Fix allOf partial override to inherit parent constraints and add --allof-merge-mode option (#2671)

* Fix type inheritance for nested array items in allOf partial overrides

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* docs: update command help in README

🤖 Generated by GitHub Actions

* feat: implement GraphQL schema parser to generate Python data models

* Fix tags initialization to use set literals for unique items in allOf partial overrides

* Fix tags initialization to use set literals for unique items in allOf partial overrides

* Add tests for OpenAPI allOf with parent bool property and multiple parents with same property

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

Author: 3 people

README.md

@@ -362,6 +362,12 @@ Options:
   --url URL             Input file URL. `--input` is ignored when `--url` is used
 
 Typing customization:
+  --allof-merge-mode {constraints,all,none}
+                        Mode for field merging in allOf schemas. ''constraints'': merge only
+                        constraints (minItems, maxItems, pattern, etc.) from parent
+                        (default). ''all'': merge constraints plus annotations (default,
+                        examples) from parent. ''none'': do not merge any fields from parent
+                        properties.
   --base-class BASE_CLASS
                         Base Class (default: pydantic.BaseModel)
   --disable-future-imports

docs/index.md

@@ -354,6 +354,12 @@ Options:
   --url URL             Input file URL. `--input` is ignored when `--url` is used
 
 Typing customization:
+  --allof-merge-mode {constraints,all,none}
+                        Mode for field merging in allOf schemas. ''constraints'': merge only
+                        constraints (minItems, maxItems, pattern, etc.) from parent
+                        (default). ''all'': merge constraints plus annotations (default,
+                        examples) from parent. ''none'': do not merge any fields from parent
+                        properties.
   --base-class BASE_CLASS
                         Base Class (default: pydantic.BaseModel)
   --disable-future-imports

src/datamodel_code_generator/__init__.py

@@ -270,6 +270,19 @@ class AllExportsCollisionStrategy(Enum):
     FullPrefix = "full-prefix"
 
 
+class AllOfMergeMode(Enum):
+    """Mode for field merging in allOf schemas.
+
+    constraints: Merge only constraint fields (minItems, maxItems, pattern, etc.) from parent.
+    all: Merge constraints plus annotation fields (default, examples) from parent.
+    none: Do not merge any fields from parent properties.
+    """
+
+    Constraints = "constraints"
+    All = "all"
+    NoMerge = "none"
+
+
 class GraphQLScope(Enum):
     """Scopes for GraphQL model generation."""
 
@@ -415,6 +428,7 @@ def generate(  # noqa: PLR0912, PLR0913, PLR0914, PLR0915
     use_title_as_name: bool = False,
     use_operation_id_as_name: bool = False,
     use_unique_items_as_set: bool = False,
+    allof_merge_mode: AllOfMergeMode = AllOfMergeMode.Constraints,
     http_headers: Sequence[tuple[str, str]] | None = None,
     http_ignore_tls: bool = False,
     use_annotated: bool = False,
@@ -657,6 +671,7 @@ def get_header_and_first_line(csv_file: IO[str]) -> dict[str, Any]:
         use_title_as_name=use_title_as_name,
         use_operation_id_as_name=use_operation_id_as_name,
         use_unique_items_as_set=use_unique_items_as_set,
+        allof_merge_mode=allof_merge_mode,
         http_headers=http_headers,
         http_ignore_tls=http_ignore_tls,
         use_annotated=use_annotated,

src/datamodel_code_generator/__main__.py

@@ -24,6 +24,7 @@
     DEFAULT_SHARED_MODULE_NAME,
     AllExportsCollisionStrategy,
     AllExportsScope,
+    AllOfMergeMode,
     DataclassArguments,
     DataModelType,
     Error,
@@ -421,6 +422,7 @@ def validate_all_exports_collision_strategy(cls, values: dict[str, Any]) -> dict
     use_title_as_name: bool = False
     use_operation_id_as_name: bool = False
     use_unique_items_as_set: bool = False
+    allof_merge_mode: AllOfMergeMode = AllOfMergeMode.Constraints
     http_headers: Optional[Sequence[tuple[str, str]]] = None  # noqa: UP045
     http_ignore_tls: bool = False
     use_annotated: bool = False
@@ -867,6 +869,7 @@ def main(args: Sequence[str] | None = None) -> Exit:  # noqa: PLR0911, PLR0912,
             use_title_as_name=config.use_title_as_name,
             use_operation_id_as_name=config.use_operation_id_as_name,
             use_unique_items_as_set=config.use_unique_items_as_set,
+            allof_merge_mode=config.allof_merge_mode,
             http_headers=config.http_headers,
             http_ignore_tls=config.http_ignore_tls,
             use_annotated=config.use_annotated,

