the term identity comparison must be defined for triple terms

[lisp]

the sparql 1.2 document delegates to this one for a query processor's definition of term equality.
the sections in this document for iri and literals include the expected definitions.
that definition is missing from the section on triple terms (https://www.w3.org/TR/rdf12-concepts/#section-triple-terms).

[franconi]

Identity for IRIs and literals is defined from a pure syntactic point of view, e.g.:
IRI equality: Two IRIs are the same if and only if they consist of the same sequence of Unicode code points
I'd say that term equality between triple terms is defined by the notion of graph isomorphism recasted for triple terms.

[franconi]

...and the notion of graph equality (also based on graph isomorphism) should consider the SAME mapping M used for establishing triple term equality.

[pfps]

I'd say that term equality between triple terms is defined by the notion of graph isomorphism recasted for triple terms.

Except that it is easier, because there is no choice involved.

[pfps]

A triple term is a 3-tuple. Identity for tuples is derived from identity of their components, i.e., if the subject, predicate, and object of "two" triple terms are the same then "they" are the same. Of course, this only works this easily if a triple term cannot contain itself, but this is the case.

[hartig]

PR #161 has added a definition of triple equality and a sentence in the Triple Terms section saying:

Since triple terms are triples, equality of triple terms is the same as triple equality.

This should address the issue here.