fix: lexical-to-value mapping of rdf:HTML literals (#2483)

Use strict mode when parsing `rdf:HTML` literals. This ensures that when
[lexical-to-value
mapping](https://www.w3.org/TR/rdf11-concepts/#dfn-lexical-to-value-mapping)
(i.e. parsing) of a literal with `rdf:HTML` data type occurs, a value will
only be assigned if the lexical form is a valid HTML5 fragment.
Otherwise, i.e. for invalid fragments, no value will be associated with
the literal
[[ref](https://www.w3.org/TR/rdf11-concepts/#section-Graph-Literal)] and
the literal will be ill-typed.

---------

Co-authored-by: WhiteGobo <[email protected]>

Author: WhiteGobo

rdflib/term.py

@@ -1641,16 +1641,17 @@ def _parseXML(xmlstring: str) -> xml.dom.minidom.Document:  # noqa: N802
 def _parseHTML(htmltext: str) -> xml.dom.minidom.DocumentFragment:  # noqa: N802
     try:
         import html5lib
-
-        parser = html5lib.HTMLParser(tree=html5lib.treebuilders.getTreeBuilder("dom"))
-        retval = parser.parseFragment(htmltext)
-        retval.normalize()
-        return retval
     except ImportError:
         raise ImportError(
             "HTML5 parser not available. Try installing"
             + " html5lib <http://code.google.com/p/html5lib>"
         )
+    parser = html5lib.HTMLParser(
+        tree=html5lib.treebuilders.getTreeBuilder("dom"), strict=True
+    )
+    retval = parser.parseFragment(htmltext)
+    retval.normalize()
+    return retval
 
 
 def _writeXML(  # noqa: N802

test/test_literal/test_xmlliterals.py

@@ -100,6 +100,11 @@ def testHTML():
     assert l2.value is not None, "xml must have been parsed"
     assert l2.datatype == RDF.HTML, "literal must have right datatype"
 
+    l3 = Literal("<invalid", datatype=RDF.HTML)
+    assert l3.value is None, "invalid html must not be parsed"
+    assert l3.datatype == RDF.HTML, "literal must have right datatype"
+    assert str(l3) == "<invalid", "invalid html must not be normalized"
+
     assert l1 != l2
     assert not l1.eq(l2)