Optimize code generation data processing (#1020)

* Optimize code generation data processing * Add SQL script generation * Update code gen table scripts * Update types * Fix model jinja and type conversion
2026-03-13 09:31:31 +08:00 · 2026-01-20 18:27:42 +08:00
parent b2785dd46f
commit 1a6aba6105
22 changed files with 467 additions and 298 deletions
--- a/backend/plugin/code_generator/crud/crud_gen.py
+++ b/backend/plugin/code_generator/crud/crud_gen.py
@@ -1,6 +1,6 @@
 from collections.abc import Sequence

-from sqlalchemy import Row, RowMapping, text
+from sqlalchemy import RowMapping, text
 from sqlalchemy.ext.asyncio import AsyncSession

 from backend.common.enums import DataBaseType
@@ -21,57 +21,75 @@ class CRUDGen:
        """
        if DataBaseType.mysql == settings.DATABASE_TYPE:
            sql = """
-            SELECT table_name AS table_name, table_comment AS table_comment
-            FROM information_schema.tables
-            WHERE table_name NOT LIKE 'sys_gen_%'
-            AND table_schema = :table_schema;
+            SELECT TABLE_NAME AS TABLE_NAME,
+                table_comment AS table_comment
+            FROM
+                information_schema.TABLES
+            WHERE
+                TABLE_NAME NOT LIKE'sys_gen_%'
+                AND table_schema = :table_schema;
            """
            stmt = text(sql).bindparams(table_schema=table_schema)
        else:
            sql = """
-            SELECT c.relname AS table_name, obj_description(c.oid) AS table_comment
-            FROM pg_class c
-            LEFT JOIN pg_namespace n ON n.oid = c.relnamespace
-            WHERE c.relkind = 'r'
-            AND n.nspname = 'public' -- schema 通常是 'public'
-            AND c.relname NOT LIKE 'sys_gen_%';
+            SELECT
+                c.relname AS TABLE_NAME,
+                obj_description (c.OID) AS table_comment
+            FROM
+                pg_class c
+                LEFT JOIN pg_namespace n ON n.OID = c.relnamespace
+            WHERE
+                c.relkind = 'r'
+                AND c.relname NOT LIKE'sys_gen_%'
+                AND n.nspname = :table_schema;
            """
-            stmt = text(sql)
+            stmt = text(sql).bindparams(table_schema='public')
        result = await db.execute(stmt)
        return result.mappings().all()

    @staticmethod
-    async def get_table(db: AsyncSession, table_name: str) -> Row[tuple]:
+    async def get_table(db: AsyncSession, table_schema: str, table_name: str) -> RowMapping | None:
        """
        获取表信息

        :param db: 数据库会话
+        :param table_schema: 数据库 schema 名称
        :param table_name: 表名
        :return:
        """
        if DataBaseType.mysql == settings.DATABASE_TYPE:
            sql = """
-            SELECT table_name AS table_name, table_comment AS table_comment
-            FROM information_schema.tables
-            WHERE table_name NOT LIKE 'sys_gen_%'
-            AND table_name = :table_name;
+            SELECT TABLE_NAME AS TABLE_NAME,
+                table_comment AS table_comment
+            FROM
+                information_schema.TABLES
+            WHERE
+                TABLE_NAME NOT LIKE'sys_gen_%'
+                AND TABLE_NAME = :table_name
+                AND table_schema = :table_schema;
            """
+            stmt = text(sql).bindparams(table_schema=table_schema, table_name=table_name)
        else:
            sql = """
-            SELECT c.relname AS table_name, obj_description(c.oid) AS table_comment
-            FROM pg_class c
-            LEFT JOIN pg_namespace n ON n.oid = c.relnamespace
-            WHERE c.relkind = 'r'
-            AND n.nspname = 'public' -- schema 通常是 'public'
-            AND c.relname = :table_name
-            AND c.relname NOT LIKE 'sys_gen_%';
+            SELECT
+                c.relname AS table_name,
+                obj_description (c.OID) AS table_comment
+            FROM
+                pg_class c
+                LEFT JOIN pg_namespace n ON n.OID = c.relnamespace
+            WHERE
+                c.relkind = 'r'
+                AND c.relname NOT LIKE'sys_gen_%'
+                AND c.relname = :table_name
+                AND n.nspname = :table_schema;
            """
-        stmt = text(sql).bindparams(table_name=table_name)
+            stmt = text(sql).bindparams(table_schema='public', table_name=table_name)
        result = await db.execute(stmt)
-        return result.fetchone()
+        row = result.fetchone()
+        return row._mapping if row else None

