Merge pull request #343 from incf-nidash/mktmps

Write test temp files to temp directories

tests/core/test_provone.py

@@ -23,7 +23,7 @@ def doc():
     return d1
 
 
-def test_ispartof(doc):
+def test_ispartof(doc, tmp_path) -> None:
     workflow_1ex1 = doc.processExec(
         "dcterms:identifier:wf1_ex1",
         "2013-08-21 13:37:54",
@@ -37,11 +37,11 @@ def test_ispartof(doc):
     doc.isPartOf(pe1, workflow_1ex1)
 
     # save a turtle file
-    with open("test.ttl", "w") as f:
+    with open(tmp_path / "test.ttl", "w") as f:
         f.write(doc.serialize(format="rdf", rdf_format="ttl"))
 
 
-def test_used(doc):
+def test_used(doc, tmp_path) -> None:
     pe1 = doc.processExec(
         "dcterms:identifier:e1_ex1", "2013-08-21 13:37:53", "2013-08-21 13:37:53"
     )
@@ -56,11 +56,11 @@ def test_used(doc):
     doc.used(pe1, dt1)
 
     # save a turtle file
-    with open("test.ttl", "w") as f:
+    with open(tmp_path / "test.ttl", "w") as f:
         f.write(doc.serialize(format="rdf", rdf_format="ttl"))
 
 
-def test_wasderivedfrom(doc):
+def test_wasderivedfrom(doc, tmp_path) -> None:
     dt1 = doc.data(
         "dcterms:identifier:defparam1",
         {
@@ -73,27 +73,27 @@ def test_wasderivedfrom(doc):
     doc.wasDerivedFrom(dt1, dt2)
 
     # save a turtle file
-    with open("test.ttl", "w") as f:
+    with open(tmp_path / "test.ttl", "w") as f:
         f.write(doc.serialize(format="rdf", rdf_format="ttl"))
 
 
-def test_dataonlink(doc):
+def test_dataonlink(doc, tmp_path) -> None:
     dt2 = doc.data("dcterms:identifier:defparam2", {"rdfs:label": "filename"})
     dl1 = doc.dataLink("dcterms:identifier:e1_e2DL")
     # save a turtle file
-    with open("test.ttl", "w") as f:
+    with open(tmp_path / "test.ttl", "w") as f:
         f.write(doc.serialize(format="rdf", rdf_format="ttl"))
     doc.dataOnLink(dt2, dl1)
 
 
-def test_wasgeneratedby(doc):
+def test_wasgeneratedby(doc, tmp_path) -> None:
     dt2 = doc.data("dcterms:identifier:defparam2", {"rdfs:label": "filename"})
     pe1 = doc.processExec(
         "dcterms:identifier:e1_ex1", "2013-08-21 13:37:53", "2013-08-21 13:37:53"
     )
     doc.wasGeneratedBy(dt2, pe1)
     # save a turtle file
-    with open("test.ttl", "w") as f:
+    with open(tmp_path / "test.ttl", "w") as f:
         f.write(doc.serialize(format="rdf", rdf_format="ttl"))
 
 
@@ -135,9 +135,9 @@ def test_dltoinport(doc):
     doc.DLToInPort(dl1, i1)
 
 
-def test_documentserialize(doc):
+def test_documentserialize(doc, tmp_path) -> None:
     # save a turtle file
-    with open("test.ttl", "w") as f:
+    with open(tmp_path / "test.ttl", "w") as f:
         f.write(doc.serialize(format="rdf", rdf_format="ttl"))
 
 

