Updated dataset

pyproject.toml

@@ -110,8 +110,8 @@ good-names = [
     "s", "p", "o",
     # Namespaces
     "EX",
-    # dict, file, ...
-    "d", "f",
+    # dict, value, file, ...
+    "d", "v", "f",
 ]
 
 [tool.pytest.ini_options]

tests/dataset/test_dataset.py

@@ -12,78 +12,72 @@
 inputdir = testdir / "input"
 
 
-if True:
-    # def test_save_and_load_dataset():
-    """Test save_dataset() and load_dataset()."""
-    # pylint: disable=too-many-locals,invalid-name
+def test_get_context():
+    """Test get_context()."""
+    from tripper.dataset import get_context
 
-    from tripper import DCAT, RDF, Triplestore
-    from tripper.dataset import load_datadoc, load_dataset, save_dataset
+    context = get_context()
+    assert isinstance(context, dict)
+    assert "@version" in context
+    assert len(context) > 20
 
-    ts = Triplestore("rdflib")
 
-    # Load data documentation
-    # datadoc = load_datadoc(inputdir / "datasets.yaml")
-    datadoc = load_datadoc(inputdir / "semdata.yaml")
-    assert isinstance(datadoc, dict)
-    assert "@context" in datadoc
+def test_get_prefixes():
+    """Test get_prefixes()."""
+    from tripper.dataset import get_prefixes
 
-    prefixes = datadoc["prefixes"]
-    for dataset in datadoc["datasets"]:
-        ds = save_dataset(ts, dataset, prefixes=prefixes)
-        repr1 = set(ts.triples())
+    prefixes = get_prefixes()
+    assert prefixes["dcat"] == "http://www.w3.org/ns/dcat#"
+    assert prefixes["emmo"] == "https://w3id.org/emmo#"
 
-    # Load back dict representation from the triplestore
-    SEMDATA = ts.namespaces["semdata"]
-    d = load_dataset(ts, iri=SEMDATA["sample3/pos1_01_grid_200x"])
 
-    # Should the prefix be expanded?
-    assert ds["@id"] == "semdata:sample3/pos1_01_grid_200x"
+def test_get_shortnames():
+    """Test get_shortnames()."""
+    from tripper.dataset import get_shortnames
+
+    # Short names that are not equal to the last component of the IRI
+    exceptions = (
+        "datasets",
+        "datamodel",
+        "prefixes",
+        "configuration",
+        "statements",
+        "@type",
+    )
 
+    shortnames = get_shortnames()
     assert (
-        "semdata:sample3/pos1_01_grid_200x",
-        RDF.type,
-        DCAT.Dataset,
-    ) in repr1
-    assert d["@id"] == (
-        "http://sintef.no/data/matchmaker/SEM/sample3/pos1_01_grid_200x"
+        shortnames["https://w3id.org/emmo/domain/oteio#hasDataSet"]
+        == "datasets"
     )
-    # # Store the new dict representation to another triplestore
-    # ts2 = Triplestore("rdflib")
-    # ds2 = save_dataset(ts2, d)
-    # repr2 = set(ts.triples())
-    #
-    # # Ensure that both dict and triplestore representations are equal
-    # assert ds2 == ds
-    # assert repr2 == repr1
-    #
-    # # Load dataset using SPARQL
-    # dd = load_dataset_sparql(ts, iri=EX.mydata)
-    # assert dd == d
+
+    for k, v in shortnames.items():
+        if v not in exceptions:
+            assert k.rsplit("#", 1)[-1].rsplit("/", 1)[-1] == v
 
 
 # if True:
-def test_datadoc():
-    """Test storing data documentation to triplestore."""
-    # pylint: disable=unused-variable
-    import io
-    import json
+def test_save_and_load_dataset():
+    """Test save_dataset() and load_dataset()."""
+    # pylint: disable=too-many-locals,invalid-name
 
     from tripper import Triplestore
-    from tripper.dataset import load_datadoc
+    from tripper.dataset import load_dataset, save_datadoc
 
     ts = Triplestore("rdflib")
-    d = load_datadoc(inputdir / "datasets.yaml")
 
-    f = io.StringIO(json.dumps(d))
-    ts2 = Triplestore(backend="rdflib")
-    ts2.parse(f, format="json-ld")
+    # Load data documentation into triplestore
+    datadoc = save_datadoc(ts, inputdir / "semdata.yaml")
+    assert isinstance(datadoc, dict)
+    assert "@context" in datadoc
 
-    # Add triples from temporary triplestore
-    ts.add_triples(ts2.triples())
-    # ts2.close()  # explicit close ts2
+    # Load back dict representation from the triplestore
+    SEMDATA = ts.namespaces["semdata"]
+    iri = SEMDATA["SEM_cement_batch2/77600-23-001/77600-23-001_5kV_400x_m001"]
+    d = load_dataset(ts, iri)
 
-    print(ts.serialize())
+    # Should the prefix be expanded?
+    assert d["@id"] == iri
 
 
 def test_fuseki():

tripper/testutils.py → tests/paths.py

@@ -1,12 +1,11 @@
-"""Motule primarly intended to be imported by tests.
+"""Defines paths for tests.
 