    @staticmethod
-    async def get_all_columns(db: AsyncSession, table_schema: str, table_name: str) -> Sequence[Row[tuple]]:
+    async def get_all_columns(db: AsyncSession, table_schema: str, table_name: str) -> Sequence[RowMapping]:
        """
        获取所有列信息

@@ -82,54 +100,73 @@ class CRUDGen:
        """
        if DataBaseType.mysql == settings.DATABASE_TYPE:
            sql = """
-            SELECT column_name AS column_name,
-            CASE WHEN column_key = 'PRI' THEN 1 ELSE 0 END AS is_pk,
-            CASE WHEN is_nullable = 'NO' OR column_key = 'PRI' THEN 0 ELSE 1 END AS is_nullable,
-            ordinal_position AS sort, column_comment AS column_comment,
-            column_type AS column_type FROM information_schema.columns
-            WHERE table_schema = :table_schema
-            AND table_name = :table_name
-            AND column_name <> 'id'
-            AND column_name <> 'created_time'
-            AND column_name <> 'updated_time'
-            ORDER BY sort;
+            SELECT COLUMN_NAME AS COLUMN_NAME,
+                CASE
+                    WHEN column_key = 'PRI' THEN
+                        1
+                    ELSE
+                        0
+                END AS is_pk,
+                CASE
+                    WHEN is_nullable = 'NO'
+                        OR column_key = 'PRI' THEN
+                        0
+                    ELSE
+                        1
+                END AS is_nullable,
+                ordinal_position AS sort,
+                column_comment AS column_comment,
+                column_type AS column_type
+            FROM
+                information_schema.COLUMNS
+            WHERE
+                COLUMN_NAME <> 'id'
+                AND COLUMN_NAME <> 'created_time'
+                AND COLUMN_NAME <> 'updated_time'
+                AND TABLE_NAME = :table_name
+                AND table_schema = :table_schema
+            ORDER BY
+                sort;
            """
            stmt = text(sql).bindparams(table_schema=table_schema, table_name=table_name)
        else:
            sql = """
-            SELECT a.attname AS column_name,
-            CASE WHEN EXISTS (
-            SELECT 1
-            FROM pg_constraint c
-            WHERE c.conrelid = t.oid
-            AND c.contype = 'p'
-            AND a.attnum = ANY(c.conkey)
-            ) THEN 1 ELSE 0 END AS is_pk,
-            CASE WHEN a.attnotnull OR EXISTS (
-            SELECT 1
-            FROM pg_constraint c
-            WHERE c.conrelid = t.oid
-            AND c.contype = 'p'
-            AND a.attnum = ANY(c.conkey)
-            ) THEN 0 ELSE 1 END AS is_nullable,
-            a.attnum AS sort,
-            col_description(t.oid, a.attnum) AS column_comment,
-            pg_catalog.format_type(a.atttypid, a.atttypmod) AS column_type
-            FROM pg_attribute a
-            JOIN pg_class t ON a.attrelid = t.oid
-            JOIN pg_namespace n ON n.oid = t.relnamespace
-            WHERE n.nspname = 'public'  -- 根据你的实际情况修改 schema 名称，通常是 'public'
-            AND t.relname = :table_name
-            AND a.attnum > 0
-            AND NOT a.attisdropped
-            AND a.attname <> 'id'
-            AND a.attname <> 'created_time'
-            AND a.attname <> 'updated_time'
-            ORDER BY sort;
-            """
-            stmt = text(sql).bindparams(table_name=table_name)
+            SELECT
+                a.attname AS COLUMN_NAME,
+                CASE
+                    WHEN EXISTS (SELECT 1 FROM pg_constraint c WHERE c.conrelid = t.OID AND c.contype = 'p' AND a.attnum = ANY (c.conkey)) THEN
+                        1
+                    ELSE
+                        0
+                END AS is_pk,
+                CASE
+                    WHEN a.attnotnull
+                        OR EXISTS (SELECT 1 FROM pg_constraint c WHERE c.conrelid = t.OID AND c.contype = 'p' AND a.attnum = ANY (c.conkey)) THEN
+                        0
+                    ELSE
+                        1
+                END AS is_nullable,
+                a.attnum AS sort,
+                col_description (t.OID, a.attnum) AS column_comment,
+                pg_catalog.format_type (a.atttypid, a.atttypmod) AS column_type
+            FROM
+                pg_attribute a
+                JOIN pg_class t ON a.attrelid = t.
+                OID JOIN pg_namespace n ON n.OID = t.relnamespace
+            WHERE
+                a.attnum > 0
+                AND NOT a.attisdropped
+                AND a.attname <> 'id'
+                AND a.attname <> 'created_time'
+                AND a.attname <> 'updated_time'
+                AND t.relname = :table_name
+                AND n.nspname = :table_schema
+            ORDER BY
+                sort;
+            """  # noqa: E501
+            stmt = text(sql).bindparams(table_schema='public', table_name=table_name)
        result = await db.execute(stmt)
-        return result.fetchall()
+        return result.mappings().all()


