Extract metadata from CITATION.cff

bids2openminds/converter.py

@@ -1,5 +1,6 @@
 import warnings
 import os
+import yaml
 import pandas as pd
 from ancpbids import BIDSLayout
 from ancpbids.query import Artifact
@@ -18,6 +19,7 @@
     ("participants.tsv", "participants", ".tsv"),
     ("participants.json", "participants", ".json"),
     ("CHANGES", None, None),
+    ("CITATION.cff", "CITATION", None),
     ("README", None, None),
     ("README.md", None, None),
 ]
@@ -77,10 +79,14 @@ def convert(input_path,  save_output=False, output_path=None, multiple_files=Fal
 
     subjects_id = bids_layout.get_subjects()
 
-    # imprting the dataset description file containing some of the
+    # importing the dataset description file containing some of the metadata
     dataset_description_path = utility.table_filter(layout_df, "description")
-
     dataset_description = utility.read_json(dataset_description_path.iat[0, 0])
+    citation_path = utility.table_filter(layout_df, "CITATION")
+    citation = None
+    if not citation_path.empty:
+        with open(citation_path.iat[0, 0], encoding="utf-8") as fp:
+            citation = yaml.safe_load(fp) 
 
     [subjects_dict, subject_state_dict, subjects_list] = main.create_subjects(
         subjects_id, layout_df, bids_layout, collection)
@@ -92,7 +98,7 @@ def convert(input_path,  save_output=False, output_path=None, multiple_files=Fal
         layout_df, input_path, collection)
 
     dataset_version = main.create_dataset_version(
-        bids_layout, dataset_description, layout_df, subjects_list, file_repository, behavioral_protocols, collection)
+        bids_layout, citation, dataset_description, layout_df, subjects_list, file_repository, behavioral_protocols, collection)
 
     dataset = main.create_dataset(
         dataset_description, dataset_version, collection)

bids2openminds/main.py

@@ -48,8 +48,35 @@ def create_openminds_person(full_name):
 
 
 def create_persons(dataset_description, collection):
+    # citation.cff case
+    if "authors" in dataset_description:
+        person_list = dataset_description["authors"]
+        openminds_list=[]
+        for person in person_list:
+            person_orcid = None
+            person_affiliation = None
+            person_contact_information = None
+            if 'orcid' in person:
+                person_orcid = [omcore.ORCID(identifier=person['orcid'])]
+            if 'email' in person:
+                person_contact_information = omcore.ContactInformation(email=person['email'])
+            if 'affiliation' in person:
+                # Handle multiple affiliations separated by semicolon
+                affiliation_list = [item.strip() for item in person['affiliation'].split(';')]
+                person_affiliation = []
+                for affiliation in affiliation_list:
+                    person_affiliation.append(omcore.Affiliation(
+                        member_of=omcore.Organization(full_name=affiliation)))
+
+            openminds_person = omcore.Person(
+                affiliations=person_affiliation, digital_identifiers=person_orcid, given_name=person['given-names'],
+                family_name=person['family-names'], contact_information=person_contact_information)
+            openminds_list.append(openminds_person)
+            collection.add(openminds_person)
+        return openminds_list
 
-    if "Authors" in dataset_description:
+    # dataset_description.json case
+    elif "Authors" in dataset_description:
         person_list = dataset_description["Authors"]
     else:
         return None
@@ -185,21 +212,39 @@ def create_openminds_age(data_subject):
         return None
 
 
-def create_dataset_version(bids_layout, dataset_description, layout_df, studied_specimens, file_repository, behavioral_protocols, collection):
+def create_dataset_version(bids_layout, citation, dataset_description, layout_df, studied_specimens, file_repository, behavioral_protocols, collection):
 