src/datamodel_code_generator/arguments.py

@@ -18,6 +18,7 @@
     DEFAULT_SHARED_MODULE_NAME,
     AllExportsCollisionStrategy,
     AllExportsScope,
+    AllOfMergeMode,
     DataclassArguments,
     DataModelType,
     InputFileType,
@@ -422,6 +423,15 @@ def start_section(self, heading: str | None) -> None:
     action="store_true",
     default=None,
 )
+typing_options.add_argument(
+    "--allof-merge-mode",
+    help="Mode for field merging in allOf schemas. "
+    "'constraints': merge only constraints (minItems, maxItems, pattern, etc.) from parent (default). "
+    "'all': merge constraints plus annotations (default, examples) from parent. "
+    "'none': do not merge any fields from parent properties.",
+    choices=[m.value for m in AllOfMergeMode],
+    default=None,
+)
 typing_options.add_argument(
     "--use-type-alias",
     help="Use TypeAlias instead of root models (experimental)",

src/datamodel_code_generator/parser/base.py

@@ -27,6 +27,7 @@
     DEFAULT_SHARED_MODULE_NAME,
     AllExportsCollisionStrategy,
     AllExportsScope,
+    AllOfMergeMode,
     Error,
     ReadOnlyWriteOnlyModelType,
     ReuseScope,
@@ -698,6 +699,7 @@ def __init__(  # noqa: PLR0913, PLR0915
         use_title_as_name: bool = False,
         use_operation_id_as_name: bool = False,
         use_unique_items_as_set: bool = False,
+        allof_merge_mode: AllOfMergeMode = AllOfMergeMode.Constraints,
         http_headers: Sequence[tuple[str, str]] | None = None,
         http_ignore_tls: bool = False,
         use_annotated: bool = False,
@@ -796,6 +798,7 @@ def __init__(  # noqa: PLR0913, PLR0915
         self.use_title_as_name: bool = use_title_as_name
         self.use_operation_id_as_name: bool = use_operation_id_as_name
         self.use_unique_items_as_set: bool = use_unique_items_as_set
+        self.allof_merge_mode: AllOfMergeMode = allof_merge_mode
         self.dataclass_arguments = dataclass_arguments
 
         if base_path:

src/datamodel_code_generator/parser/graphql.py

@@ -16,6 +16,7 @@
 
 from datamodel_code_generator import (
     DEFAULT_SHARED_MODULE_NAME,
+    AllOfMergeMode,
     DataclassArguments,
     DefaultPutDict,
     LiteralType,
@@ -151,6 +152,7 @@ def __init__(  # noqa: PLR0913
         use_title_as_name: bool = False,
         use_operation_id_as_name: bool = False,
         use_unique_items_as_set: bool = False,
+        allof_merge_mode: AllOfMergeMode = AllOfMergeMode.Constraints,
         http_headers: Sequence[tuple[str, str]] | None = None,
         http_ignore_tls: bool = False,
         use_annotated: bool = False,
@@ -245,6 +247,7 @@ def __init__(  # noqa: PLR0913
             use_title_as_name=use_title_as_name,
             use_operation_id_as_name=use_operation_id_as_name,
             use_unique_items_as_set=use_unique_items_as_set,
+            allof_merge_mode=allof_merge_mode,
             http_headers=http_headers,
             http_ignore_tls=http_ignore_tls,
             use_annotated=use_annotated,

src/datamodel_code_generator/parser/jsonschema.py

@@ -23,6 +23,7 @@
 
 from datamodel_code_generator import (
     DEFAULT_SHARED_MODULE_NAME,
+    AllOfMergeMode,
     DataclassArguments,
     InvalidClassNameError,
     ReadOnlyWriteOnlyModelType,
@@ -562,6 +563,7 @@ def __init__(  # noqa: PLR0913
         use_title_as_name: bool = False,
         use_operation_id_as_name: bool = False,
         use_unique_items_as_set: bool = False,
+        allof_merge_mode: AllOfMergeMode = AllOfMergeMode.Constraints,
         http_headers: Sequence[tuple[str, str]] | None = None,
         http_ignore_tls: bool = False,
         use_annotated: bool = False,
@@ -655,6 +657,7 @@ def __init__(  # noqa: PLR0913
             use_title_as_name=use_title_as_name,
             use_operation_id_as_name=use_operation_id_as_name,
             use_unique_items_as_set=use_unique_items_as_set,
+            allof_merge_mode=allof_merge_mode,
             http_headers=http_headers,
             http_ignore_tls=http_ignore_tls,
             use_annotated=use_annotated,
@@ -1365,6 +1368,68 @@ def _is_list_with_any_item_type(self, data_type: DataType | None) -> bool:  # no
                 break
         return item_type.type == ANY
 
