feat: add check_publication_list_format check

CHANGELOG.md

@@ -57,6 +57,7 @@
 * Added `check_image_series_starting_frame_without_external_file` to verify that `starting_frame` is not set when `external_file` is not used in an `ImageSeries`. [#235](https://github.com/NeurodataWithoutBorders/nwbinspector/issues/235)
 * Added `check_sweeptable_deprecated` to detect usage of the deprecated `SweepTable` in NWB files with schema version >= 2.4.0, which should use `IntracellularRecordingsTable` instead. [#657](https://github.com/NeurodataWithoutBorders/nwbinspector/issues/657)
 * Added `check_time_series_data_is_not_empty` to detect empty `.data` fields in `TimeSeries` containers, which often indicate incomplete data entry or conversion errors. Skips `ImageSeries` with `external_file` set, where empty data is intentional. [#668](https://github.com/NeurodataWithoutBorders/nwbinspector/pull/668)
+* Added `check_publication_list_format` to detect comma-separated DOIs/URLs in `related_publications` entries that should be separate list entries. [#419](https://github.com/NeurodataWithoutBorders/nwbinspector/issues/419)
 
 ### Improvements
 * Added documentation to API and CLI docs on how to use the dandi config option. [#624](https://github.com/NeurodataWithoutBorders/nwbinspector/pull/624)

docs/best_practices/nwbfile_metadata.rst

@@ -181,7 +181,12 @@ of the form ``'doi: ###'`` or as an external link of the form ``'http://dx.doi.o
 This allows metadata collection programs, such as those on the :dandi-archive:`DANDI archive <>` to easily form direct
 hyperlinks to the publications.
 
-Check function: :py:meth:`~nwbinspector.checks._nwbfile_metadata.check_doi_publications`
+Each publication should be a separate entry in the list. Do not combine multiple DOIs or URLs into a single
+comma-separated string. For example, use ``["https://doi.org/10.1234/abc", "https://doi.org/10.5678/def"]`` instead of
+``["https://doi.org/10.1234/abc,https://doi.org/10.5678/def"]``.
+
+Check functions: :py:meth:`~nwbinspector.checks._nwbfile_metadata.check_doi_publications` and
+:py:meth:`~nwbinspector.checks._nwbfile_metadata.check_publication_list_format`
 
 
 

src/nwbinspector/checks/__init__.py

@@ -55,6 +55,7 @@
     check_institution,
     check_keywords,
     check_processing_module_name,
+    check_publication_list_format,
     check_session_id_no_slashes,
     check_session_start_time_future_date,
     check_session_start_time_old_date,
@@ -151,6 +152,7 @@
     "check_session_start_time_future_date",
     "check_processing_module_name",
     "check_session_start_time_old_date",
+    "check_publication_list_format",
     "check_optogenetic_stimulus_site_has_optogenetic_series",
     "check_excitation_lambda_in_nm",
     "check_plane_segmentation_image_mask_shape_against_ref_images",

src/nwbinspector/checks/_nwbfile_metadata.py

@@ -168,6 +168,35 @@ def check_doi_publications(nwbfile: NWBFile) -> Optional[Iterable[InspectorMessa
     return None
 
 
+@register_check(importance=Importance.BEST_PRACTICE_VIOLATION, neurodata_type=NWBFile)
+def check_publication_list_format(nwbfile: NWBFile) -> Optional[Iterable[InspectorMessage]]:
+    """
+    Check if related_publications entries contain comma-separated values that should be separate list entries.
+
+    Best Practice: :ref:`best_practice_doi_publications`
+    """
+    if not nwbfile.related_publications:
+        return None
+    for publication in nwbfile.related_publications:
+        publication = publication.decode() if isinstance(publication, bytes) else publication
+        # Check for comma-separated DOIs or URLs within a single entry
+        # Look for patterns like "doi:xxx,doi:yyy" or "https://doi.org/xxx,https://doi.org/yyy"
+        if "," in publication:
+            # Check if the comma appears to separate multiple DOIs/URLs
+            parts = [p.strip() for p in publication.split(",")]
+            doi_indicators = ["doi:", "doi.org/"]
+            doi_like_parts = [part for part in parts if any(indicator in part.lower() for indicator in doi_indicators)]
+            if len(doi_like_parts) > 1:
+                yield InspectorMessage(
+                    message=(
+                        f"Metadata /general/related_publications contains a comma-separated list '{publication}'. "
+                        "Each publication should be a separate entry in the list, not combined in a single string."
+                    )
+                )
+
+    return None
+
+
 @register_check(importance=Importance.CRITICAL, neurodata_type=Subject)
 def check_subject_age(subject: Subject) -> Optional[InspectorMessage]:
     """Check if the Subject age is in ISO 8601 or our extension of it for ranges."""