     # Fetch the dataset type from dataset description file
 
     # dataset_type=bids2openminds_instance(dataset_description.get("DatasetType",None))
-
-    # Fetch the digitalIdentifier from dataset description file
-
-    if "DatasetDOI" in dataset_description:
-        digital_identifier = omcore.DOI(
-            identifier=dataset_description["DatasetDOI"])
+    license = None
+    digital_identifier = None
+    name = None
+    version_identifier = None
+    # General rules on the usage of CITATION.cff and dataset_description.json
+    # https://bids-specification.readthedocs.io/en/stable/modality-agnostic-files/dataset-description.html#citationcff
+    if citation:
+        if 'doi' in citation:
+            digital_identifier = omcore.DOI(identifier=citation['doi'])
+        if 'license' in citation:
+            license = omcore.License.by_name(citation['license'])
+            if license is None:
+                warn(f"Could not resolve license '{citation['license']}' "
+                    "to an openMINDS License."
+                )
+        if 'title' in citation:
+            name = citation['title']
+        if 'version' in citation:
+            version_identifier = citation['version']
+        authors = create_persons(citation, collection)
     else:
-        digital_identifier = None
+        # if CITATION.cff is present, the "Authors" field of dataset_description.json MUST be omitted
+        authors = create_persons(dataset_description, collection)
 
-    authors = create_persons(dataset_description, collection)
+    name = dataset_description["Name"] if name is None else name
+    # Fetch the digitalIdentifier from dataset description file
+    if digital_identifier is None and "DatasetDOI" in dataset_description:
+        digital_identifier = omcore.DOI(identifier=dataset_description["DatasetDOI"])
 
     if "Acknowledgements" in dataset_description:
         other_contribution = dataset_description["Acknowledgements"]
@@ -235,19 +280,20 @@ def create_dataset_version(bids_layout, dataset_description, layout_df, studied_
     experimental_approaches = create_approaches(layout_df)
 
     dataset_version = omcore.DatasetVersion(
+        authors=authors,
+        behavioral_protocols=behavioral_protocols,
+        data_types=dataset_type,
         digital_identifier=digital_identifier,
         experimental_approaches=experimental_approaches,
-        short_name=dataset_description["Name"],
-        full_name=dataset_description["Name"],
-        studied_specimens=studied_specimens,
-        authors=authors,
-        techniques=techniques,
+        full_name=name,
         how_to_cite=how_to_cite,
+        license=license,
         repository=file_repository,
-        behavioral_protocols=behavioral_protocols,
-        data_types=dataset_type
+        short_name=name,
+        studied_specimens=studied_specimens,
+        techniques=techniques,
+        version_identifier=version_identifier,
         # other_contributions=other_contribution  # needs to be a Contribution object
-        # version_identifier
     )
 
     collection.add(dataset_version)

pyproject.toml

@@ -6,7 +6,8 @@ dependencies = [
   "openminds >= 0.2.3",
   "click>=8.1",
   "pandas",
-  "nameparser >= 1.1.3"
+  "nameparser >= 1.1.3",
+  "PyYAML >= 6.0"
 ]
 requires-python = ">=3.9"
 authors = [

test/test_dataset_version.py

@@ -0,0 +1,63 @@
+import pytest
+
+import pandas as pd
+from openminds import Collection
+import openminds.v4.core as omcore
+
+from bids2openminds.main import create_dataset_version
+
+# Mock BIDS layout DataFrame
+layout_df = pd.DataFrame({
+    "suffix": ["T1w", "T2w", "bold"],
+    "path": ["sub-01/anat/sub-01_T1w.nii.gz", "sub-01/anat/sub-01_T2w.nii.gz", "sub-01/func/sub-01_task-rest_bold.nii.gz"],
+    "datatype": ["anat", "anat", "func"]
+})
+
+
+def test_create_dataset_version_citation():
+    # Mock loaded CITATION.cff
+    citation = {
+        "title": "My Dataset",
+        "version": "1.05.9",
+        "doi": "10.5281/zenodo.123456",
+        "license": "Apache-2.0"
+    }
+
+    citation_dataset_version = create_dataset_version(
+        "", citation, {}, layout_df, [], [], [], Collection()
+    )
+
+    expected = omcore.DatasetVersion(
+        digital_identifier=omcore.DOI(identifier="10.5281/zenodo.123456"),
+        full_name="My Dataset",
+        short_name="My Dataset",
+        license=omcore.License.apache_2_0,
+        version_identifier="1.05.9"
+    )
+
+    for field in ["full_name", "short_name", "version_identifier", "license"]:
+        assert getattr(citation_dataset_version, field) == getattr(expected, field)
+    assert citation_dataset_version.digital_identifier.identifier == expected.digital_identifier.identifier
+
+
+def test_create_dataset_version_without_citation():
+    # Mock missing CITATION.cff
+    citation = None
+    dataset_description = {
+        "Name": "My Dataset",
+        "DatasetDOI": "10.5281/zenodo.123456"
+    }
+
+    dataset_version = create_dataset_version(
+        "", citation, dataset_description, layout_df, [], [], [], Collection()
+    )
+
+    expected = omcore.DatasetVersion(
+        digital_identifier=omcore.DOI(identifier="10.5281/zenodo.123456"),
+        full_name="My Dataset",
+        short_name="My Dataset"
+    )
+
+    for field in ["full_name", "short_name"]:
+        assert getattr(dataset_version, field) == getattr(expected, field)
+    assert dataset_version.digital_identifier.identifier == expected.digital_identifier.identifier

