[Backport 8.x] Correct typing hints for the FunctionScore query

elasticsearch_dsl/query.py

@@ -612,9 +612,9 @@ class FunctionScore(Query):
 
     name = "function_score"
     _param_defs = {
+        "functions": {"type": "score_function", "multi": True},
         "query": {"type": "query"},
         "filter": {"type": "query"},
-        "functions": {"type": "score_function", "multi": True},
     }
 
     def __init__(
@@ -623,11 +623,7 @@ def __init__(
         boost_mode: Union[
             Literal["multiply", "replace", "sum", "avg", "max", "min"], "DefaultType"
         ] = DEFAULT,
-        functions: Union[
-            Sequence["types.FunctionScoreContainer"],
-            Sequence[Dict[str, Any]],
-            "DefaultType",
-        ] = DEFAULT,
+        functions: Union[Sequence[ScoreFunction], "DefaultType"] = DEFAULT,
         max_boost: Union[float, "DefaultType"] = DEFAULT,
         min_score: Union[float, "DefaultType"] = DEFAULT,
         query: Union[Query, "DefaultType"] = DEFAULT,

elasticsearch_dsl/types.py

@@ -19,7 +19,7 @@
 
 from elastic_transport.client_utils import DEFAULT, DefaultType
 
-from elasticsearch_dsl import Query, function
+from elasticsearch_dsl import Query
 from elasticsearch_dsl.document_base import InstrumentedField
 from elasticsearch_dsl.utils import AttrDict
 
@@ -688,74 +688,6 @@ def __init__(
         super().__init__(kwargs)
 
 
-class FunctionScoreContainer(AttrDict[Any]):
-    """
-    :arg exp: Function that scores a document with a exponential decay,
-        depending on the distance of a numeric field value of the document
-        from an origin.
-    :arg gauss: Function that scores a document with a normal decay,
-        depending on the distance of a numeric field value of the document
-        from an origin.
-    :arg linear: Function that scores a document with a linear decay,
-        depending on the distance of a numeric field value of the document
-        from an origin.
-    :arg field_value_factor: Function allows you to use a field from a
-        document to influence the score. It’s similar to using the
-        script_score function, however, it avoids the overhead of
-        scripting.
-    :arg random_score: Generates scores that are uniformly distributed
-        from 0 up to but not including 1. In case you want scores to be
-        reproducible, it is possible to provide a `seed` and `field`.
-    :arg script_score: Enables you to wrap another query and customize the
-        scoring of it optionally with a computation derived from other
-        numeric field values in the doc using a script expression.
-    :arg filter:
-    :arg weight:
-    """
-
-    exp: Union[function.DecayFunction, DefaultType]
-    gauss: Union[function.DecayFunction, DefaultType]
-    linear: Union[function.DecayFunction, DefaultType]
-    field_value_factor: Union[function.FieldValueFactorScore, DefaultType]
-    random_score: Union[function.RandomScore, DefaultType]
-    script_score: Union[function.ScriptScore, DefaultType]
-    filter: Union[Query, DefaultType]
-    weight: Union[float, DefaultType]
-
-    def __init__(
-        self,
-        *,
-        exp: Union[function.DecayFunction, DefaultType] = DEFAULT,
-        gauss: Union[function.DecayFunction, DefaultType] = DEFAULT,
-        linear: Union[function.DecayFunction, DefaultType] = DEFAULT,
-        field_value_factor: Union[
-            function.FieldValueFactorScore, DefaultType
-        ] = DEFAULT,
-        random_score: Union[function.RandomScore, DefaultType] = DEFAULT,
-        script_score: Union[function.ScriptScore, DefaultType] = DEFAULT,
-        filter: Union[Query, DefaultType] = DEFAULT,
-        weight: Union[float, DefaultType] = DEFAULT,
-        **kwargs: Any,
-    ):
-        if exp is not DEFAULT:
-            kwargs["exp"] = exp
-        if gauss is not DEFAULT:
-            kwargs["gauss"] = gauss
-        if linear is not DEFAULT:
-            kwargs["linear"] = linear
-        if field_value_factor is not DEFAULT:
-            kwargs["field_value_factor"] = field_value_factor
-        if random_score is not DEFAULT:
-            kwargs["random_score"] = random_score
-        if script_score is not DEFAULT:
-            kwargs["script_score"] = script_score
-        if filter is not DEFAULT:
-            kwargs["filter"] = filter
-        if weight is not DEFAULT:
-            kwargs["weight"] = weight
-        super().__init__(kwargs)
-
-
 class FuzzyQuery(AttrDict[Any]):
     """
     :arg value: (required) Term you wish to find in the provided field.

tests/test_query.py

@@ -562,6 +562,33 @@ def test_function_score_to_dict() -> None:
     assert d == q.to_dict()
 
 
+def test_function_score_class_based_to_dict() -> None:
+    q = query.FunctionScore(
+        query=query.Match(title="python"),
+        functions=[
+            function.RandomScore(),
+            function.FieldValueFactor(
+                field="comment_count",
+                filter=query.Term(tags="python"),
+            ),
+        ],
+    )
+
+    d = {
+        "function_score": {
+            "query": {"match": {"title": "python"}},
+            "functions": [
+                {"random_score": {}},
+                {
+                    "filter": {"term": {"tags": "python"}},
+                    "field_value_factor": {"field": "comment_count"},
+                },
+            ],
+        }
+    }
+    assert d == q.to_dict()
+
+
 def test_function_score_with_single_function() -> None:
     d = {
         "function_score": {

utils/generator.py

@@ -200,6 +200,12 @@ def get_python_type(self, schema_type, for_response=False):
             ):
                 # QueryContainer maps to the DSL's Query class
                 return "Query", {"type": "query"}
+            elif (
+                type_name["namespace"] == "_types.query_dsl"
+                and type_name["name"] == "FunctionScoreContainer"
+            ):
+                # FunctionScoreContainer maps to the DSL's ScoreFunction class
+                return "ScoreFunction", {"type": "score_function"}
             elif (
                 type_name["namespace"] == "_types.aggregations"
                 and type_name["name"] == "Buckets"

utils/templates/query.py.tpl

@@ -174,7 +174,6 @@ class {{ k.name }}({{ parent }}):
            shortcut property. Until the code generator can support shortcut
            properties directly that solution is added here #}
         "filter": {"type": "query"},
-        "functions": {"type": "score_function", "multi": True},
     {% endif %}
     }
     {% endif %}

utils/templates/types.py.tpl

@@ -20,7 +20,7 @@ from typing import Any, Dict, Literal, Mapping, Sequence, Union
 from elastic_transport.client_utils import DEFAULT, DefaultType
 
 from elasticsearch_dsl.document_base import InstrumentedField
-from elasticsearch_dsl import function, Query
+from elasticsearch_dsl import Query
 from elasticsearch_dsl.utils import AttrDict
 
 PipeSeparatedFlags = str