Store the dataset featured field as a bool

pyproject.toml

@@ -8,7 +8,7 @@ authors = [{name = "Opendata Team", email = "[email protected]"}]
 readme = "README.md"
 dynamic = ["version", "description"]
 dependencies = [
-    "dependency-injector==4.41.0",
+    "dependency-injector==4.42.0b1",
     "elasticsearch==7.15.0",
     "elasticsearch_dsl==7.4.0",
     "factory-boy==3.2.1",
@@ -20,7 +20,7 @@ dependencies = [
     "pydantic==1.9.0",
     "pytest==6.2.5",
     "pytest-flask==1.2.0",
-    "markdown==3.3.3",
+    "markdown==3.7",
     "beautifulsoup4==4.10.0",
     # pinned to a known working version, since 3.1.3 will fail while running tests
     "werkzeug==3.0.4",

tests/test_api.py

@@ -17,7 +17,7 @@ def test_api_dataset_index_unindex(app, client, faker):
         'views': faker.random_int(),
         'followers': faker.random_int(),
         'reuses': faker.random_int(),
-        'featured': faker.random_int(min=0, max=1),
+        'featured': faker.pybool(),
         'resources_count': faker.random_int(min=1, max=15),
         'organization': {
             'id': faker.md5(),
@@ -104,7 +104,7 @@ def test_api_dataset_index_on_another_index(app, client, search_client, faker):
         'views': faker.random_int(),
         'followers': faker.random_int(),
         'reuses': faker.random_int(),
-        'featured': faker.random_int(min=0, max=1),
+        'featured': faker.pybool(),
         'resources_count': faker.random_int(min=1, max=15),
         'organization': {
             'id': faker.md5(),

tests/test_consumers.py

@@ -21,7 +21,7 @@ def test_parse_dataset_obj():
         'views': 7806,
         'followers': 72,
         'reuses': 45,
-        'featured': 0,
+        'featured': False,
         'resources_count': 10,
         'organization': {
             'id': '534fff8ea3a7292c64a77f02',

udata_search_service/domain/entities.py

@@ -49,7 +49,7 @@ class Dataset(EntityBase):
     views: int
     followers: int
     reuses: int
-    featured: int
+    featured: bool
     resources_count: int
     concat_title_org: str
     description: str

udata_search_service/infrastructure/consumers.py

@@ -33,7 +33,6 @@ def load_from_dict(cls, data):
         data["reuses"] = log2p(data.get("reuses", 0))
         data["orga_followers"] = log2p(data.get("orga_followers", 0))
         data["orga_sp"] = 4 if data.get("orga_sp", 0) else 1
-        data["featured"] = 4 if data.get("featured", 0) else 1
 
         return super().load_from_dict(data)
 

udata_search_service/infrastructure/search_clients.py

@@ -7,7 +7,7 @@
 
 from elasticsearch import Elasticsearch
 from elasticsearch.exceptions import NotFoundError
-from elasticsearch_dsl import Date, Document, Float, Integer, Keyword, Text, tokenizer, token_filter, analyzer, query
+from elasticsearch_dsl import Boolean, Date, Document, Float, Integer, Keyword, Text, tokenizer, token_filter, analyzer, query
 from elasticsearch_dsl.connections import connections
 from udata_search_service.domain.entities import Dataset, Organization, Reuse, Dataservice
 from udata_search_service.config import Config
@@ -139,7 +139,7 @@ class SearchableDataset(IndexDocument):
     views = Float()
     followers = Float()
     reuses = Float()
-    featured = Integer()
+    featured = Boolean()
     resources_count = Integer()
     concat_title_org = Text(analyzer=dgv_analyzer)
     temporal_coverage_start = Date()
@@ -272,7 +272,15 @@ def query_datasets(self, query_text: str, offset: int, page_size: int, filters:
             query.SF("field_value_factor", field="views", factor=4, modifier='sqrt', missing=1),
             query.SF("field_value_factor", field="followers", factor=4, modifier='sqrt', missing=1),
             query.SF("field_value_factor", field="orga_followers", factor=1, modifier='sqrt', missing=1),
-            query.SF("field_value_factor", field="featured", factor=1, modifier='sqrt', missing=1),
+            query.SF("script_score",
+                     script={
+                         "source": """
+                            if (doc['featured'].value) {
+                                return 2
+                            } else {
+                                return 1
+                            }
+                        """})
         ]
 
         if query_text:

udata_search_service/presentation/api.py

@@ -163,7 +163,7 @@ class DatasetToIndex(BaseModel):
     followers: int
     reuses: int
     resources_count: Optional[int] = 0
-    featured: Optional[int] = 0
+    featured: Optional[bool] = False
     format: Optional[list] = []
     schema_: Optional[list] = Field([], alias="schema")
     extras: Optional[dict] = {}