 gen_dao: CRUDGen = CRUDGen()
--- a/backend/plugin/code_generator/enums.py
+++ b/backend/plugin/code_generator/enums.py
@@ -19,34 +19,34 @@ class GenMySQLColumnType(StrEnum):
    DateTime = 'datetime'  # DATETIME
    DECIMAL = 'Decimal'
    DOUBLE = 'float'
-    Double = 'float'  # DOUBLE
    DOUBLE_PRECISION = 'float'
+    Double = 'float'  # DOUBLE
    Enum = 'Enum'  # Enum()
    FLOAT = 'float'
    Float = 'float'  # FLOAT
    INT = 'int'  # INTEGER
    INTEGER = 'int'
    Integer = 'int'  # INTEGER
-    Interval = 'timedelta'  # DATETIME
+    Interval = 'timedelta'  # INTERVAL
    JSON = 'dict'
    LargeBinary = 'bytes'  # BLOB
    NCHAR = 'str'
    NUMERIC = 'Decimal'
-    Numeric = 'Decimal'  # NUMERIC
    NVARCHAR = 'str'  # String
+    Numeric = 'Decimal'  # NUMERIC
    PickleType = 'bytes'  # BLOB
    REAL = 'float'
    SMALLINT = 'int'
    SmallInteger = 'int'  # SMALLINT
    String = 'str'  # String
    TEXT = 'str'
-    Text = 'str'  # TEXT
    TIME = 'time'
-    Time = 'time'  # TIME
    TIMESTAMP = 'datetime'
+    Text = 'str'  # TEXT
+    Time = 'time'  # TIME
+    UUID = 'str | UUID'
    Unicode = 'str'  # String
    UnicodeText = 'str'  # TEXT
-    UUID = 'str | UUID'
    Uuid = 'str'  # CHAR(32)
    VARBINARY = 'bytes'
    VARCHAR = 'str'  # String
@@ -77,6 +77,8 @@ class GenPostgreSQLColumnType(StrEnum):
    BOOLEAN = 'bool'
    Boolean = 'bool'  # BOOLEAN
    CHAR = 'str'
+    CHARACTER = 'str'  # CHAR
+    CHARACTER_VARYING = 'str'  # CHARACTER VARYING
    CLOB = 'str'
    DATE = 'date'
    Date = 'date'  # DATE
@@ -84,8 +86,8 @@ class GenPostgreSQLColumnType(StrEnum):
    DateTime = 'datetime'  # TIMESTAMP WITHOUT TIME ZONE
    DECIMAL = 'Decimal'
    DOUBLE = 'float'
-    Double = 'float'  # DOUBLE PRECISION
    DOUBLE_PRECISION = 'float'  # DOUBLE PRECISION
+    Double = 'float'  # DOUBLE PRECISION
    Enum = 'Enum'  # Enum(name='enum')
    FLOAT = 'float'
    Float = 'float'  # FLOAT
@@ -97,21 +99,25 @@ class GenPostgreSQLColumnType(StrEnum):
    LargeBinary = 'bytes'  # BYTEA
    NCHAR = 'str'
    NUMERIC = 'Decimal'
-    Numeric = 'Decimal'  # NUMERIC
    NVARCHAR = 'str'  # String
+    Numeric = 'Decimal'  # NUMERIC
    PickleType = 'bytes'  # BYTEA
    REAL = 'float'
    SMALLINT = 'int'
    SmallInteger = 'int'  # SMALLINT
    String = 'str'  # String
    TEXT = 'str'
-    Text = 'str'  # TEXT
    TIME = 'time'  # TIME WITHOUT TIME ZONE
-    Time = 'time'  # TIME WITHOUT TIME ZONE
+    TIME_WITHOUT_TIME_ZONE = 'time'  # TIME WITHOUT TIME ZONE
+    TIME_WITH_TIME_ZONE = 'time'  # TIME WITH TIME ZONE
    TIMESTAMP = 'datetime'  # TIMESTAMP WITHOUT TIME ZONE
+    TIMESTAMP_WITHOUT_TIME_ZONE = 'datetime'  # TIMESTAMP WITHOUT TIME ZONE
+    TIMESTAMP_WITH_TIME_ZONE = 'datetime'  # TIMESTAMP WITH TIME ZONE
+    Text = 'str'  # TEXT
+    Time = 'time'  # TIME WITHOUT TIME ZONE
+    UUID = 'str | UUID'
    Unicode = 'str'  # String
    UnicodeText = 'str'  # TEXT
-    UUID = 'str | UUID'
    Uuid = 'str'
    VARBINARY = 'bytes'
    VARCHAR = 'str'  # String
--- a/backend/plugin/code_generator/model/business.py
+++ b/backend/plugin/code_generator/model/business.py
@@ -11,19 +11,14 @@ class GenBusiness(Base):
    __tablename__ = 'gen_business'

