net here is CI — the swc_core v72 major bump could carry breaking API changes. At review time the Integration/Unit test jobs (ubuntu/macos/windows) are still pending; please confirm they go green before merging, since a green build is the only real validation for a transform-behavior-preserving bump.
  • Security: Socket Security checks passed; the dependency reduction is a positive.

LGTM pending CI. 👍

@claude

claude Bot commented Jul 6, 2026

Copy link
Copy Markdown

Review: build: Update swc_core to v72.0.0

Reviewed as a mechanical dependency bump. Overall this looks clean and correct. 👍

Summary

  • Workspace swc_core 71.0.5 → 72.0.0, with the aligned swc_ecma_* crates bumped (minifier 56→57, transforms 56→57, transforms_base 44→45, transforms_react 50→51, transforms_testing 48→49).
  • Internal Rust crate versions bumped 11.0.0 → 12.0.0 (and mut_cjs_exports 22→23, swc_icu_messageformat_parser 30→31, swc_formatjs_transform 34→35), npm packages bumped as minor (x.14.1 → x.15.0), CHANGELOGs and Cargo.lock regenerated.
  • Changeset bumpy-lights-press.md is correctly consumed/deleted on the branch (it only shows as churn in the merge diff), and the CHANGELOG/package.json edits match its minor bump list. ✅

What's good

  • Version consistency: Cargo.toml workspace refs, per-crate version =, the { path, version } internal references, and the lockfile all line up. No stragglers.
  • Nice cleanup side effect: swc_core v72 dropped its vergen dependency, which lets Cargo prune a whole subtree from Cargo.lockvergen, vergen-lib, derive_builder(_core/_macro), darling(_core/_macro), getset, proc-macro-error(_attr), fnv, ident_case. Fewer transitive deps and one less syn 1.0 consumer. Good outcome.

Things worth confirming before merge

  • Major bump with zero source changes. There are no .rs changes, so this relies entirely on swc_core v72 being source-compatible with how these plugins use it. The Check (swc_ast_unknown) job already passed, but the Integration Test and per-OS Test jobs were still pending at review time. Since a swc_core major can shift AST/transform behavior, I'd hold merge until the full integration/test matrix is green — the plugin .wasm snapshot/fixture tests are the real signal here, not just cargo check.
  • Pre-existing lockfile drift (informational). Several crates moved 10.0.0 → 12.0.0 in Cargo.lock while their Cargo.toml was already at 11.0.0 — i.e. the lockfile had drifted behind the manifests before this PR. This PR incidentally corrects it, so no action needed, but it suggests Cargo.lock wasn't regenerated on the previous bump.

Nits

  • The generated CHANGELOG entries are all the single line "build: Update swc_core to v72.0.0", which is fine for a bump but conveys nothing to downstream consumers about behavioral changes coming from swc_core v72. If v72 changes any user-visible transform output, a one-line pointer would help.

No blocking issues from a static review — approving in spirit pending green CI. 🚀

🤖 Automated review. I focused on version consistency, lockfile integrity, and changeset correctness; I did not build the full workspace locally.

@kdy1 kdy1 enabled auto-merge (squash) July 6, 2026 10:52
@claude

claude Bot commented Jul 6, 2026

Copy link
Copy Markdown

PR Review — build: Update swc_core to v72.0.0

Reviewed the full diff. This is a clean, mechanical dependency bump and looks good to merge. Summary of what I checked:

✅ Version consistency

  • swc_core 71.0.5 → 72.0.0 in root Cargo.toml, with the companion swc_ecma_* crates bumped in lockstep (minifier 56→57, transforms 56→57, transforms_base 44→45, transforms_react 50→51, transforms_testing 48→49).
  • Local crates get a major bump (11.0.0 → 12.0.0, formatjs 34→35), consistent with the Cargo.toml path-dep references and the regenerated Cargo.lock (on-disk lockfile matches — e.g. single swc_emotion @ 12.0.0; the apparent duplicate entries in gh pr diff are just a merge-diff artifact).
  • npm packages get a minor bump (x.14.1 → x.15.0) with matching changeset/CHANGELOG entries (changeset 4b382e7).

✅ Behavioral change is contained and verified

The only non-metadata source change is the snapshot update in styled-jsx/.../attribute-generation-modes/output.{swc,lightningcss}.js: the dynamic _JSXStyle.dynamic(...) selector IDs changed (e.g. 655fcf303bd006a4 → 7113a62c00624bef), while the static jsx-* hashes are unchanged. The new IDs are applied consistently across every usage within the file (call site + <_JSXStyle id=...>), and both output variants were regenerated identically, so correctness is preserved. Test - ubuntu-latest passing confirms this is the only fixture whose output shifted and that all other dynamic-selector fixtures still match.

Minor notes (non-blocking)

  • These dynamic IDs are internal runtime identifiers (not user-facing CSS class names), so a minor npm bump is reasonable — just flagging that consumers pinning on exact generated output would see a diff.
  • At review time a few checks were still pending (Test - windows-latest, Integration Test (macos-latest)); worth confirming they go green before merge given this touches every package.

Nothing else stood out — no correctness, performance, or security concerns. 👍

@kdy1 kdy1 merged commit 280f487 into main Jul 6, 2026
11 checks passed
@kdy1 kdy1 deleted the kdy1/bump branch July 6, 2026 11:13
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

1 participant