Validate date strings when transforming MITAardvark records

tests/fixtures/aardvark_records.jsonl

@@ -1,3 +1,3 @@
-{"dct_accessRights_s": "Access rights", "dct_references_s": "", "dct_title_s": "Test title 1", "gbl_mdModified_dt": "", "gbl_mdVersion_s": "", "gbl_resourceClass_sm": "", "gbl_suppressed_b": false, "id": "mit:123", "locn_geometry": ""}
-{"dct_accessRights_s": "Access rights", "dct_references_s": "", "dct_title_s": "Test title 2", "gbl_mdModified_dt": "", "gbl_mdVersion_s": "", "gbl_resourceClass_sm": "", "gbl_suppressed_b": false, "id": "ogm:456", "locn_geometry": ""}
-{"dct_accessRights_s": "Access rights", "dct_references_s": "", "dct_title_s": "Test title 3", "gbl_mdModified_dt": "", "gbl_mdVersion_s": "", "gbl_resourceClass_sm": "", "gbl_suppressed_b": true, "id": "ogm:789", "locn_geometry": ""}
+{"dct_accessRights_s": "Access rights", "dct_references_s": "{\"http://schema.org/url\": \"https://geodata.libraries.mit.edu/record/abc:123\"}", "dct_title_s": "Test title 1", "gbl_mdModified_dt": "", "gbl_mdVersion_s": "", "gbl_resourceClass_sm": "", "gbl_suppressed_b": false, "id": "mit:123", "locn_geometry": ""}
+{"dct_accessRights_s": "Access rights", "dct_references_s": "{\"http://schema.org/url\": \"https://geodata.libraries.mit.edu/record/abc:123\"}", "dct_title_s": "Test title 2", "gbl_mdModified_dt": "", "gbl_mdVersion_s": "", "gbl_resourceClass_sm": "", "gbl_suppressed_b": false, "id": "ogm:456", "locn_geometry": ""}
+{"dct_accessRights_s": "Access rights", "dct_references_s": "{\"http://schema.org/url\": \"https://geodata.libraries.mit.edu/record/abc:123\"}", "dct_title_s": "Test title 3", "gbl_mdModified_dt": "", "gbl_mdVersion_s": "", "gbl_resourceClass_sm": "", "gbl_suppressed_b": true, "id": "ogm:789", "locn_geometry": ""}

tests/sources/json/test_aardvark.py

@@ -1,3 +1,4 @@
+import json
 from pathlib import Path
 
 import pytest
@@ -32,7 +33,7 @@ def test_aardvark_get_required_fields_returns_expected_values(aardvark_records):
     transformer = MITAardvark("cool-repo", aardvark_records)
     assert transformer.get_required_fields(next(aardvark_records)) == {
         "source": "A Cool Repository",
-        "source_link": "https://example.com/gismit:123",
+        "source_link": "https://geodata.libraries.mit.edu/record/abc:123",
         "timdex_record_id": "cool-repo:123",
         "title": "Test title 1",
     }
@@ -42,12 +43,21 @@ def test_aardvark_transform_returns_timdex_record(aardvark_records):
     transformer = MITAardvark("cool-repo", aardvark_records)
     assert next(transformer) == timdex.TimdexRecord(
         source="A Cool Repository",
-        source_link="https://example.com/gismit:123",
+        source_link="https://geodata.libraries.mit.edu/record/abc:123",
         timdex_record_id="cool-repo:123",
         title="Test title 1",
-        citation="Test title 1. Geospatial data. https://example.com/gismit:123",
+        citation="Test title 1. Geospatial data. "
+        "https://geodata.libraries.mit.edu/record/abc:123",
         content_type=["Geospatial data"],
         rights=[timdex.Rights(description="Access rights", kind="Access")],
+        links=[
+            timdex.Link(
+                url="https://geodata.libraries.mit.edu/record/abc:123",
+                kind="Website",
+                restrictions=None,
+                text="Website",
+            )
+        ],
     )
 
 
@@ -137,6 +147,39 @@ def test_aardvark_get_dates_success(aardvark_record_all_fields):
     ]
 
 
+def test_aardvark_get_dates_drops_dates_with_invalid_strings(
+    caplog, aardvark_record_all_fields
+):
+    caplog.set_level("DEBUG")
+    record = next(aardvark_record_all_fields)
+    record["dct_issued_s"] = "1933?"  # dropped
+    record["dct_temporal_sm"] = [
+        "2000-01-01",
+        "1999",
+        "approximately 1569",  # dropped
+        "absolute junky date",  # dropped
+    ]
+    record["gbl_dateRange_drsim"] = [
+        "[1943 TO 1946]",
+        "[apples TO oranges]",  # logged and dropped
+    ]
+    assert MITAardvark.get_dates(record, "123") == [
+        timdex.Date(kind="Coverage", note=None, range=None, value="2000-01-01"),
+        timdex.Date(kind="Coverage", note=None, range=None, value="1999"),
+        timdex.Date(kind="Coverage", note=None, range=None, value="1943"),
+        timdex.Date(kind="Coverage", note=None, range=None, value="1944"),
+        timdex.Date(kind="Coverage", note=None, range=None, value="1945"),
+        timdex.Date(kind="Coverage", note=None, range=None, value="1946"),
+        timdex.Date(
+            kind="Coverage",
+            note=None,
+            range=timdex.DateRange(gt=None, gte="1943", lt=None, lte="1946"),
+            value=None,
+        ),
+    ]
+    assert "Unable to parse date range string" in caplog.text
+
+
 def test_aardvark_parse_solr_date_range_string_success():
     assert MITAardvark.parse_solr_date_range_string("[1932 TO 1937]", "123") == (
         "1932",
@@ -277,3 +320,39 @@ def test_aardvark_get_subjects_success(aardvark_record_all_fields):
         timdex.Subject(value=["Dataset"], kind="Subject scheme not provided"),
         timdex.Subject(value=["Vector data"], kind="Subject scheme not provided"),
     ]
+
+
+def test_aardvark_record_get_source_link_success(
+    aardvark_record_all_fields,
+):
+    record = next(aardvark_record_all_fields)
+    url_from_aardvark_record = "https://geodata.libraries.mit.edu/record/abc:123"
+    record["dct_references_s"] = json.dumps(
+        {"http://schema.org/url": url_from_aardvark_record}
+    )
+    assert (
+        MITAardvark.get_source_link(
+            "None",
+            "abc:123",
+            record,
+        )
+        == url_from_aardvark_record
+    )
+
+
+def test_aardvark_record_get_source_link_bad_dct_references_s_raises_error(
+    aardvark_record_all_fields,
+):
+    record = next(aardvark_record_all_fields)
+    record["dct_references_s"] = json.dumps(
+        {"missing data": "from aardvark from geoharvester"}
+    )
+    with pytest.raises(
+        ValueError,
+        match="Could not locate a kind=Website link to pull the source link from.",
+    ):
+        MITAardvark.get_source_link(
+            "None",
+            "abc:123",
+            record,
+        )