    id: Mapped[id_key] = mapped_column(init=False)
-    app_name: Mapped[str] = mapped_column(sa.String(64), comment='应用名称（英文）')
-    table_name: Mapped[str] = mapped_column(sa.String(256), unique=True, comment='表名称（英文）')
-    doc_comment: Mapped[str] = mapped_column(sa.String(256), comment='文档注释（用于函数/参数文档）')
+    app_name: Mapped[str] = mapped_column(sa.String(64), comment='应用名称')
+    table_name: Mapped[str] = mapped_column(sa.String(256), unique=True, comment='表名称')
+    doc_comment: Mapped[str] = mapped_column(sa.String(256), comment='文档注释')
    table_comment: Mapped[str | None] = mapped_column(sa.String(256), default=None, comment='表描述')
-    # relate_model_fk: Mapped[int | None] = mapped_column(default=None, comment='关联表外键')
-    class_name: Mapped[str | None] = mapped_column(sa.String(64), default=None, comment='基础类名（默认为英文表名称）')
-    schema_name: Mapped[str | None] = mapped_column(
-        sa.String(64), default=None, comment='Schema 名称 (默认为英文表名称)'
-    )
-    filename: Mapped[str | None] = mapped_column(sa.String(64), default=None, comment='基础文件名（默认为英文表名称）')
-    default_datetime_column: Mapped[bool] = mapped_column(default=True, comment='是否存在默认时间列')
-    api_version: Mapped[str] = mapped_column(sa.String(32), default='v1', comment='代码生成 api 版本，默认为 v1')
-    gen_path: Mapped[str | None] = mapped_column(
-        sa.String(256), default=None, comment='代码生成路径（默认为 app 根路径）'
-    )
+    class_name: Mapped[str | None] = mapped_column(sa.String(64), default=None, comment='基础类名')
+    schema_name: Mapped[str | None] = mapped_column(sa.String(64), default=None, comment='Schema 名称')
+    filename: Mapped[str | None] = mapped_column(sa.String(64), default=None, comment='基础文件名')
+    datetime_mixin: Mapped[bool] = mapped_column(default=True, comment='是否包含时间 Mixin 列')
+    api_version: Mapped[str] = mapped_column(sa.String(32), default='v1', comment='API 版本')
+    gen_path: Mapped[str | None] = mapped_column(sa.String(256), default=None, comment='生成路径')
    remark: Mapped[str | None] = mapped_column(UniversalText, default=None, comment='备注')
--- a/backend/plugin/code_generator/plugin.toml
+++ b/backend/plugin/code_generator/plugin.toml
@@ -1,6 +1,6 @@
 [plugin]
 summary = '代码生成'
-version = '0.0.7'
+version = '0.1.0'
 description = '生成通用业务代码'
 author = 'wu-clan'

