Merge pull request #467 from UUDigitalHumanitieslab/feature/reo-migra…

…tion

Feature/reo migration

backend/ontology/conftest.py

@@ -0,0 +1,114 @@
+import pytest
+from items import namespace as item
+from rdf.ns import DCTERMS, OA, RDF, RDFS, SCHEMA, SKOS, XSD
+from rdf.utils import graph_from_triples
+from rdflib import BNode, Literal
+from staff import namespace as staff
+from vocab import namespace as vocab
+
+from . import namespace as READIT
+
+BLESSINGTON = BNode()
+READ_FRENCH_POEMS = BNode()
+
+
+@pytest.fixture
+def linked_item_triples():
+    return (
+        graph_from_triples((
+            (item.subject1, RDF.type, READIT.reader),
+            (item.subject1, READIT.had_response, item.subject2),
+            (item.subject1, READIT.had_response, item.subject3),
+            (item.subject1, READIT.had_response, item.subject5),
+
+            (item.subject2, RDF.type, READIT.reading_response),
+            (item.subject2, RDFS.label, Literal("remember")),
+            (item.subject2, DCTERMS.type, Literal("example data")),
+
+            (item.subject3, RDF.type, READIT.reading_response),
+            (item.subject3, SKOS.prefLabel, Literal("Comment on book")),
+            (item.subject3, DCTERMS.creator, staff.dhdevelopers),
+            (item.subject3, READIT.involved, item.subject4),
+
+            (item.subject4, RDF.type, READIT.content),
+            (item.subject4, SKOS.prefLabel, Literal("Title of book")),
+            (item.subject4, DCTERMS.creator, staff.dhdevelopers),
+
+            (item.subject5, RDF.type, READIT.reading_response),
+            (item.subject5, RDFS.label, Literal("thought about")),
+        )),
+        graph_from_triples((
+            (item.subject1, RDF.type, READIT.reader),
+            (item.subject1, READIT.had_response, item.subject5),
+
+            (item.subject2, RDF.type, READIT.reading_response),
+            (item.subject2, RDFS.label, Literal("remember")),
+            (item.subject2, DCTERMS.type, Literal("example data")),
+
+            (item.subject3, RDF.type, READIT.reading_response),
+            (item.subject3, SKOS.prefLabel, Literal("Comment on book")),
+            (item.subject3, DCTERMS.creator, staff.dhdevelopers),
+
+            (item.subject4, RDF.type, READIT.content),
+            (item.subject4, SKOS.prefLabel, Literal("Title of book")),
+            (item.subject4, DCTERMS.creator, staff.dhdevelopers),
+
+            (item.subject5, RDF.type, READIT.reading_response),
+            (item.subject5, RDFS.label, Literal("thought about")),
+        ))
+    )
+
+
+@pytest.fixture
+def color_triples():
+    return (
+        graph_from_triples(
+            ((READIT.EXAMPLE, RDF.type, RDFS.Class),
+             (READIT.EXAMPLE_CHILD, RDF.type, RDFS.Class),
+                (READIT.EXAMPLE_CHILD, RDFS.subClassOf, READIT.EXAMPLE))
+        ),
+        graph_from_triples(
+            ((READIT.EXAMPLE, RDF.type, RDFS.Class),
+             (READIT.EXAMPLE, SCHEMA.color, Literal("#009e74")),
+                (READIT.EXAMPLE_CHILD, RDF.type, RDFS.Class),
+                (READIT.EXAMPLE_CHILD, RDFS.subClassOf, READIT.EXAMPLE),
+                (READIT.EXAMPLE_CHILD, SCHEMA.color, Literal("#009e74"))
+             )
+        )
+    )
+
+
+@pytest.fixture
+def object_triples():
+    return (
+        graph_from_triples((
+            (BLESSINGTON, RDF.type, READIT.reader),
+            (BLESSINGTON, READIT.carries_out, READ_FRENCH_POEMS),
+            (READ_FRENCH_POEMS, RDF.type, READIT.reading_session)
+        )),
+        graph_from_triples((
+            (BLESSINGTON, RDF.type, READIT.person),
+            (BLESSINGTON, READIT.carries_out, READ_FRENCH_POEMS),
+            (READ_FRENCH_POEMS, RDF.type, READIT.reading_session)
+        ))
+    )
+
+
+@pytest.fixture
+def annotation_triples():
+    return (
+        graph_from_triples((
+            (item.subject1, RDF.type, OA.Annotation),
+            (item.subject1, OA.hasBody, READIT.reader),
+            (item.subject2, RDF.type, OA.Annotation),
+            (item.subject2, OA.hasBody, READIT.otherClass)
+        )),
+        graph_from_triples((
+            (item.subject1, RDF.type, OA.Annotation),
+            (item.subject1, OA.hasBody, READIT.reader),
+            (item.subject1, vocab.needsVerification,
+             Literal("true", datatype=XSD.boolean)),
+            (item.subject2, RDF.type, OA.Annotation),
+            (item.subject2, OA.hasBody, READIT.otherClass)
+        ))
+    )

