Ethan Wu
Ethan Wu
I agree. Since things won't be unexcepted when calling the module with `\u2028`, etc., I think simply closing the pr is ok. But I don't know how the ES2016 environment...
ESLint when env is ES2017 doesn't report errors. I will close it first due to it is unnecessary.
It shouldn't be a network error, as the log isn't reported after start.
I've implemented a small feature [here](https://github.com/7086cmd/rolldown/tree/feat/http-resolve/crates/rolldown_plugin_http_resolve) (as a native plugin), which uses `oxc` to transform remote file imports into URLs. While native support might be overkill, we could alternatively develop...
> If it is going to become a built-in feature of Rolldown, caching (and cleaning) of remote files may also need to be considered to speed up repeated builds during...
So strange. Local code are formatted, but online repo reported errors.
@brc-dd It works. Thanks!
It appears that the test262 breakage is caused by the `void x` declaration. However, surprisingly, this pass does not address the `void x` issue at all!
It appears that the `hoistTransitiveImports` option is currently set to false by default in Rolldown’s behavior (related: rollup/rollup#3332). Is it necessary to support the option being set to `true`?
It should be a error throwed by Rollup. I have improved the report message in Rolldown btw.