Semantics of this feature in HTML/the Web

[littledan]

Note: The semantics of this proposal on the Web would be standardized in WHATWG/HTML, not in this repository. This issue is purely for requirements-gathering/brainstorming.

The README has examples of using the type: module attribute, but the semantics of this are not really defined anywhere. I think there's been some confusion in the discussion in issues as different people make assumptions about different semantics. Here's how I am currently picturing that this would work on the Web/in HTML:

• When JavaScript loads a module, either through an import statement or dynamic import(), it passes an optional additional parameter to the embedder, of the options bag of module attributes.
• When HTML gets these module attributes, it ignores unrecognized attributes (for reasons discussed in How should unrecognized attributes or types be handled? #21) and just looks at type:. If the type is present and unrecognized, module fetching and parsing is aborted with an error.
• Otherwise, the module is fetched, and the MIME type is compared with the type, and if no type is provided, then JavaScript is implicitly used. Each known type has a set of MIME types that it permits. If the MIME type is not contained in the type's set, then module fetching and parsing also errors out. If the module had already been fetched previously and lived in the cache, then the MIME type is cached along with it and used for this comparison.
• The module is then parsed and processed according to its MIME type (which may provide more detailed information than simply the type:). The type: is not used at this point--it simply provided an additional check before the parser was invoked.

Any thoughts on this logic?

[littledan]

Note, the above talks about how integration with type: would work, but I'd imagine the same logic would apply with out-of-band data, the type indicated as part of the module specifier, or with a single piece of additional data rather than key/value pairs.

[ljharb]

An unrecognized type should be ignored, as if none was provided, for the same reasons an unrecognized attribute is ignored. If the env can’t load the module, it will fail later.

[littledan]

@ljharb In this thread, we're talking about one specific environment--the Web. Are you saying that, if the type is not recognized, the Web should require that the MIME type be JavaScript? I don't understand the motivation for this behavior.

[ljharb]

I’m saying that if explicit type info is provided but unrecognized, it should fall back to the mime type. Obviously if the mime type is unrecognized it should act as it does now.

[littledan]

Based on the discussion at WICG/webcomponents#839 , my understanding is that Mozilla and WebKit don't find it acceptable to choose the module interpretation based on the MIME type, at least when choosing between JavaScript and JSON/CSS/HTML (though WebAssembly may provide a point of flexibility). Within this repository, I'd like to work within that assumption. I'd recommend following up on that W3C thread if you want to see if that constraint can be loosened.

[ljharb]

I think I’m not being clear :-)

I’m saying that an import with an inline unrecognized type should behave the same as an import without an inline unrecognized type. If that means the only type that can be imported without an inline type is javascript, that’s fine.

[littledan]

Let's centralize discussion of the handling of unrecognized types in #27 . I'd rather not repeat ourselves in several different threads!

[xtuc]

I think that on the web it's important to set the Accept header when fetching the resource. For example:

import u from "https://a.cf/sven.jpeg" as json

The developer is requesting from the server a file of type json. The host does that by sending an the corresponding Accept header.

[dandclark]

The module is then parsed and processed according to its MIME type (which may provide more detailed information than simply the type:). The type: is not used at this point--it simply provided an additional check before the parser was invoked.

If we want to allow overlapping MIME type sets, as discussed in #16 (image/foobar+json etc), then in this step it is the type attribute, rather than the MIME type, that is what must be used to determine what kind of module to instantiate, or else we're left with some ambiguity.

([the MIME type] may provide more detailed information than simply the type)

Could you elaborate on this a bit? For modules, I'm not sure what information could be gleaned from the MIME type other than which type of module we should instantiate.

[littledan]

Well, as @xtuc mentioned, if JS BinAst has a different MIME type, it may require different processing, even if it's also a JS module.

[jfparadis]

Is "script" the only tag that will needs this feature? In particular, does it make sense to loading images or templates (already supported via the template tag) using a script tag?

[littledan]

@jfparadis Could you say more about what you mean by that? I could imagine other sorts of attributes passed to other subresources (e.g., fetch options or integrity), and these could be passed through other tags (or out-of-band manifests).