--- a/backend/plugin/code_generator/schema/business.py
+++ b/backend/plugin/code_generator/schema/business.py
@@ -1,8 +1,10 @@
 from datetime import datetime

-from pydantic import ConfigDict, Field
+from pydantic import ConfigDict, Field, field_validator

+from backend.common.exception import errors
 from backend.common.schema import SchemaBase
+from backend.utils.pattern_validate import is_english_identifier


 class GenBusinessSchemaBase(SchemaBase):
@@ -15,11 +17,19 @@ class GenBusinessSchemaBase(SchemaBase):
    class_name: str | None = Field(None, description='用于 python 代码基础类名')
    schema_name: str | None = Field(None, description='用于 python Schema 代码基础类名')
    filename: str | None = Field(None, description='用于 python 代码基础文件名')
-    default_datetime_column: bool = Field(True, description='是否存在默认时间列')
-    api_version: str = Field('v1', description='代码生成 api 版本')
-    gen_path: str | None = Field(None, description='代码生成路径')
+    datetime_mixin: bool = Field(True, description='是否包含时间 Mixin 列')
+    api_version: str = Field('v1', description='API 版本')
+    gen_path: str | None = Field(None, description='生成路径（默认在 backend/app 目录下）')
    remark: str | None = Field(None, description='备注')

+    @field_validator('app_name', 'table_name')
+    @classmethod
+    def validate_english_only(cls, v: str) -> str:
+        """验证英文字段"""
+        if not is_english_identifier(v):
+            raise errors.RequestError(msg='必须以英文字母开头且只能包含英文字母和下划线')
+        return v
+

 class CreateGenBusinessParam(GenBusinessSchemaBase):
    """创建代码生成业务参数"""
--- a/backend/plugin/code_generator/schema/column.py
+++ b/backend/plugin/code_generator/schema/column.py
@@ -19,8 +19,8 @@ class GenColumnSchemaBase(SchemaBase):

    @field_validator('type')
    @classmethod
-    def type_update(cls, v: str) -> str:
-        """更新列类型"""
+    def normalize_type(cls, v: str) -> str:
+        """规范化类型"""
        return sql_type_to_sqlalchemy(v)


--- a/backend/plugin/code_generator/service/gen_service.py
+++ b/backend/plugin/code_generator/service/gen_service.py
@@ -21,6 +21,7 @@ from backend.plugin.code_generator.schema.business import CreateGenBusinessParam
 from backend.plugin.code_generator.schema.column import CreateGenColumnParam
 from backend.plugin.code_generator.schema.gen import ImportParam
 from backend.plugin.code_generator.service.column_service import gen_column_service
+from backend.plugin.code_generator.utils.format_code import format_python_code
 from backend.plugin.code_generator.utils.gen_template import gen_template
 from backend.plugin.code_generator.utils.type_conversion import sql_type_to_pydantic

@@ -50,7 +51,7 @@ class GenService:
        :return:
        """

-        table_info = await gen_dao.get_table(db, obj.table_name)
+        table_info = await gen_dao.get_table(db, obj.table_schema, obj.table_name)
        if not table_info:
            raise errors.NotFoundError(msg='数据库表不存在')

@@ -58,13 +59,13 @@ class GenService:
        if business_info:
            raise errors.ConflictError(msg='已存在相同数据库表业务')