 It defines some directories and some utility functions that can be used
 with or without conftest.
 """
 
 from pathlib import Path
 
-rootdir = Path(__file__).resolve().parent.parent
-testdir = rootdir / "tests"
+testdir = Path(__file__).resolve().parent
 ontodir = testdir / "ontologies"
 outdir = testdir / "output"

tests/test_namespace.py

@@ -14,9 +14,10 @@
 def test_namespaces() -> None:
     """Test namespaces."""
     pytest.importorskip("rdflib")
+    from paths import ontodir
+
     from tripper import RDF, Namespace
     from tripper.errors import NoSuchIRIError
-    from tripper.testutils import ontodir
 
     assert str(RDF) == "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
     assert RDF.type == "http://www.w3.org/1999/02/22-rdf-syntax-ns#type"
@@ -72,9 +73,10 @@ def test_namespaces() -> None:
 def test_triplestore_arg() -> None:
     """Test triplestore argument of Namespace.__init__()."""
     pytest.importorskip("rdflib")
+    from paths import ontodir
+
     from tripper import RDF, Namespace, Triplestore
     from tripper.errors import NamespaceError, NoSuchIRIError
-    from tripper.testutils import ontodir
 
     assert str(RDF) == "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
     assert RDF.type == "http://www.w3.org/1999/02/22-rdf-syntax-ns#type"

tests/test_triplestore.py

@@ -463,8 +463,9 @@ def test_find_literal_triples() -> None:
     """Test finding literals."""
     pytest.importorskip("rdflib")
 
+    from paths import ontodir
+
     from tripper import RDF, XSD, Literal, Triplestore
-    from tripper.testutils import ontodir
 
     ts = Triplestore("rdflib")
     FAM = ts.bind("ex", "http://onto-ns.com/ontologies/examples/family#")

tests/test_utils.py

@@ -297,9 +297,10 @@ def test_random_string():
 def test_extend_namespace():
     """Test extend namespace()"""
     pytest.importorskip("rdflib")
+    from paths import ontodir
+
     from tripper import Namespace
     from tripper.errors import NoSuchIRIError
-    from tripper.testutils import ontodir
     from tripper.utils import extend_namespace
 
     FOOD = Namespace(
@@ -327,3 +328,25 @@ def test_extend_namespace():
     EX = Namespace("http://example.com#")
     with pytest.raises(TypeError):
         extend_namespace(EX, {"Item": EX + "Item"})
+
+
+def test_AttrDict():
+    """Test AttrDict."""
+    from tripper.utils import AttrDict
+
+    d = AttrDict(a=1, b=2)
+    assert d.a == 1
+
+    with pytest.raises(KeyError):
+        d.c  # pylint: disable=pointless-statement
+
+    d.c = 3
+    assert d.c == 3
+
+    d.get = 4
+    assert d["get"] == 4
+    assert d.get("get") == 4  # pylint: disable=not-callable
+
+    d2 = AttrDict({"a": "A"})
+    assert d2.a == "A"
+    assert d2 == {"a": "A"}

tripper/context/0.2/context.json

@@ -81,8 +81,8 @@
     "datamodel": "oteio:hasDatamodel",
     "hasDataSink": "oteio:hasDataSink",
     "storeURL": "oteio:storeURL",
-    "mappingURL": "oteio.mappingURL",
-    "mappingFormat": "oteio.mappingFormat",
+    "mappingURL": "oteio:mappingURL",
+    "mappingFormat": "oteio:mappingFormat",
 
     "subject": "rdf:subject",
     "predicate": "rdf:predicate",

tripper/dataset/__init__.py

@@ -3,9 +3,9 @@
 from .dataset import (
     get_context,
     get_prefixes,
-    load_datadoc,
+    get_shortnames,
     load_dataset,
-    prepare_datadoc,
+    read_datadoc,
+    save_datadoc,
     save_dataset,
-    update_dataset,
 )