fix(core): add missing Subject to dataverse export (#2420)

renku/core/commands/providers/dataverse.py

@@ -34,6 +34,7 @@
 from renku.core.commands.providers.doi import DOIProvider
 from renku.core.models.datasets import Dataset, DatasetFile, DatasetSchema
 from renku.core.models.provenance.agents import PersonSchema
+from renku.core.utils import communication
 from renku.core.utils.doi import extract_doi, is_doi
 from renku.core.utils.requests import retry
 
@@ -48,6 +49,24 @@
 DATAVERSE_EXPORTER = "schema.org"
 
 
+DATAVERSE_SUBJECTS = [
+    "Agricultural Sciences",
+    "Arts and Humanities",
+    "Astronomy and Astrophysics",
+    "Business and Management",
+    "Chemistry",
+    "Computer and Information Science",
+    "Earth and Environmental Sciences",
+    "Engineering",
+    "Law",
+    "Mathematical Sciences",
+    "Medicine, Health and Life Sciences",
+    "Physics",
+    "Social Sciences",
+    "Other",
+]
+
+
 class _DataverseDatasetSchema(DatasetSchema):
     """Schema for Dataverse datasets."""
 
@@ -411,15 +430,26 @@ def export(self, publish, **kwargs):
 
     def _get_dataset_metadata(self):
         authors, contacts = self._get_creators()
+        subject = self._get_subject()
         metadata_template = Template(DATASET_METADATA_TEMPLATE)
         metadata = metadata_template.substitute(
             name=_escape_json_string(self.dataset.title),
             authors=json.dumps(authors),
             contacts=json.dumps(contacts),
             description=_escape_json_string(self.dataset.description),
+            subject=subject,
         )
         return json.loads(metadata)
 
+    def _get_subject(self):
+        text_prompt = "Subject of this dataset: \n\n"
+        text_prompt += "\n".join(f"{s}\t[{i}]" for i, s in enumerate(DATAVERSE_SUBJECTS, start=1))
+        text_prompt += "\n\nSubject"
+
+        selection = communication.prompt(text_prompt, type=int, default=len(DATAVERSE_SUBJECTS)) or 0
+
+        return DATAVERSE_SUBJECTS[selection - 1]
+
     def _get_creators(self):
         authors = []
         contacts = []

renku/core/commands/providers/dataverse_metadata_templates.py

@@ -56,7 +56,7 @@
                         "typeName": "dsDescription"
                     },
                     {
-                        "value": [],
+                        "value": ["${subject}"],
                         "typeClass": "controlledVocabulary",
                         "multiple": true,
                         "typeName": "subject"

tests/cli/test_integration_datasets.py

@@ -479,15 +479,16 @@ def test_renku_dataset_import_missing_lfs_objects(runner, project):
 @pytest.mark.integration
 @flaky(max_runs=10, min_passes=1)
 @pytest.mark.parametrize(
-    "provider,params,output",
+    "provider,params,output,input",
     [
-        ("zenodo", [], "zenodo.org/deposit"),
-        ("dataverse", ["--dataverse-name", "sdsc-test-dataverse"], "doi:"),
-        ("olos", ["--dlcm-server", "https://sandbox.dlcm.ch/"], "sandbox.dlcm.ch/ingestion/preingest/deposits/"),
+        ("zenodo", [], "zenodo.org/deposit", None),
+        ("dataverse", ["--dataverse-name", "sdsc-test-dataverse"], "doi:", "1"),
+        ("olos", ["--dlcm-server", "https://sandbox.dlcm.ch/"], "sandbox.dlcm.ch/ingestion/preingest/deposits/", None),
     ],
 )
 def test_dataset_export_upload_file(
-    runner, project, tmpdir, client, zenodo_sandbox, dataverse_demo, olos_sandbox, provider, params, output
+    runner, project, tmpdir, client, zenodo_sandbox, dataverse_demo, olos_sandbox, provider, params, output,
+    input,
 ):
     """Test successful uploading of a file to Zenodo/Dataverse deposit."""
     result = runner.invoke(cli, ["dataset", "create", "my-dataset"])
@@ -511,7 +512,9 @@ def test_dataset_export_upload_file(
     data_repo.git.add(update=True)
     data_repo.index.commit("metadata updated")
 