transmogrifier/config.py

@@ -106,13 +106,13 @@
     },
     "gismit": {
         "name": "MIT GIS Resources",
-        "base-url": "https://search.libraries.mit.edu/record/",
+        "base-url": "None",
         "transform-class": "transmogrifier.sources.json.aardvark.MITAardvark",
     },
     "gisogm": {
         "name": "OpenGeoMetadata GIS Resources",
-        "base-url": "https://search.libraries.mit.edu/record/",
-        "transform-class": "transmogrifier.sources.json.aardvark.OGMAardvark",
+        "base-url": "None",
+        "transform-class": "transmogrifier.sources.json.aardvark.MITAardvark",
     },
     "researchdatabases": {
         "name": "Research Databases",

transmogrifier/sources/json/aardvark.py

@@ -3,6 +3,7 @@
 import re
 
 import transmogrifier.models as timdex
+from transmogrifier.helpers import validate_date
 from transmogrifier.sources.transformer import JSON, JSONTransformer
 
 logger = logging.getLogger(__name__)
@@ -31,25 +32,33 @@ def get_main_titles(cls, source_record: dict) -> list[str]:
 
     @classmethod
     def get_source_link(
-        cls, source_base_url: str, source_record_id: str, source_record: dict[str, JSON]
+        cls,
+        _source_base_url: str,
+        source_record_id: str,
+        source_record: dict[str, JSON],
     ) -> str:
         """
         Class method to set the source link for the item.
 
         May be overridden by source subclasses if needed.
 
-        Default behavior is to concatenate the source base URL + source record id.
+        Unlike other Transmogrifier sources that dynamically build a source link,
+        MITAardvark files are expected to have a fully formed and appropriate source link
+        in the metadata already.  This method relies on that data.
 
         Args:
-            source_base_url: Source base URL.
+            _source_base_url: Source base URL.  Not used for MITAardvark transforms.
             source_record_id: Record identifier for the source record.
             source_record: A BeautifulSoup Tag representing a single XML record.
                 - not used by default implementation, but could be useful for subclass
                     overrides
         """
-        return source_base_url + cls.get_timdex_record_id(
-            "gismit", source_record_id, source_record
-        )
+        links = cls.get_links(source_record, source_record_id)
+        url_links = [link for link in links if link.kind == "Website"]
+        if len(url_links) == 1:
+            return url_links[0].url
+        message = "Could not locate a kind=Website link to pull the source link from."
+        raise ValueError(message)
 
     @classmethod
     def get_timdex_record_id(
@@ -183,12 +192,25 @@ def get_contributors(source_record: dict) -> list[timdex.Contributor]:
 
     @classmethod
     def get_dates(cls, source_record: dict, source_record_id: str) -> list[timdex.Date]:
-        """Get values from source record for TIMDEX dates field."""
-        return (
+        """Get values from source record for TIMDEX dates field.
+
+        This method aggregates dates from a variety of Aardvark fields.  Once aggregated,
+        the results are filtered to allow only well formed DateRanges or validated date
+        strings.
+        """
+        dates = (
             cls._issued_dates(source_record)
             + cls._coverage_dates(source_record)
             + cls._range_dates(source_record, source_record_id)
         )
+        return [
+            date
+            for date in dates
+            # skip value validation for DateRange type dates
+            if isinstance(date.range, timdex.DateRange)
+            # validate date string if not None
+            or (date.value is not None and validate_date(date.value, source_record_id))
+        ]
 
     @classmethod
     def _issued_dates(cls, source_record: dict) -> list[timdex.Date]:
@@ -228,9 +250,13 @@ def _range_dates(
         """Get values for issued dates."""
         range_dates = []
         for date_range_string in source_record.get("gbl_dateRange_drsim", []):
-            date_range_values = cls.parse_solr_date_range_string(
-                date_range_string, source_record_id
-            )
+            try:
+                date_range_values = cls.parse_solr_date_range_string(
+                    date_range_string, source_record_id
+                )
+            except ValueError as exc:
+                logger.warning(exc)
+                continue
             range_dates.append(
                 timdex.Date(
                     kind="Coverage",
@@ -292,6 +318,7 @@ def get_links(source_record: dict, source_record_id: str) -> list[timdex.Link]:
                         url=link.get("url"), kind="Download", text=link.get("label")
                     )
                     for link in links_object.get("http://schema.org/downloadUrl", [])
+                    if isinstance(link.get("url", {}), str)
                 ]
             )
             if schema_url := links_object.get("http://schema.org/url"):