+    def _merge_property_schemas(self, parent_dict: dict[str, Any], child_dict: dict[str, Any]) -> dict[str, Any]:
+        """Merge parent and child property schemas for allOf."""
+        if self.allof_merge_mode == AllOfMergeMode.NoMerge:
+            return child_dict.copy()
+
+        non_merged_fields: set[str] = set()
+        if self.allof_merge_mode == AllOfMergeMode.Constraints:
+            non_merged_fields = {"default", "examples", "example"}
+
+        result = {key: value for key, value in parent_dict.items() if key not in non_merged_fields}
+
+        for key, value in child_dict.items():
+            if key in result and isinstance(result[key], dict) and isinstance(value, dict):
+                result[key] = self._merge_property_schemas(result[key], value)
+            else:
+                result[key] = value
+        return result
+
+    def _merge_properties_with_parent_constraints(
+        self, child_obj: JsonSchemaObject, parent_refs: list[str]
+    ) -> JsonSchemaObject:
+        """Merge child properties with parent property constraints for allOf inheritance."""
+        if not child_obj.properties:
+            return child_obj
+
+        parent_properties: dict[str, JsonSchemaObject] = {}
+        for ref in parent_refs:
+            try:
+                parent_schema = self._load_ref_schema_object(ref)
+            except Exception:  # pragma: no cover  # noqa: BLE001, S112
+                continue
+            if parent_schema.properties:
+                for prop_name, prop_schema in parent_schema.properties.items():
+                    if isinstance(prop_schema, JsonSchemaObject) and prop_name not in parent_properties:
+                        parent_properties[prop_name] = prop_schema
+
+        if not parent_properties:
+            return child_obj
+
+        merged_properties: dict[str, JsonSchemaObject | bool] = {}
+        for prop_name, child_prop in child_obj.properties.items():
+            if not isinstance(child_prop, JsonSchemaObject):
+                merged_properties[prop_name] = child_prop
+                continue
+
+            parent_prop = parent_properties.get(prop_name)
+            if parent_prop is None:
+                merged_properties[prop_name] = child_prop
+                continue
+
+            parent_dict = parent_prop.dict(exclude_unset=True, by_alias=True)
+            child_dict = child_prop.dict(exclude_unset=True, by_alias=True)
+            merged_dict = self._merge_property_schemas(parent_dict, child_dict)
+            merged_properties[prop_name] = self.SCHEMA_OBJECT_TYPE.parse_obj(merged_dict)
+
+        merged_obj_dict = child_obj.dict(exclude_unset=True, by_alias=True)
+        merged_obj_dict["properties"] = {
+            k: v.dict(exclude_unset=True, by_alias=True) if isinstance(v, JsonSchemaObject) else v
+            for k, v in merged_properties.items()
+        }
+        return self.SCHEMA_OBJECT_TYPE.parse_obj(merged_obj_dict)
+
     def _get_inherited_field_type(self, prop_name: str, base_classes: list[Reference]) -> DataType | None:
         """Get the data type for an inherited property from parent schemas."""
         for base in base_classes:
@@ -1662,6 +1727,8 @@ def _parse_all_of_item(  # noqa: PLR0912, PLR0913, PLR0917
         required: list[str],
         union_models: list[Reference],
     ) -> None:
+        parent_refs = [item.ref for item in obj.allOf if item.ref]
+
         for all_of_item in obj.allOf:  # noqa: PLR1702
             if all_of_item.ref:  # $ref
                 ref_schema = self._load_ref_schema_object(all_of_item.ref)
@@ -1681,9 +1748,11 @@ def _parse_all_of_item(  # noqa: PLR0912, PLR0913, PLR0917
                     if ref.path not in {b.path for b in base_classes}:
                         base_classes.append(ref)
             else:
+                # Merge child properties with parent constraints before processing
+                merged_item = self._merge_properties_with_parent_constraints(all_of_item, parent_refs)
                 module_name = get_module_name(name, None, treat_dot_as_module=self.treat_dot_as_module)
                 object_fields = self.parse_object_fields(
-                    all_of_item,
+                    merged_item,
                     path,
                     module_name,
                     class_name=name,

src/datamodel_code_generator/parser/openapi.py

@@ -19,6 +19,7 @@
 
 from datamodel_code_generator import (
     DEFAULT_SHARED_MODULE_NAME,
+    AllOfMergeMode,
     DataclassArguments,
     Error,
     LiteralType,
@@ -235,6 +236,7 @@ def __init__(  # noqa: PLR0913
         use_title_as_name: bool = False,
         use_operation_id_as_name: bool = False,
         use_unique_items_as_set: bool = False,
+        allof_merge_mode: AllOfMergeMode = AllOfMergeMode.Constraints,
         http_headers: Sequence[tuple[str, str]] | None = None,
         http_ignore_tls: bool = False,
         use_annotated: bool = False,
@@ -328,6 +330,7 @@ def __init__(  # noqa: PLR0913
             use_title_as_name=use_title_as_name,
             use_operation_id_as_name=use_operation_id_as_name,
             use_unique_items_as_set=use_unique_items_as_set,
+            allof_merge_mode=allof_merge_mode,
             http_headers=http_headers,
             http_ignore_tls=http_ignore_tls,
             use_annotated=use_annotated,

tests/data/expected/main/openapi/allof_materialize_defaults.py

@@ -0,0 +1,19 @@
+# generated by datamodel-codegen:
+#   filename:  allof_materialize_defaults.yaml
+#   timestamp: 2019-07-26T00:00:00+00:00
+
+from __future__ import annotations
+
+from typing import Optional
+
+from pydantic import BaseModel, conint, constr
+
+
+class Parent(BaseModel):
+    name: Optional[constr(min_length=1)] = 'parent_default'
+    count: Optional[conint(ge=0)] = 10
+
+
+class Child(Parent):
+    name: Optional[constr(min_length=1, max_length=100)] = 'parent_default'
+    count: Optional[conint(ge=0, le=1000)] = 10