Correct some uses of 'URI' to 'URI reference' in structuring

[AdamWill]

While dealing with an issue in a set of schemas:
https://pagure.io/fedora-ci/messages/issue/67
I got rather confused by these docs until I went and read the
official spec drafts. In several places while discussing $id
and $ref, these docs claim these must be "URIs", but this is
not what the official draft says. It says they must be "URI-
references". From section 8.2, referring to $id:

"If present, the value for this keyword MUST be a string, and
MUST represent a valid URI-reference [RFC3986]."

From section 8.3, referring to $ref:

"The value of the "$ref" property MUST be a URI Reference."

This is a crucial and significant difference, as we can see in
RFC3986:

"A URI-reference is either a URI or a relative reference. If
the URI-reference's prefix does not match the syntax of a scheme
followed by its colon separator, then the URI-reference is a
relative reference."

That is, there is no such thing as a "relative URI" (since RFC
3986 came in, anyway, and clarified this as compared to earlier
RFCs). URIs are absolute by definition. If the doc says something
must be a 'URI' then the most obvious reading is that it must be
an absolute-URI as defined in RFC3986. If it can in fact be a
relative reference, we must make this clear.

Signed-off-by: Adam Williamson [email protected]

[awwright]

All of this seems to be an improvement.

We don't use URI-reference with the hyphen too much, but it's not wrong and might even be better than a space.

[AdamWill]

I used that form as it's what the RFC uses in defining the term. The JSON Schema draft is, as the quotes I provided show, inconsistent on what form it uses. :P

[AdamWill]

Ping?

[AdamWill]

ping again...

[Relequestual]

@AdamWill Sorry it's taken so long to approve and merge! Thanks for your contribution!

Feel free to highlight or fix any other issues. We're trying to focus on the test suite for 2019-09 currently, so expect a similar time frame to approve and merge.