Reworked _add_extra_field function to properly insert hvd categories and appl. legislation

ckanext/odsh/profiles/odsh_dcat_de_profile.py

@@ -11,14 +11,18 @@ from ckanext.dcatde.profiles import DCATdeProfile, DCATDE, DCAT, DCATDE_1_0, DCA
 
 import ckanext.odsh.helpers as helpers_odsh
 import ckanext.odsh.collection.helpers as helpers_collection
-import ast
+import logging
+
+log = logging.getLogger(__name__)
 
 DCT = rdflib.namespace.Namespace("http://purl.org/dc/terms/")
 DCAT = rdflib.namespace.Namespace("http://www.w3.org/ns/dcat#")
 DCATAP = rdflib.namespace.Namespace("http://data.europa.eu/r5r/")
-DCATDE_1_0_2 = rdflib.namespace.Namespace("http://dcat-ap.de/def/dcatde/1.0.2/")
+DCATDE_1_0_2 = rdflib.namespace.Namespace(
+    "http://dcat-ap.de/def/dcatde/1.0.2/")
 ADMS = rdflib.namespace.Namespace("http://www.w3.org/ns/adms#")
 
+
 class ODSHDCATdeProfile(DCATdeProfile):
 
     # from RDF
@@ -64,12 +68,14 @@ class ODSHDCATdeProfile(DCATdeProfile):
     def _parse_applicable_legislation(self, dataset_dict, dataset_ref):
         value = self._object(dataset_ref, DCT.applicableLegislation)
         if value:
-            ds_utils.insert_new_extras_field(dataset_dict, 'applicableLegislation', value)
+            ds_utils.insert_new_extras_field(
+                dataset_dict, 'applicableLegislation', value)
 
     def _parse_hvd_category(self, dataset_dict, dataset_ref):
         value = self._object(dataset_ref, DCT.hvdCategory)
         if value:
-            ds_utils.insert_new_extras_field(dataset_dict, 'hvdCategory', value)
+            ds_utils.insert_new_extras_field(
+                dataset_dict, 'hvdCategory', value)
 
     def _belongs_to_collection(self, dataset_dict, dataset_ref):
         dct_is_version_of = self._object(dataset_ref, DCT.isVersionOf)
@@ -94,9 +100,12 @@ class ODSHDCATdeProfile(DCATdeProfile):
         self._add_license_attribution_by_text(dataset_dict, dataset_ref)
         self._add_type(dataset_dict, dataset_ref)
         self._add_modified_and_issued(dataset_dict, dataset_ref)
-        self._add_extra_field(dataset_dict, dataset_ref, 'reference', DCT.references)
-        self._add_extra_field(dataset_dict, dataset_ref, 'applicableLegislation', DCATAP.applicableLegislation)
-        self._add_extra_field(dataset_dict, dataset_ref, 'hvdCategory', DCATAP.hvdCategory)
+        self._add_extra_field(dataset_dict, dataset_ref,
+                              'reference', DCT.references)
+        self._add_extra_field(dataset_dict, dataset_ref,
+                              'applicableLegislation', DCATAP.applicableLegislation)
+        self._add_extra_field(dataset_dict, dataset_ref,
+                              'hvdCategory', DCATAP.hvdCategory)
         self._add_version(dataset_dict, dataset_ref)
         if self._is_dataset_collection(dataset_dict):
             self._remove_predefined_collection_members()
@@ -126,29 +135,56 @@ class ODSHDCATdeProfile(DCATdeProfile):
                      rdflib.Literal(licenseAttributionByText))
                 )
 