test/test_person.py

@@ -1,8 +1,11 @@
 # test for create_openminds_person function in the main
 import pytest
-from bids2openminds.main import create_openminds_person
+
+from openminds import Collection
 import openminds.v4.core as omcore
 
+from bids2openminds.main import create_openminds_person, create_persons
+
 # Test data: (full_name, given_name, family_name)
 example_names = [("John Ronald Reuel Tolkien", "John Ronald Reuel", "Tolkien"),
                  ("Bilbo Baggins", "Bilbo", "Baggins"),
@@ -28,6 +31,60 @@ def test_create_openminds_person(full_name, given_name, family_name):
     # assert openminds_person_object == bids2openminds_person_object
 
 
+def test_create_openminds_person_citation_explicit():
+    citation = {
+        "authors": [
+            {
+                "family-names": "Awart",
+                "given-names": "Peter",
+                "affiliation": "Place1",
+                "orcid": "https://orcid.org/1234-5678-9123-4567"
+            },
+            {
+                "family-names": "Detienne",
+                "given-names": "Franck",
+                "affiliation": "Place1; Place2"
+            }
+        ]
+    }
+
+    persons = create_persons(citation, Collection())
+    author1 = persons[0]
+    author2 = persons[1]
+
+    # Expected objects
+    expected_author1 = omcore.Person(
+        given_name="Peter",
+        family_name="Awart",
+        affiliations=[
+            omcore.Affiliation(member_of=omcore.Organization(full_name="Place1"))
+        ],
+        digital_identifiers=[omcore.ORCID(identifier="https://orcid.org/1234-5678-9123-4567")]
+    )
+
+    expected_author2 = omcore.Person(
+        given_name="Franck",
+        family_name="Detienne",
+        affiliations=[
+            omcore.Affiliation(member_of=omcore.Organization(full_name="Place1")),
+            omcore.Affiliation(member_of=omcore.Organization(full_name="Place2"))
+        ]
+    )
+
+    assert author1.given_name == expected_author1.given_name
+    assert author1.family_name == expected_author1.family_name
+    assert author1.digital_identifiers[0].identifier == expected_author1.digital_identifiers[0].identifier
+    assert author2.given_name == expected_author2.given_name
+    assert author2.family_name == expected_author2.family_name
+
+    # Single affiliation
+    assert author1.affiliations[0].member_of.full_name == expected_author1.affiliations[0].member_of.full_name
+
+    # Multiple affiliations
+    assert author2.affiliations[0].member_of.full_name == expected_author2.affiliations[0].member_of.full_name
+    assert author2.affiliations[1].member_of.full_name == expected_author2.affiliations[1].member_of.full_name
+
+
 @pytest.mark.parametrize("full_not_name", example_not_names)
 def test_create_openminds_person_not_names(full_not_name):
     bids2openminds_person_object = create_openminds_person(full_not_name)