arpakitlib 1.8.250__tar.gz → 1.8.251__tar.gz

{arpakitlib-1.8.250 → arpakitlib-1.8.251}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: arpakitlib
-Version: 1.8.250
+Version: 1.8.251
 Summary: arpakitlib
 License: Apache-2.0
 Keywords: arpakitlib,arpakit,arpakit-company,arpakitcompany,arpakit_company

arpakitlib-1.8.251/arpakitlib/ar_pydantic_schema_from_sqlalchemy_model.py

@@ -0,0 +1,149 @@
+# arpakit
+import datetime as dt
+from typing import Any, Optional, get_type_hints
+
+from pydantic import BaseModel, Field
+from sqlalchemy import inspect
+from sqlalchemy.orm import ColumnProperty
+from sqlalchemy.sql.sqltypes import (
+    Boolean, Integer, BigInteger, SmallInteger,
+    String, Text, Unicode, UnicodeText,
+    DateTime, Date, Time,
+    Float, Numeric, DECIMAL, LargeBinary, JSON
+)
+
+_ARPAKIT_LIB_MODULE_VERSION = "3.0"
+
+_SQLA_TYPE_MAP = {
+    Boolean: bool,
+    Integer: int,
+    BigInteger: int,
+    SmallInteger: int,
+    Float: float,
+    Numeric: float,
+    DECIMAL: float,
+    String: str,
+    Unicode: str,
+    Text: str,
+    UnicodeText: str,
+    LargeBinary: bytes,
+    JSON: dict,
+    DateTime: dt.datetime,
+    Date: dt.date,
+    Time: dt.time,
+}
+
+
+def _python_type_from_col(col) -> type | str:
+    try:
+        return col.type.python_type
+    except Exception:
+        for sa_t, py_t in _SQLA_TYPE_MAP.items():
+            if isinstance(col.type, sa_t):
+                return py_t
+        return Any
+
+
+def _collect_properties_with_types(model_class: type) -> dict[str, Any]:
+    """
+    Находит все @property в классе и вытаскивает их возвращаемый тип.
+    Если тип не удаётся получить — подставляем Any.
+    """
+    props: dict[str, Any] = {}
+    for name, attr in vars(model_class).items():
+        if isinstance(attr, property):
+            try:
+                hints = get_type_hints(attr.fget) if attr.fget else {}
+                ret_type = hints.get("return", Any)
+            except Exception:
+                ret_type = Any
+            props[name] = ret_type
+    return props
+
+
+def pydantic_schema_from_sqlalchemy_model(
+        sqlalchemy_model: type,
+        *,
+        name: str | None = None,
+        base_model: type[BaseModel] = BaseModel,
+        include_column_defaults: bool = False,
+        exclude_column_names: list[str] | None = None,
+        include_properties: bool = False,
+        include_property_names: list[str] | None = None,
+        exclude_property_names: list[str] | None = None,
+) -> type[BaseModel]:
+    """
+    Генерирует Pydantic-модель из колонок SQLAlchemy-модели и (опционально) из @property.
+
+    - include_column_defaults: добавлять ли default/server_default у колонок.
+    - exclude_column_names: список имён колонок, которые нужно пропустить.
+
+    - include_properties: включать ли свойства (@property). По умолчанию False.
+    - include_property_names: whitelist имён свойств (если задан, берём только их).
+    - exclude_property_names: blacklist имён свойств (исключаются после whitelist'а).
+    """
+    mapper = inspect(sqlalchemy_model).mapper
+    model_name = name or f"{sqlalchemy_model.__name__}Schema"
+
+    annotations: dict[str, Any] = {}
+    attrs: dict[str, Any] = {}
+
+    exclude_column_names = set(exclude_column_names or [])
+    include_property_names = set(include_property_names or [])
+    exclude_property_names = set(exclude_property_names or [])
+
+    # 1) Колонки
+    for prop in mapper.attrs:
+        if not isinstance(prop, ColumnProperty):
+            continue
+        if prop.key in exclude_column_names:
+            continue
+
+        column = prop.columns[0]
+        column_type = _python_type_from_col(column)
+
+        # Аннотация типа
+        if column.nullable:
+            annotations[prop.key] = Optional[column_type]  # type: ignore[name-defined]
+        else:
+            annotations[prop.key] = column_type
+
+        # Дефолты, если нужно
+        if include_column_defaults:
+            default_value = None
+            if column.default is not None and getattr(column.default, "is_scalar", False):
+                default_value = column.default.arg
+            elif column.server_default is not None and getattr(column.server_default.arg, "text", None):
+                default_value = column.server_default.arg.text
+
+            if default_value is not None:
+                attrs[prop.key] = Field(default=default_value)
+
+    # 2) Свойства (@property)
+    if include_properties:
+        property_name_to_type = _collect_properties_with_types(sqlalchemy_model)
+
+        # whitelist (если задан)
+        if include_property_names:
+            property_name_to_type = {
+                k: v
+                for k, v in property_name_to_type.items()
+                if k in include_property_names
+            }
+        else:
+            property_name_to_type = dict(property_name_to_type)
+
+        # blacklist
+        if exclude_property_names:
+            for property_name in list(property_name_to_type.keys()):
+                if property_name in exclude_property_names:
+                    property_name_to_type.pop(name, None)
+
+        # Добавляем аннотации свойств (колонки имеют приоритет)
+        for property_name, property_type in property_name_to_type.items():
+            if property_name in annotations:
+                continue
+            annotations[property_name] = property_type
+
+    attrs["__annotations__"] = annotations
+    return type(model_name, (base_model,), attrs)

