add include_in_payload to optimize rest queries where we dont really …

docker-compose.yml

@@ -175,7 +175,7 @@ services:
         aliases:
           - "nfcomposes3"
     ports:
-      - "127.0.0.1:${SEAWEEDFS_DEV_OUTSIDE_PORT:-8001}:8000"
+      - "127.0.0.1:${SEAWEEDFS_DEV_OUTSIDE_PORT:-8000}:8000"
     volumes:
       - "./devenv/s3_config/s3.json:/etc/seaweedfs/s3.json"
       - "seaweedfs_buckets:/data"

skipper/skipper/dataseries/storage/contract/base.py

@@ -16,7 +16,7 @@
 from rest_framework.generics import get_object_or_404
 from rest_framework.request import Request
 from rest_framework.serializers import SerializerMetaclass
-from typing import Dict, Any, Optional, Union, Tuple
+from typing import Dict, Any, Optional, Union, Tuple, List
 from uuid import UUID
 
 from skipper.core.serializers.base import BaseSerializer
@@ -244,7 +244,7 @@ def impl_internal_to_representation(
             self,
             data_point: Any,
             data_series: DataSeries,
-            external_id_as_dimension_identifier: bool
+            external_id_as_dimension_identifier: bool,
     ) -> Dict[str, Any]:
         raise NotImplementedError()
 

skipper/skipper/dataseries/storage/contract/view.py

@@ -53,6 +53,8 @@ def get_changes_since(self) -> Optional[datetime.datetime]: ...
 
     def should_include_versions(self) -> bool: ...
 
+    def get_include_in_payload(self) -> Optional[List[str]]: ...
+
 
 class StorageViewAdapter(metaclass=ABCMeta):
     """

skipper/skipper/dataseries/storage/dynamic_sql/queries/create_view.py

@@ -86,6 +86,7 @@ def create_view_as(
 
             data_sql = data_series_as_sql_table(
                 data_series,
+                include_in_payload=None,
                 data_series_query_info=data_series_query_info,
                 resolve_dimension_external_ids=identify_dimensions_by_external_id
             )

skipper/skipper/dataseries/storage/dynamic_sql/queries/display.py

@@ -80,6 +80,7 @@ def render_pagination_data_select(include_pagination_data: bool) -> str:
 
 
 def _data_series_as_sql_table(
+        include_in_payload: Optional[List[str]],
         payload_as_json: bool,
         point_in_time: bool,
         changes_since: bool,
@@ -101,6 +102,9 @@ def _data_series_as_sql_table(
                                                     and len(_data_series_query_info.dimensions.keys()) > 0
 
     all_select_infos = select_infos(_data_series_query_info)
+    if include_in_payload is not None:
+        all_select_infos = [select_info for select_info in all_select_infos if select_info.unescaped_display_id in include_in_payload]
+
     central_table_sql = single_data_series_as_sql_table(
         all_select_infos=all_select_infos,
         data_series_query_info=_data_series_query_info,
@@ -222,14 +226,15 @@ def _data_series_as_sql_table(
 
 def data_series_as_sql_table(
         data_series: DataSeries,
+        include_in_payload: List[str],
         payload_as_json: bool = False,
         point_in_time: bool = False,
         changes_since: bool = False,
         include_versions: bool = False,
         filter_str: str = '',
         resolve_dimension_external_ids: bool = False,
         data_series_query_info: Optional[DataSeriesQueryInfo] = None,
-        use_materialized: Optional[bool] = None
+        use_materialized: Optional[bool] = None,
 ) -> str:
     _data_series_query_info: DataSeriesQueryInfo
     if data_series_query_info is None:
@@ -238,6 +243,7 @@ def data_series_as_sql_table(
         _data_series_query_info = data_series_query_info
 
     sql = _data_series_as_sql_table(
+        include_in_payload=include_in_payload,
         payload_as_json=payload_as_json,
         point_in_time=point_in_time,
         changes_since=changes_since,

skipper/skipper/dataseries/storage/dynamic_sql/queries/materialize_view.py

@@ -56,7 +56,7 @@ def materialize_view_as(schema_name: str, overwrite: bool, data_series_id: str,
                 )
             if not exists_already:
                 query = f"""CREATE MATERIALIZED VIEW IF NOT EXISTS {schema_name}.{escaped_view_name}
-                    AS {data_series_as_sql_table(data_series)}"""
+                    AS {data_series_as_sql_table(data_series, include_in_payload=None)}"""
                 query_params: Dict[str, Any] = {select_info.payload_variable_name: select_info.unescaped_display_id for
                                                 select_info in select_infos(compute_data_series_query_info(data_series))}
                 cursor.execute(