-    def _add_extra_field(self, dataset_dict, dataset_ref, field_name, dcat_property):
-        field_value_str = helpers_odsh.odsh_extract_value_from_extras(dataset_dict.get('extras'), field_name)
+    def _add_extra_field(self, dataset_dict, dataset_ref, field_name, rdf_property):
+        '''
+        Adds extra fields from a dataset dictionary to a RDF graph.
+
+        Parameters:
+        - dataset_dict: A dictionary containing dataset information.
+        - dataset_ref: The reference to the dataset in the RDF graph.
+        - field_name: The name of the field to extract from the dataset dictionary.
+        - rdf_property: The DCAT property to which the field value corresponds.
+
+        This function extracts the value of the specified field from the dataset dictionary,
+        and adds it to the RDF graph with the given dataset reference and DCAT property.
+        If the field value is a list of values represented as a string enclosed in curly braces,
+        it splits the string and adds each value individually to the graph.
+        If the field value is a single value, it is added directly to the graph.
+
+        Note: This function includes a workaround since CKAN's extras do not natively support
+        lists. It treats lists represented as strings enclosed in curly braces as a single string.
+        '''
+        field_value_str = dataset_dict.get(field_name)
+
+        if not field_value_str:
+            return
 
         try:
-            # Attempt to parse the string as a Python literal
-            field_value = ast.literal_eval(field_value_str)
-            if not isinstance(field_value, set):
-                # If it's not a set, treat it as a single value
-                raise ValueError("Not a set")
-        except (ValueError, SyntaxError):
-            # If parsing fails, treat the entire string as a single value
-            field_value = field_value_str
+            # Remove curly braces from the string
+            values_string = field_value_str.strip('{}')
 
-        if isinstance(field_value, set):
-            for value in field_value:
-                self.g.set(
-                    (dataset_ref, dcat_property, rdflib.URIRef(value))
+            # Split the string by commas
+            field_values = values_string.split(',')
+
+            # Trim any extra whitespace from each URL
+            field_values = [value.strip() for value in field_values]
+
+            # Check if the field value is a list
+            if len(field_values) > 1:
+                for value in field_values:
+                    self.g.add(
+                        (dataset_ref, rdf_property, rdflib.URIRef(value))
                     )
             else:
+                # Treat it as a single value
                 self.g.set(
-                (dataset_ref, dcat_property, rdflib.URIRef(field_value))
+                    (dataset_ref, rdf_property, rdflib.URIRef(field_values[0]))
+                )
+        except (ValueError, SyntaxError):
+            # If parsing fails, treat the entire string as a single value
+            self.g.set(
+                (dataset_ref, rdf_property, rdflib.URIRef(field_value_str))
             )
-
 
     def _add_modified_and_issued(self, dataset_dict, dataset_ref):
         '''
@@ -176,7 +212,8 @@ class ODSHDCATdeProfile(DCATdeProfile):
         '''
         Adds CKAN isReplacedBy extra field to dublin core isReplacedBy and sets version information.
         '''
-        new_version = helpers_odsh.odsh_extract_value_from_extras(dataset_dict.get('extras'), 'is_replaced_by')
+        new_version = helpers_odsh.odsh_extract_value_from_extras(
+            dataset_dict.get('extras'), 'is_replaced_by')
         if new_version:
             self.g.set(
                 (dataset_ref, DCTERMS.isReplacedBy,
@@ -184,7 +221,8 @@ class ODSHDCATdeProfile(DCATdeProfile):
                  )
             )
 
-        version_notes = helpers_odsh.odsh_extract_value_from_extras(dataset_dict.get('extras'), 'version_notes')
+        version_notes = helpers_odsh.odsh_extract_value_from_extras(
+            dataset_dict.get('extras'), 'version_notes')
         if version_notes:
             self.g.set(
                 (dataset_ref, ADMS.versionNotes,
@@ -194,8 +232,7 @@ class ODSHDCATdeProfile(DCATdeProfile):
 
     def _add_type(self, dataset_dict, dataset_ref):
         '''
-        adds the type if there is a known mapping from ckan type to
-        dct:type
+        Adds the type if there is a known mapping from ckan type to dct:type
         '''
         ckan_type = self._get_ckan_type(dataset_dict)
         dct_type = helpers_odsh.map_ckan_type_to_dct_type(ckan_type)