{arpakitlib-1.8.250 → arpakitlib-1.8.251}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "arpakitlib"
-version = "1.8.250"
+version = "1.8.251"
 description = "arpakitlib"
 authors = [
     { name = "arpakit_company", email = "support@arpakit.com" }

arpakitlib-1.8.250/arpakitlib/ar_pydantic_schema_from_sqlalchemy_model.py

@@ -1,93 +0,0 @@
-# arpakit
-import datetime as dt
-from typing import Any, Optional
-
-from pydantic import BaseModel, Field
-from sqlalchemy import inspect
-from sqlalchemy.orm import ColumnProperty
-from sqlalchemy.sql.sqltypes import (
-    Boolean, Integer, BigInteger, SmallInteger,
-    String, Text, Unicode, UnicodeText,
-    DateTime, Date, Time,
-    Float, Numeric, DECIMAL, LargeBinary, JSON
-)
-_ARPAKIT_LIB_MODULE_VERSION = "3.0"
-
-_SQLA_TYPE_MAP = {
-    Boolean: bool,
-    Integer: int,
-    BigInteger: int,
-    SmallInteger: int,
-    Float: float,
-    Numeric: float,
-    DECIMAL: float,
-    String: str,
-    Unicode: str,
-    Text: str,
-    UnicodeText: str,
-    LargeBinary: bytes,
-    JSON: dict,
-    DateTime: dt.datetime,
-    Date: dt.date,
-    Time: dt.time,
-}
-
-
-def _python_type_from_col(col) -> type | str:
-    try:
-        return col.type.python_type
-    except Exception:
-        for sa_t, py_t in _SQLA_TYPE_MAP.items():
-            if isinstance(col.type, sa_t):
-                return py_t
-        return Any
-
-
-def pydantic_schema_from_sqlalchemy_model(
-        sqlalchemy_model: type,
-        *,
-        name: str | None = None,
-        base_model: type[BaseModel] = BaseModel,
-        include_defaults: bool = False,
-        exclude_column_names: list[str] | None = None,
-) -> type[BaseModel]:
-    """
-    Генерирует Pydantic-модель только из колонок SQLAlchemy-модели.
-    - include_defaults: добавлять ли default/server_default.
-    - exclude_column_names: список имён колонок, которые нужно пропустить.
-    """
-    mapper = inspect(sqlalchemy_model).mapper
-    model_name = name or f"{sqlalchemy_model.__name__}Schema"
-
-    annotations: dict[str, Any] = {}
-    attrs: dict[str, Any] = {}
-    exclude_column_names = set(exclude_column_names or [])
-
-    for prop in mapper.attrs:
-        if not isinstance(prop, ColumnProperty):
-            continue
-        if prop.key in exclude_column_names:
-            continue
-
-        col = prop.columns[0]
-        py_t = _python_type_from_col(col)
-
-        # Аннотация типа
-        if col.nullable:
-            annotations[prop.key] = Optional[py_t]  # type: ignore[name-defined]
-        else:
-            annotations[prop.key] = py_t
-
-        # Если нужно — добавляем дефолт
-        if include_defaults:
-            default_value = None
-            if col.default is not None and col.default.is_scalar:
-                default_value = col.default.arg
-            elif col.server_default is not None and getattr(col.server_default.arg, "text", None):
-                default_value = col.server_default.arg.text
-
-            if default_value is not None:
-                attrs[prop.key] = Field(default=default_value)
-
-    attrs["__annotations__"] = annotations
-    return type(model_name, (base_model,), attrs)