-
-
Notifications
You must be signed in to change notification settings - Fork 7.5k
feat: add base option to import.meta.glob #20163
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
…ob-import.spec.ts
Co-authored-by: Ben McCann <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
amazing! thank you!
alias support would be a nice-to-have. I wouldn't block this PR on it, but it would be nice to add in some future release
Making a note for myself in case we come back to try to add the feature later, it seems one of the difficulties is that resolveId only works with files today whereas we need a directory for import.meta.glob: #18510 (comment)
sapphi-red
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good to me!
Co-authored-by: 翠 <[email protected]>
|
/ecosystem-ci run |
commit: |
|
📝 Ran ecosystem CI on
✅ nuxt, ladle, laravel, marko, qwik, rakkas, quasar, react-router, vite-plugin-vue, vite-plugin-pwa, unocss, vite-plugin-svelte, vuepress, vite-setup-catalogue, vite-environment-examples, waku, storybook, vite-plugin-react, vitest, vitepress |
## [1.0.0-beta.19] - 2025-06-22 ### 🚀 Features - support `OutputOptions#manualChunks` (#5037) by @hyf0 - advanced-chunks: support `advancedChunks#gruop#name` to be function (#5035) by @hyf0 - rolldown_plugin_import_glob: align with `vitejs/vite#20163` (#5034) by @shulaoda - rust/advanced-chunks: support `MatchGroup#name` to be dynamic (#5033) by @hyf0 - rolldown_plugin_build_import_analysis: align with `vitejs/vite#20117` (#5027) by @shulaoda - rolldown_plugin_build_import_analysis: align with `vitejs/vite#20115` (#5020) by @shulaoda - add validation warning for advanced chunks options without groups (#5009) by @sapphi-red ### 🐛 Bug Fixes - moduleInfo is not updated when entry module is emitted by this.emitFile (#5032) by @IWANABETHATGUY - preserveEntrySignatures: false generates circular imports that hangs with TLA (#5029) by @IWANABETHATGUY - rolldown_plugin_build_import_analysis: align pure dynamic import handling with rolldown-vite (#5016) by @shulaoda - plugin/vite-resolve: normalize leading slash (#5013) by @sapphi-red - debug: `build_id` doesn't increase (#5015) by @hyf0 - side effects in this.emitFile({ type: 'chunk' }) is removed when preserveEntrySignatures: false is set (#5012) by @IWANABETHATGUY ### 🚜 Refactor - rolldown_utils: simplify `block_on_spawn_all` (#5019) by @shulaoda - use `rolldown_utils::futures::block_on` for `WatcherImpl#start` (#5018) by @shulaoda ### 📚 Documentation - jsdoc: document `experimental.attachDebugInfo` (#5028) by @hyf0 - clarify that `advancedChunks` options are in bytes (#5022) by @sapphi-red - add a note that sequential conversion may break the code (#5024) by @sapphi-red ### ⚙️ Miscellaneous Tasks - infra: clean up `dist` before building `rolldown` (#5036) by @hyf0 - Align status notice in readme with documentation (#5021) by @rijkvanzanten ### ❤️ New Contributors * @rijkvanzanten made their first contribution in [#5021](#5021) Co-authored-by: shulaoda <[email protected]>
fixes #17453
Description
This is basically #18510 removing the partial base plus alias support and clarifying docs. Aliases can start with any char, not only
@. I couldn't find a good way to support them, and I think we could release the feature to unblock SvelteKit if they only need non-aliased bases. @dominikg @benmccann could you clarify if this is the case?Added these clarifications to the docs:
We don't support full resolution in globs already. If the path start with
.or/we don't resolve them for example, so it seems support for aliases and virtual modules is already partial without taking into account the base.Sending this as separate PR to avoid modifying the other PR that has a discussion about how to support alias and virtual modules, and we may want to continue that. There is also a proposal to refactor the code but IIUC, that isn't needed if we only need to support what the docs in this PR describe.