skipper/skipper/dataseries/storage/dynamic_sql/serializers/modification.py

@@ -41,10 +41,10 @@ def impl_internal_to_representation(
         query_params: Dict[str, Any] = {select_info.payload_variable_name: select_info.unescaped_display_id for
                                         select_info in
                                         select_infos(self.get_data_series_children_query_info())}
-        _view = self.context.get('view')
 
         query_str = data_series_as_sql_table(
             data_series=data_series,
+            include_in_payload=None,
             payload_as_json=True,
             point_in_time=self.point_in_time is not None,
             include_versions=self.should_include_versions,

skipper/skipper/dataseries/storage/dynamic_sql/storage_view_adapter.py

@@ -93,6 +93,7 @@ def precompute_filter_part(
 
 
 def raw_display_data_point_query(
+        include_in_payload: Optional[List[str]],
         filter_value: Dict[str, Any],
         data_series: DataSeries,
         external_id_as_dimension_identifier: bool,
@@ -205,6 +206,7 @@ def raw_display_data_point_query(
         query_params['changes_since'] = changes_since
 
     query_str = data_series_as_sql_table(
+        include_in_payload=include_in_payload,
         data_series=data_series_obj,
         payload_as_json=True,
         point_in_time=is_point_in_time,
@@ -255,6 +257,7 @@ def access_object(
             return obj
 
         raw_query = raw_display_data_point_query(
+            include_in_payload=view.get_include_in_payload(),
             data_point_id=data_point_id,
             filter_value=view.get_filter_value(),
             data_series=view.access_data_series(),
@@ -332,6 +335,7 @@ def encode_last_id_for_pagination(self, view: BaseDataSeries_DataPointViewSet, d
     def get_next_page_query_for_pagination(self, view: BaseDataSeries_DataPointViewSet, last_query: str, limit: int,
                                            request: HttpRequest) -> RawQuerySet:  # type: ignore
         return raw_display_data_point_query(
+            include_in_payload=view.get_include_in_payload(), # we actually don't need anything in the payload here
             start_object=last_query,
             limit=limit,
             filter_value=view.get_filter_value(),
@@ -353,6 +357,7 @@ def get_prev_page_query_for_pagination(
     ) -> Optional[RawQuerySet]:  # type: ignore
         # FIXME: this could be a simpler query without as many joins
         return raw_display_data_point_query(
+            include_in_payload=view.get_include_in_payload(), # we actually don't need anything in the payload here
             start_object=last_query,
             reverse=True,
             limit=limit,

skipper/skipper/dataseries/tests/data_point/test_data_point_non_historical_lifecycle.py

@@ -156,6 +156,7 @@ class FakedFile:
     )
 
     _sql = data_series_as_sql_table(
+        include_in_payload=None,
         data_series=_data_series_obj,
         payload_as_json=True,
         point_in_time=False,

skipper/skipper/dataseries/views/datapoint/crud.py

@@ -198,6 +198,17 @@ def get_filter_value(self) -> Dict[str, Any]:
             else:
                 filter_value = {}
             return filter_value
+
+        def get_include_in_payload(self) -> Optional[List[str]]:
+            if 'include_in_payload' in self.request.GET:
+                _include_in_payload_str = self.request.GET['include_in_payload']
+                if _include_in_payload_str != None:
+                    return list(_include_in_payload_str.split(','))
+                else:
+                    return None
+            else:
+                return None
+
 
         def get_external_ids(self) -> Optional[List[str]]:
             return self.request.GET.getlist('external_id') if 'external_id' in self.request.GET else None  # type: ignore
@@ -275,6 +286,7 @@ def get_description_string(self) -> str:
             - count[=true] <br>
             - external_id=<str> (repeatable) <br>
             - identify_dimensions_by_external_id[=true] <br>
+            - include_in_payload=<str> (comma separated list of dimension/fact external ids to include in the payload) <br>
             """
             if _history:
                 doc_string = f"""

skipper/skipper/dataseries/views/metamodel/cube_sql.py

@@ -121,6 +121,7 @@ def get(self, request: Request, **kwargs: str) -> HttpResponse:
 
                 sql_template = data_series_as_sql_table(
                     data_series,
+                    include_in_payload=None,
                     payload_as_json=payload_as_json(request),
                     resolve_dimension_external_ids=use_external_id_as_dimension_identifier(
                         cast(Dict[str, Any], request.GET)