tests/experiment/conftest.py

@@ -0,0 +1,66 @@
+from __future__ import annotations
+from dataclasses import dataclass
+from typing import Optional
+import urllib.request
+import pytest
+from nidm.experiment.tools.rest import RestParser
+
+# We will test example NIDM files downloaded from
+# the GitHub dbkeator/simple2_NIDM_examples repo
+#
+# DBK: this is a bit unsafe as the TTL files in the github repo above can change and the UUID will change since they are randomly
+# generated at this point.  It's probably more robust to explicitly create these files for the time being and explicitly set the
+# UUID in the test file:
+# For example:  kwargs={Constants.NIDM_PROJECT_NAME:"FBIRN_PhaseIII",Constants.NIDM_PROJECT_IDENTIFIER:1200,Constants.NIDM_PROJECT_DESCRIPTION:"Test investigation2"}
+#               project = Project(uuid="_654321",attributes=kwargs)
+
+
+@pytest.fixture(scope="session")
+def brain_vol_files(tmp_path_factory: pytest.TempPathFactory) -> list[str]:
+    tmp_path = tmp_path_factory.mktemp("brain_vol_files")
+    urllib.request.urlretrieve(
+        "https://raw.githubusercontent.com/dbkeator/simple2_NIDM_examples/master/datasets.datalad.org/abide/RawDataBIDS/CMU_a/nidm.ttl",
+        tmp_path / "cmu_a.nidm.ttl",
+    )
+    urllib.request.urlretrieve(
+        "https://raw.githubusercontent.com/dbkeator/simple2_NIDM_examples/master/datasets.datalad.org/abide/RawDataBIDS/Caltech/nidm.ttl",
+        tmp_path / "caltech.nidm.ttl",
+    )
+    return [
+        str(tmp_path / "cmu_a.nidm.ttl"),
+        str(tmp_path / "caltech.nidm.ttl"),
+    ]
+
+
+@dataclass
+class BrainVol:
+    files: list[str]
+    restParser: RestParser
+    cmu_test_project_uuid: str
+    cmu_test_subject_uuid: str
+
+
+@pytest.fixture(scope="session")
+def brain_vol(brain_vol_files: list[str]) -> BrainVol:
+    restParser = RestParser(output_format=RestParser.OBJECT_FORMAT)
+    projects = restParser.run(brain_vol_files, "/projects")
+    cmu_test_project_uuid: Optional[str] = None
+    for p in projects:
+        proj_info = restParser.run(brain_vol_files, f"/projects/{p}")
+        if (
+            isinstance(proj_info, dict)
+            and proj_info.get("dctypes:title") == "ABIDE - CMU_a"
+        ):
+            cmu_test_project_uuid = p
+            break
+    assert cmu_test_project_uuid is not None
+    subjects = restParser.run(
+        brain_vol_files, f"/projects/{cmu_test_project_uuid}/subjects"
+    )
+    cmu_test_subject_uuid = subjects["uuid"][0]
+    return BrainVol(
+        brain_vol_files,
+        restParser,
+        cmu_test_project_uuid,
+        cmu_test_subject_uuid,
+    )

tests/experiment/test_experiment.py

@@ -1,3 +1,5 @@
+from pathlib import Path
+import pytest
 from nidm.core import Constants
 from nidm.experiment import (
     AssessmentAcquisition,
@@ -102,5 +104,6 @@ def main():
 
 
 # very simple test, just checking if main does not give any error
-def test_main():
+def test_main(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:
+    monkeypatch.chdir(tmp_path)
     main()

tests/experiment/test_experiment_basic.py

@@ -2,39 +2,31 @@
 import json
 from pathlib import Path
 import prov
-import pytest
 import rdflib
 from nidm.core import Constants
 from nidm.experiment import Project, Session
 
 
-def test_1(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:
-    monkeypatch.chdir(tmp_path)
-
+def test_1(tmp_path: Path) -> None:
     project = Project()
-
     # save a turtle file
-    with open("test.ttl", "w") as f:
+    with open(tmp_path / "test.ttl", "w") as f:
         f.write(project.serializeTurtle())
 
 
-def test_2(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:
-    monkeypatch.chdir(tmp_path)
-
+def test_2(tmp_path: Path) -> None:
     kwargs = {
         Constants.NIDM_PROJECT_NAME: "FBIRN_PhaseII",
         Constants.NIDM_PROJECT_IDENTIFIER: 9610,
         Constants.NIDM_PROJECT_DESCRIPTION: "Test investigation",
     }
     project = Project(attributes=kwargs)
 
-    with open("test.ttl", "w") as f:
+    with open(tmp_path / "test.ttl", "w") as f:
         f.write(project.serializeTurtle())
 
 
-def test_sessions_1(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:
-    monkeypatch.chdir(tmp_path)
-
+def test_sessions_1() -> None:
     project = Project()
     assert project.sessions == []
 
@@ -48,19 +40,15 @@ def test_sessions_1(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:
     assert session2.label == project.sessions[1].label
 
 
-def test_sessions_2(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:
-    monkeypatch.chdir(tmp_path)
-
+def test_sessions_2() -> None:
     project = Project()
     assert project.sessions == []
 
     session1 = Session(project)
     assert project.sessions[0].label == session1.label
 
 
-def test_sessions_3(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:
-    monkeypatch.chdir(tmp_path)
-
+def test_sessions_3() -> None:
     project1 = Project()
     project2 = Project()
 
@@ -188,7 +176,7 @@ def test_session_noparameters():
     assert len(proj.graph.get_records()) == 2
 
 
-def test_jsonld_exports():
+def test_jsonld_exports(tmp_path: Path) -> None:
     kwargs = {
         Constants.NIDM_PROJECT_NAME: "FBIRN_PhaseII",
         Constants.NIDM_PROJECT_IDENTIFIER: 9610,
@@ -197,16 +185,15 @@ def test_jsonld_exports():
     project = Project(uuid="_123456", attributes=kwargs)
 