-        table_name = table_info[0]
+        table_name = table_info['table_name']
        new_business = GenBusiness(
            **CreateGenBusinessParam(
                app_name=obj.app,
                table_name=table_name,
-                doc_comment=table_info[1] or table_name.split('_')[-1],
-                table_comment=table_info[1],
+                doc_comment=table_info['table_comment'] or table_name.split('_')[-1],
+                table_comment=table_info['table_comment'],
                class_name=to_pascal(table_name),
                schema_name=to_pascal(table_name),
                filename=table_name,
@@ -75,25 +76,27 @@ class GenService:

        column_info = await gen_dao.get_all_columns(db, obj.table_schema, table_name)
        for column in column_info:
-            column_type = column[-1].split('(')[0].upper()
+            column_type = column['column_type'].split('(')[0].upper()
            pd_type = sql_type_to_pydantic(column_type)
            await gen_column_dao.create(
                db,
                CreateGenColumnParam(
-                    name=column[0],
-                    comment=column[-2],
+                    name=column['column_name'],
+                    comment=column['column_comment'],
                    type=column_type,
-                    sort=column[-3],
-                    length=column[-1].split('(')[1][:-1] if pd_type == 'str' and '(' in column[-1] else 0,
-                    is_pk=column[1],
-                    is_nullable=column[2],
+                    sort=column['sort'],
+                    length=column['column_type'].split('(')[1][:-1]
+                    if pd_type == 'str' and '(' in column['column_type']
+                    else 0,
+                    is_pk=column['is_pk'],
+                    is_nullable=column['is_nullable'],
                    gen_business_id=new_business.id,
                ),
                pd_type=pd_type,
            )

    @staticmethod
-    async def render_tpl_code(*, db: AsyncSession, business: GenBusiness) -> dict[str, str]:
+    async def _render_tpl_code(*, db: AsyncSession, business: GenBusiness) -> dict[str, str]:
        """
        渲染模板代码

@@ -106,10 +109,16 @@ class GenService:
            raise errors.NotFoundError(msg='代码生成模型表为空')

        gen_vars = gen_template.get_vars(business, gen_models)
-        return {
-            tpl_path: await gen_template.get_template(tpl_path).render_async(**gen_vars)
-            for tpl_path in gen_template.get_template_files()
-        }
+        template_mapping = gen_template.get_template_path_mapping(business)
+
+        rendered_codes = {}
+        for template_path, output_path in template_mapping.items():
+            code = await gen_template.get_template(template_path).render_async(**gen_vars)
+            if output_path.endswith('.py'):
+                code = await format_python_code(code)
+            rendered_codes[output_path] = code
+
+        return rendered_codes

    async def preview(self, *, db: AsyncSession, pk: int) -> dict[str, bytes]:
        """
@@ -119,33 +128,20 @@ class GenService:
        :param pk: 业务 ID
        :return:
        """
-
        business = await gen_business_dao.get(db, pk)
        if not business:
            raise errors.NotFoundError(msg='业务不存在')

-        tpl_code_map = await self.render_tpl_code(db=db, business=business)
-
        codes = {}
-        for tpl_path, code in tpl_code_map.items():
-            if tpl_path.startswith('python'):
-                rootpath = f'fastapi_best_architecture/backend/app/{business.app_name}'
-                template_name = tpl_path.split('/')[-1]
-                filepath = None
-                match template_name:
-                    case 'api.jinja':
-                        filepath = f'{rootpath}/api/{business.api_version}/{business.filename}.py'
-                    case 'crud.jinja':
-                        filepath = f'{rootpath}/crud/crud_{business.filename}.py'
-                    case 'model.jinja':
-                        filepath = f'{rootpath}/model/{business.filename}.py'
-                    case 'schema.jinja':
-                        filepath = f'{rootpath}/schema/{business.filename}.py'
-                    case 'service.jinja':
-                        filepath = f'{rootpath}/service/{business.filename}_service.py'
+        backend_path = 'fastapi_best_architecture/backend/app/'

-                if filepath:
-                    codes[filepath] = code.encode('utf-8')
+        init_files = gen_template.get_init_files(business)
+        for filepath, content in init_files.items():
+            codes[f'{backend_path}{filepath}'] = content.encode('utf-8')
+
+        rendered_codes = await self._render_tpl_code(db=db, business=business)
+        for filepath, code in rendered_codes.items():
+            codes[f'{backend_path}{filepath}'] = code.encode('utf-8')

        return codes

@@ -158,15 +154,20 @@ class GenService:
        :param pk: 业务 ID
        :return:
        """
-
        business = await gen_business_dao.get(db, pk)
        if not business:
            raise errors.NotFoundError(msg='业务不存在')