-    result = runner.invoke(cli, ["dataset", "export", "my-dataset", provider] + params)
+    result = runner.invoke(
+        cli, ["dataset", "export", "my-dataset", provider] + params, input=input, catch_exceptions=False
+    )
 
     assert 0 == result.exit_code, result.output + str(result.stderr_bytes)
     assert "Exported to:" in result.output
@@ -521,15 +524,16 @@ def test_dataset_export_upload_file(
 @pytest.mark.integration
 @flaky(max_runs=10, min_passes=1)
 @pytest.mark.parametrize(
-    "provider,params,output",
+    "provider,params,output,input",
     [
-        ("zenodo", [], "zenodo.org/deposit"),
-        ("dataverse", ["--dataverse-name", "sdsc-test-dataverse"], "doi:"),
-        ("olos", ["--dlcm-server", "https://sandbox.dlcm.ch/"], "sandbox.dlcm.ch/ingestion/preingest/deposits/"),
+        ("zenodo", [], "zenodo.org/deposit", None),
+        ("dataverse", ["--dataverse-name", "sdsc-test-dataverse"], "doi:", "1"),
+        ("olos", ["--dlcm-server", "https://sandbox.dlcm.ch/"], "sandbox.dlcm.ch/ingestion/preingest/deposits/", None),
     ],
 )
 def test_dataset_export_upload_tag(
-    runner, project, tmpdir, client, zenodo_sandbox, dataverse_demo, olos_sandbox, provider, params, output
+        runner, project, tmpdir, client, zenodo_sandbox, dataverse_demo, olos_sandbox, provider, params, output,
+    input,
 ):
     """Test successful uploading of a file to Zenodo/Dataverse deposit."""
     result = runner.invoke(cli, ["dataset", "create", "my-dataset"])
@@ -568,19 +572,19 @@ def test_dataset_export_upload_tag(
     result = runner.invoke(cli, ["dataset", "tag", "my-dataset", "2.0"])
     assert 0 == result.exit_code, result.output + str(result.stderr_bytes)
 
-    result = runner.invoke(cli, ["dataset", "export", "my-dataset", provider, "-t", "2.0"] + params)
+    result = runner.invoke(cli, ["dataset", "export", "my-dataset", provider, "-t", "2.0"] + params, input=input)
 
     assert 0 == result.exit_code
     assert "Exported to:" in result.output
     assert output in result.output
 
-    result = runner.invoke(cli, ["dataset", "export", "my-dataset", provider, "-t", "1.0"] + params)
+    result = runner.invoke(cli, ["dataset", "export", "my-dataset", provider, "-t", "1.0"] + params, input=input)
 
     assert 0 == result.exit_code, result.output + str(result.stderr_bytes)
     assert "Exported to:" in result.output
     assert output in result.output
 
-    result = runner.invoke(cli, ["dataset", "export", "my-dataset", provider] + params, input="1")  # HEAD
+    result = runner.invoke(cli, ["dataset", "export", "my-dataset", provider] + params, input=f"0\n{input}")  # HEAD
 
     assert 0 == result.exit_code, result.output + str(result.stderr_bytes)
     assert "Exported to:" in result.output
@@ -590,15 +594,16 @@ def test_dataset_export_upload_tag(
 @pytest.mark.integration
 @flaky(max_runs=10, min_passes=1)
 @pytest.mark.parametrize(
-    "provider,params,output",
+    "provider,params,output,input",
     [
-        ("zenodo", [], "zenodo.org/deposit"),
-        ("dataverse", ["--dataverse-name", "sdsc-test-dataverse"], "doi:"),
-        ("olos", ["--dlcm-server", "https://sandbox.dlcm.ch/"], "sandbox.dlcm.ch/ingestion/preingest/deposits/"),
+        ("zenodo", [], "zenodo.org/deposit", None),
+        ("dataverse", ["--dataverse-name", "sdsc-test-dataverse"], "doi:", "1"),
+        ("olos", ["--dlcm-server", "https://sandbox.dlcm.ch/"], "sandbox.dlcm.ch/ingestion/preingest/deposits/", None),
     ],
 )
 def test_dataset_export_upload_multiple(
-    runner, project, tmpdir, client, zenodo_sandbox, dataverse_demo, olos_sandbox, provider, params, output
+        runner, project, tmpdir, client, zenodo_sandbox, dataverse_demo, olos_sandbox, provider, params, output,
+    input,
 ):
     """Test successful uploading of a files to Zenodo deposit."""
     result = runner.invoke(cli, ["dataset", "create", "my-dataset"])
@@ -625,7 +630,7 @@ def test_dataset_export_upload_multiple(
     data_repo.git.add(update=True)
     data_repo.index.commit("metadata updated")
 
-    result = runner.invoke(cli, ["dataset", "export", "my-dataset", provider] + params)
+    result = runner.invoke(cli, ["dataset", "export", "my-dataset", provider] + params, input=input)
 
     assert 0 == result.exit_code, result.output + str(result.stderr_bytes)
     assert "Exported to:" in result.output

tests/cli/test_save.py

@@ -51,7 +51,7 @@ def test_save_with_remote(runner, project, client_with_remote, tmpdir_factory):
         fp.write("tracked file")
 
     result = runner.invoke(cli, ["save", "-m", "save changes", "tracked"], catch_exceptions=False)
-    breakpoint()
+
     assert 0 == result.exit_code
     assert "tracked" in result.output
     assert "save changes" in client_with_remote.repo.head.commit.message