Fix CheckBoxTree rendering by aliasing react-checkbox-tree to its ESM bundle

[dpage]

Summary

• react-checkbox-tree v2 (bumped to ^2.0.1 in fix: Bump runtime and development dependencies #9870) marks the package as "type": "module" while still pointing its require exports condition at a UMD bundle. Babel transforms our import statements to require(), so webpack picks the UMD file and then treats it as ESM (because of "type": "module"). The UMD wrapper's module.exports = factory(...) never runs in that context and the default export is undefined — which is why dialogs that mount CheckBoxTree (Tools → Import/Export Servers and similar) fail with "Element type is invalid".
• Add a webpack resolve.alias for react-checkbox-tree pointing at the package's ESM bundle (lib/index.esm.js). That file is explicitly exposed by the package's own "./*": "./*" exports passthrough, has a real default export, and is what webpack would have picked if our build didn't down-transform imports to requires.
• The $ suffix makes it an exact-match alias, so any future subpath imports (e.g. CSS) continue to resolve normally.

Test plan

• yarn run linter — clean
• yarn run jest — all 140 suites / 824 tests pass
• yarn run bundle:dev and yarn run bundle both compile cleanly (only the pre-existing asset-size advisory warnings)
• Bundle inspection: before the fix, app.bundle.js contained node_modules/react-checkbox-tree/lib/index.js (the UMD wrapper); after the fix, only lib/index.esm.js is bundled and the webpackUniversalModuleDefinition wrapper is gone
• Manual smoke test: Tools → Import Servers and Tools → Export Servers render their dialogs without the "Element type is invalid" error

Closes #9972

Summary by CodeRabbit

• Chores
  • Updated webpack configuration to resolve compatibility issues with the react-checkbox-tree dependency by ensuring the correct module bundle is loaded during the build process.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 3d7a9277-8805-4587-af65-bbafcb43cf70

📥 Commits

Reviewing files that changed from the base of the PR and between 0d11dbc and ecd242a.

📒 Files selected for processing (1)

• web/webpack.shim.js

---

Walkthrough

A webpack alias configuration is added to web/webpack.shim.js that directs react-checkbox-tree module requests to the package's ESM entry file, bypassing resolution issues caused by the package's "type": "module" declaration and broken CommonJS main entry.

Changes

Webpack Alias for react-checkbox-tree

Layer / File(s)	Summary
Webpack react-checkbox-tree ESM alias web/webpack.shim.js	resolveAlias map adds react-checkbox-tree$ pointing to the package's ESM bundle (react-checkbox-tree/lib/index.esm.js) with inline comments explaining the workaround for the package's module type and entry point configuration.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~5 minutes

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and concisely summarizes the main change: fixing CheckBoxTree rendering through a webpack alias to react-checkbox-tree's ESM bundle.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[KijongHan]

Hey @dpage ! Thanks for looking into this one. It looks like we had the same idea actually, I opened a PR for this last week over in #9973.

Could you please take a look so we can avoid duplicating the review effort? Keen to get your feedback and your help in getting it across the line