Is star in import syntax actually necessary for resolution?

[ncthbrt]

Since we're revising the import syntax to support both symbolic modules and files, I was wondering if we could look at the star syntax?

I've implemented module imports without needing the star, so I was wondering if there was some limitation of file imports that prevented us from omitting it?

[k2d222]

I was against the star import, but it was the only way to import many items scoped in a namespace with the current import proposal.

But if we suport importing modules directly, then I'm OK with removing the star.

[mighdoll]

Not sure if we should drop star imports, but one clarifying point that I missed in my first pass implementation is that the star imports are more like module imports than wildcard imports.

import ./foo/*; does not bring a fn bar inside foo.wgsl into scope. It only brings foo into scope, the call site needs to use foo.bar().

So star imports are module imports. If we cook up a different syntax in #25 to handle the additional needs of namespaces aka submodules in the import syntax, sure seems like star imports would be on the chopping block.

I'd suggest we use #25 to discuss the changes needed for addressing namespaces, and then come back to #35.