     # save a turtle file
-    with open("test.json", "w") as f:
+    with open(tmp_path / "test.json", "w") as f:
         f.write(project.serializeJSONLD())
 
     # load in JSON file
-    with open("test.json") as json_file:
+    with open(tmp_path / "test.json") as json_file:
         data = json.load(json_file)
 
     assert data["Identifier"]["@value"] == "9610"
     # WIP  Read back in json-ld file and check that we have the project info
-    # remove("test.json")
 
 
 def test_project_trig_serialization():

tests/experiment/test_map_vars_to_terms.py

@@ -1,15 +1,20 @@
+from dataclasses import dataclass
 import json
-from os.path import join
-import tempfile
+from pathlib import Path
 import pandas as pd
 import pytest
 from nidm.experiment.Utils import map_variables_to_terms
 
 
-@pytest.fixture(scope="module", autouse="True")
-def setup():
-    global DATA, REPROSCHEMA_JSON_MAP, BIDS_SIDECAR
+@dataclass
+class Setup:
+    data: pd.DataFrame
+    reproschema_json_map: dict
+    bids_sidecar: dict
 
+
+@pytest.fixture(scope="module")
+def setup() -> Setup:
     temp = {
         "participant_id": [
             "100",
@@ -27,9 +32,9 @@ def setup():
         "sex": ["m", "m", "f", "m", "f", "f", "f", "f", "m", "m"],
     }
 
-    DATA = pd.DataFrame(temp)
+    data = pd.DataFrame(temp)
 
-    REPROSCHEMA_JSON_MAP = json.loads(
+    reproschema_json_map = json.loads(
         """
         {
             "DD(source='participants.tsv', variable='participant_id')": {
@@ -89,7 +94,7 @@ def setup():
         }"""
     )
 
-    BIDS_SIDECAR = json.loads(
+    bids_sidecar = json.loads(
         """
         {
             "age": {
@@ -128,23 +133,26 @@ def setup():
                 ]
             }
         }
-
         """
     )
 
+    return Setup(
+        data=data,
+        reproschema_json_map=reproschema_json_map,
+        bids_sidecar=bids_sidecar,
+    )
+
 
-def test_map_vars_to_terms_BIDS():
+def test_map_vars_to_terms_BIDS(setup: Setup, tmp_path: Path) -> None:
     """
     This function will test the Utils.py "map_vars_to_terms" function with a BIDS-formatted
     JSON sidecar file
     """
 
-    global DATA, BIDS_SIDECAR
-
     column_to_terms, cde = map_variables_to_terms(
-        df=DATA,
-        json_source=BIDS_SIDECAR,
-        directory=tempfile.gettempdir(),
+        df=setup.data,
+        json_source=setup.bids_sidecar,
+        directory=str(tmp_path),
         assessment_name="test",
         bids=True,
     )
@@ -197,7 +205,7 @@ def test_map_vars_to_terms_BIDS():
     )
 
     # now check the JSON sidecar file created by map_variables_to_terms which should match BIDS format
-    with open(join(tempfile.gettempdir(), "nidm_annotations.json")) as fp:
+    with open(tmp_path / "nidm_annotations.json") as fp:
         bids_sidecar = json.load(fp)
 
     assert "age" in bids_sidecar.keys()
@@ -237,18 +245,16 @@ def test_map_vars_to_terms_BIDS():
     assert len(results) == 20
 
 
-def test_map_vars_to_terms_reproschema():
+def test_map_vars_to_terms_reproschema(setup: Setup, tmp_path: Path) -> None:
     """
     This function will test the Utils.py "map_vars_to_terms" function with a reproschema-formatted
     JSON sidecar file
     """
 
-    global DATA, REPROSCHEMA_JSON_MAP
-
     column_to_terms, cde = map_variables_to_terms(
-        df=DATA,
-        json_source=REPROSCHEMA_JSON_MAP,
-        directory=tempfile.gettempdir(),
+        df=setup.data,
+        json_source=setup.reproschema_json_map,
+        directory=str(tmp_path),
         assessment_name="test",
     )
 
@@ -300,7 +306,7 @@ def test_map_vars_to_terms_reproschema():
     )
 
     # now check the JSON mapping file created by map_variables_to_terms which should match Reproschema format
-    with open(join(tempfile.gettempdir(), "nidm_annotations.json")) as fp:
+    with open(tmp_path / "nidm_annotations_annotations.json") as fp:
         json.load(fp)
 
     assert "DD(source='test', variable='age')" in column_to_terms.keys()