tests/unit_tests/test_nwbfile_metadata.py

@@ -23,6 +23,7 @@
     check_institution,
     check_keywords,
     check_processing_module_name,
+    check_publication_list_format,
     check_session_id_no_slashes,
     check_session_start_time_future_date,
     check_session_start_time_old_date,
@@ -277,6 +278,110 @@ def test_check_doi_publications_multiple_fail():
     ]
 
 
+def test_check_publication_list_format_pass():
+    """Test that properly formatted publications pass the check."""
+    nwbfile = NWBFile(
+        session_description="",
+        identifier=str(uuid4()),
+        session_start_time=datetime.now().astimezone(),
+        related_publications=["https://doi.org/10.1234/abc", "https://doi.org/10.5678/def"],
+    )
+    assert check_publication_list_format(nwbfile) is None
+
+
+def test_check_publication_list_format_pass_single_comma_in_title():
+    """Test that a single publication with a comma in the title passes (not multiple DOIs)."""
+    nwbfile = NWBFile(
+        session_description="",
+        identifier=str(uuid4()),
+        session_start_time=datetime.now().astimezone(),
+        related_publications=["Some publication title, with comma"],
+    )
+    assert check_publication_list_format(nwbfile) is None
+
+
+def test_check_publication_list_format_pass_no_publications():
+    """Test that no related_publications passes the check."""
+    nwbfile = NWBFile(
+        session_description="",
+        identifier=str(uuid4()),
+        session_start_time=datetime.now().astimezone(),
+    )
+    assert check_publication_list_format(nwbfile) is None
+
+
+def test_check_publication_list_format_fail_comma_separated_doi_urls():
+    """Test detection of comma-separated DOI URLs in a single entry."""
+    nwbfile = NWBFile(
+        session_description="",
+        identifier=str(uuid4()),
+        session_start_time=datetime.now().astimezone(),
+        related_publications=["https://doi.org/10.1234/abc,https://doi.org/10.5678/def"],
+    )
+    assert check_publication_list_format(nwbfile) == [
+        InspectorMessage(
+            message=(
+                "Metadata /general/related_publications contains a comma-separated list "
+                "'https://doi.org/10.1234/abc,https://doi.org/10.5678/def'. "
+                "Each publication should be a separate entry in the list, not combined in a single string."
+            ),
+            importance=Importance.BEST_PRACTICE_VIOLATION,
+            check_function_name="check_publication_list_format",
+            object_type="NWBFile",
+            object_name="root",
+            location="/",
+        )
+    ]
+
+
+def test_check_publication_list_format_fail_comma_separated_doi_prefix():
+    """Test detection of comma-separated DOI prefixes in a single entry."""
+    nwbfile = NWBFile(
+        session_description="",
+        identifier=str(uuid4()),
+        session_start_time=datetime.now().astimezone(),
+        related_publications=["doi:10.1234/abc, doi:10.5678/def"],
+    )
+    assert check_publication_list_format(nwbfile) == [
+        InspectorMessage(
+            message=(
+                "Metadata /general/related_publications contains a comma-separated list "
+                "'doi:10.1234/abc, doi:10.5678/def'. "
+                "Each publication should be a separate entry in the list, not combined in a single string."
+            ),
+            importance=Importance.BEST_PRACTICE_VIOLATION,
+            check_function_name="check_publication_list_format",
+            object_type="NWBFile",
+            object_name="root",
+            location="/",
+        )
+    ]
+
+
+def test_check_publication_list_format_bytestring_fail():
+    """Test that bytestrings are properly decoded and checked."""
+    nwbfile = NWBFile(
+        session_description="",
+        identifier=str(uuid4()),
+        session_start_time=datetime.now().astimezone(),
+        related_publications=[b"https://doi.org/10.1234/abc,https://doi.org/10.5678/def"],
+    )
+    assert check_publication_list_format(nwbfile) == [
+        InspectorMessage(
+            message=(
+                "Metadata /general/related_publications contains a comma-separated list "
+                "'https://doi.org/10.1234/abc,https://doi.org/10.5678/def'. "
+                "Each publication should be a separate entry in the list, not combined in a single string."
+            ),
+            importance=Importance.BEST_PRACTICE_VIOLATION,
+            check_function_name="check_publication_list_format",
+            object_type="NWBFile",
+            object_name="root",
+            location="/",
+        )
+    ]
+
+
 def test_check_subject_sex():
     nwbfile = NWBFile(session_description="", identifier=str(uuid4()), session_start_time=datetime.now().astimezone())
     nwbfile.subject = Subject(subject_id="001")