backend/ontology/constants.py

@@ -8,11 +8,16 @@
 # External file or URL from which we're loading the RDF-encoded ontology.
 # This is hardcoded, i.e., not a setting, because migrations depend
 # on the exact contents.
-SOURCE = op.join(settings.BASE_DIR, 'ontology', 'mock-ontology.jsonld')
+OLD_SOURCE = op.join(settings.BASE_DIR, 'ontology', 'mock-ontology.jsonld')
+SOURCE = 'https://raw.githubusercontent.com/eureadit/reading-experience-ontology/master/REO_2.4.1.owl'
 
 # Format must be a string that rdflib recognizes.
-SOURCE_FORMAT = 'json-ld'
+# SOURCE_FORMAT = 'json-ld'
+SOURCE_FORMAT = 'xml'
 
 # Namespace prefix that is used in the source. If different from
 # ONTOLOGY_NS, this needs to be replaced.
-SOURCE_PREFIX = 'http://readit.example/ontology/'
+OLD_SOURCE_PREFIX = 'http://readit.example/ontology/'
+SOURCE_PREFIX = 'http://dataforhistory.org/read-it-ongoing/'
+SOURCE_PROPERTY_PREFIX = SOURCE_PREFIX + 'property/'
+SOURCE_CLASS_PREFIX = SOURCE_PREFIX + 'class/'

backend/ontology/fixture.py

@@ -2,12 +2,16 @@
 This module represents what we believe *should* be in .graph.graph().
 """
 
-from rdflib import Graph, URIRef, Literal
+import re
 
+import requests
 from rdf.ns import *
+from rdflib import Graph, Literal, URIRef
 
 from . import namespace as my
-from .constants import SOURCE, SOURCE_FORMAT, SOURCE_PREFIX, ONTOLOGY_NS
+from .constants import (OLD_SOURCE, OLD_SOURCE_PREFIX, ONTOLOGY_NS, SOURCE,
+                        SOURCE_CLASS_PREFIX, SOURCE_FORMAT, SOURCE_PREFIX,
+                        SOURCE_PROPERTY_PREFIX)
 
 REPARSE_FORMAT = 'n3'
 
@@ -39,7 +43,20 @@ def canonical_graph():
     URIRefs.
     """
     g = Graph()
-    g.parse(SOURCE, format=SOURCE_FORMAT)
-    if SOURCE_PREFIX != ONTOLOGY_NS:
-        return replace_prefix(g, SOURCE_PREFIX, ONTOLOGY_NS)
+    g.parse(OLD_SOURCE, format='json-ld')
+    if OLD_SOURCE_PREFIX != ONTOLOGY_NS:
+        return replace_prefix(g, OLD_SOURCE_PREFIX, ONTOLOGY_NS)
     return g
+
+
+def reo_graph():
+    source = requests.get(SOURCE).text
+    to_replace = (SOURCE_PROPERTY_PREFIX, SOURCE_CLASS_PREFIX,
+                  SOURCE_PREFIX)
+    replaced_source = re.sub((r'|').join(to_replace), ONTOLOGY_NS, source)
+
+    g = Graph()
+    g.parse(data=replaced_source, format=SOURCE_FORMAT)
+    return g
+
+