Interpretation of "NaN"^^xsd:double and "-0"^^xsd:double

[ektrah]

Let's say I have a graph that consists of the triple ex:a ex:p "NaN"^^xsd:double . and xsd:double is in D.

Does this graph D-entail the graph ex:a ex:p "NaN"^^xsd:double .?

The answer depends on whether <I(ex:a), I("NaN"^^xsd:double)> is in IEXT(I(ex:p)),
which I assume depends on whether I("NaN"^^xsd:double) = I("NaN"^^xsd:double).

xmlschema11-2 says:

• Equality is identity, except that 0 = −0 (although they are not identical) and NaN ≠ NaN (although NaN is of course identical to itself).

  0 and −0 are thus equivalent for purposes of enumerations, identity constraints, and minimum and maximum values.

This made me think that I("NaN"^^xsd:double) ≠ I("NaN"^^xsd:double).

rdf-semantics says:

Throughout this document, the equality sign = indicates strict identity.

This is a bit bidden in the document, but makes it clear the answer is in fact yes, I("NaN"^^xsd:double) = I("NaN"^^xsd:double).

It would be useful to have a test for that.

Similarly for the question if ex:a ex:p "-0"^^xsd:double . D-entails ex:a ex:p "+0"^^xsd:double . (The answer is no.)

[pfps]

Yes, NaN double is identical to itself.

No, 0 and -0 are not identical.

The controlling wording is in XML Schema

The ·value space· of double contains the non-zero numbers m × 2^e , where m is an integer whose absolute value is less than 2^53, and e is an integer between −1074 and 971, inclusive. In addition to these values, the ·value space· of double also contains the following ·special values·: positiveZero, negativeZero, positiveInfinity, negativeInfinity, and notANumber.