udata 10.8.4.dev37375__py2.py3-none-any.whl → 10.8.4.dev37405__py2.py3-none-any.whl

udata/core/dataservices/rdf.py

@@ -14,6 +14,7 @@ from udata.rdf import (
     TAG_TO_EU_HVD_CATEGORIES,
     contact_points_from_rdf,
     contact_points_to_rdf,
+    default_lang_value,
     namespace_manager,
     rdf_value,
     remote_url_from_rdf,
@@ -38,7 +39,9 @@ def dataservice_from_rdf(
     d = graph.resource(node)
 
     dataservice.title = rdf_value(d, DCT.title)
-    dataservice.description = sanitize_html(d.value(DCT.description) or d.value(DCT.abstract))
+    dataservice.description = sanitize_html(
+        default_lang_value(d, DCT.description) or default_lang_value(d, DCT.abstract)
+    )
 
     dataservice.base_api_url = url_from_rdf(d, DCAT.endpointURL)
 

udata/core/dataset/rdf.py

@@ -40,6 +40,7 @@ from udata.rdf import (
     TAG_TO_EU_HVD_CATEGORIES,
     contact_points_from_rdf,
     contact_points_to_rdf,
+    default_lang_value,
     namespace_manager,
     rdf_unique_values,
     rdf_value,
@@ -656,7 +657,7 @@ def resource_from_rdf(graph_or_distrib, dataset=None, is_additionnal=False):
     resource.filetype = "remote"
     resource.title = title_from_rdf(distrib, url)
     resource.url = url
-    resource.description = sanitize_html(distrib.value(DCT.description))
+    resource.description = sanitize_html(default_lang_value(distrib, DCT.description))
     resource.filesize = rdf_value(distrib, DCAT.byteSize)
     resource.mime = mime_from_rdf(distrib)
     resource.format = format_from_rdf(distrib)
@@ -731,7 +732,7 @@ def dataset_from_rdf(graph: Graph, dataset=None, node=None, remote_url_prefix: s
         raise HarvestSkipException("missing title on dataset")
 
     # Support dct:abstract if dct:description is missing (sometimes used instead)
-    description = d.value(DCT.description) or d.value(DCT.abstract)
+    description = default_lang_value(d, DCT.description) or default_lang_value(d, DCT.abstract)
     dataset.description = sanitize_html(description)
     dataset.frequency = frequency_from_rdf(d.value(DCT.accrualPeriodicity)) or dataset.frequency
     roles = [  # Imbricated list of contact points for each role

udata/rdf.py

@@ -239,11 +239,28 @@ def rdf_unique_values(resource, predicate, parse_label=False) -> set[str]:
 
 
 def rdf_value(obj, predicate, default=None, parse_label=False):
-    """Serialize the value for a predicate on a RdfResource"""
-    value = obj.value(predicate)
+    """
+    Serialize the value for a predicate on a RdfResource,
+    expecting one value only or (at most) one per language for Literals.
+    """
+    value = default_lang_value(obj, predicate)
     return serialize_value(value, parse_label=parse_label) if value else default
 
 
+def default_lang_value(obj, predicate):
+    """
+    Return the value with the default language if multiple Literal values exist in different languages.
+    """
+    candidate_values = list(obj.objects(predicate))
+    if not candidate_values:
+        return None
+    for val in candidate_values:
+        if isinstance(val, Literal) and val.language == current_app.config["DEFAULT_LANGUAGE"]:
+            return val
+    # Defaulting to the first value found
+    return candidate_values[0]
+
+
 class HTMLDetector(HTMLParser):
     def __init__(self, *args, **kwargs):
         HTMLParser.__init__(self, *args, **kwargs)