Multiple correct outcomes for tests IRI-resolution-01 and IRI-resolution-02 (in Turtle and TriG)

[afs]

RFC 3986 section 5.2.2 allows for strict and non-strict resolution:

 -- A non-strict parser may ignore a scheme in the reference
 -- if it is identical to the base URI's scheme.

The tests

• rdf/rdf11/rdf-turtle/IRI-resolution-01.ttl
• rdf/rdf11/rdf-turtle/IRI-resolution-02.ttl

are affected because they have a test for resolving http:g against an http base.

There are two correct answers: if http: is ignored, the result is the same as resolving <g> (non-strict) otherwise it is <http:g> (strict - scheme already present).

    if ((not strict) and (R.scheme == Base.scheme)) then
         undefine(R.scheme);
         endif;
    if defined(R.scheme) then
         T.scheme    = R.scheme;
         T.authority = R.authority;
         T.path      = remove_dot_segments(R.path);
         T.query     = R.query;
    . . .

IRI-resolution-02 has:

@base <http://a/bb/ccc/d/>.
...
<urn:ex:s044> <urn:ex:p> <g>.
...
<urn:ex:s084> <urn:ex:p> <http:g>.

IRI-resolution-01 has a base of <http://a/bb/ccc/d;p?q>.

[afs]

The best way to fix this is to comment out the <http:g> lines in the two tests, leaving a comment in each TTL file.

Giving alternatives results files that an implementer has to sort out is more trouble than it's worth.