Module naming and declarations

Sam Tobin-Hochstadt samth at ccs.neu.edu
Tue May 7 14:22:39 PDT 2013


On Tue, May 7, 2013 at 5:18 PM, Anne van Kesteren <annevk at annevk.nl> wrote:
> On Tue, May 7, 2013 at 2:15 PM, Sam Tobin-Hochstadt <samth at ccs.neu.edu> wrote:
>> On Tue, May 7, 2013 at 5:12 PM, Anne van Kesteren <annevk at annevk.nl> wrote:
>>> On Tue, May 7, 2013 at 2:00 PM, Sam Tobin-Hochstadt <samth at ccs.neu.edu> wrote:
>>>> 0. If we have an absolute URL, skip steps 1-3.
>>>
>>> How do you define this? We currently do not have this concept really.
>>> You could parse and compare if the serialization is equivalent I
>>> suppose, but even that would fail to detect certain cases, such as
>>> http://test/test\test ... Or http:test which is an absolute URL in
>>> theory, but treated identical to "test" if the base URL's scheme is
>>> http.
>>
>> If it's a URI, then it's treated this way. That is, if it starts
>> 'scheme:' for some scheme.
>
> I think what you're failing to understand is that there's no concept
> as "if it's a URI" within the context of the web platform, so I'm
> asking how you'd thought about going about defining that new concept.

I understood what you meant about the treatment of 'http:' URLs of
that form. I think I gave a reasonable definition for URI recognition
in the part you quote above; I could rewrite it as a JS program if
you'd like.

> Maybe seeing what I meant in action will help:
> http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%3Cimg%20src%3Dhttp%3Aimage%3E

This seems relevant for what will happen if you import from 'http:foo'
with the default semantics, but I don't see the relevance otherwise.

Sam


More information about the es-discuss mailing list