-        gen_path = business.gen_path or '.../backend/app/'
-        target_files = gen_template.get_code_gen_paths(business)
+        gen_path = business.gen_path or '<project_root>/backend/app'
+        paths = []

-        return [os.path.join(gen_path, *target_file.split('/')) for target_file in target_files]
+        init_files = gen_template.get_init_files(business)
+        paths.extend(os.path.join(gen_path, *filepath.split('/')) for filepath in init_files.keys())
+
+        template_mapping = gen_template.get_template_path_mapping(business)
+        paths.extend(os.path.join(gen_path, *filepath.split('/')) for filepath in template_mapping.values())
+
+        return paths

    async def generate(self, *, db: AsyncSession, pk: int) -> str:
        """
@@ -176,51 +177,26 @@ class GenService:
        :param pk: 业务 ID
        :return:
        """
-
        business = await gen_business_dao.get(db, pk)
        if not business:
            raise errors.NotFoundError(msg='业务不存在')

-        tpl_code_map = await self.render_tpl_code(db=db, business=business)
-        gen_path = business.gen_path or BASE_PATH / 'app'
+        gen_path = business.gen_path or str(BASE_PATH / 'app')

-        for tpl_path, code in tpl_code_map.items():
-            code_filepath = os.path.join(
-                gen_path,
-                *gen_template.get_code_gen_path(tpl_path, business).split('/'),
-            )
+        init_files = gen_template.get_init_files(business)
+        for init_filepath, init_content in init_files.items():
+            full_path = os.path.join(gen_path, *init_filepath.split('/'))
+            init_folder = anyio.Path(full_path).parent
+            await init_folder.mkdir(parents=True, exist_ok=True)
+            async with await open_file(full_path, 'w', encoding='utf-8') as f:
+                await f.write(init_content)

-            # 写入 init 文件
-            code_folder = anyio.Path(code_filepath).parent
+        rendered_codes = await self._render_tpl_code(db=db, business=business)
+        for code_filepath, code in rendered_codes.items():
+            full_path = os.path.join(gen_path, *code_filepath.split('/'))
+            code_folder = anyio.Path(full_path).parent
            await code_folder.mkdir(parents=True, exist_ok=True)
-
-            init_filepath = code_folder.joinpath('__init__.py')
-            if not await init_filepath.exists():
-                async with await open_file(init_filepath, 'w', encoding='utf-8') as f:
-                    await f.write(gen_template.init_content)
-
-            # api __init__.py
-            if 'api' in code_filepath:
-                api_init_filepath = code_folder.parent.joinpath('__init__.py')
-                async with await open_file(api_init_filepath, 'w', encoding='utf-8') as f:
-                    await f.write(gen_template.init_content)
-
-            # app __init__.py
-            if 'service' in code_filepath:
-                app_init_filepath = code_folder.parent.joinpath('__init__.py')
-                async with await open_file(app_init_filepath, 'w', encoding='utf-8') as f:
-                    await f.write(gen_template.init_content)
-
-            # model init 文件补充
-            if code_folder.name == 'model':
-                async with await open_file(init_filepath, 'a', encoding='utf-8') as f:
-                    await f.write(
-                        f'from backend.app.{business.app_name}.model.{business.table_name} '
-                        f'import {to_pascal(business.table_name)}\n',
-                    )
-
-            # 写入代码文件
-            async with await open_file(code_filepath, 'w', encoding='utf-8') as f:
+            async with await open_file(full_path, 'w', encoding='utf-8') as f:
                await f.write(code)

        return gen_path
@@ -233,41 +209,18 @@ class GenService:
        :param pk: 业务 ID
        :return:
        """
-
        business = await gen_business_dao.get(db, pk)
        if not business:
            raise errors.NotFoundError(msg='业务不存在')

        bio = io.BytesIO()
        with zipfile.ZipFile(bio, 'w') as zf:
-            tpl_code_map = await self.render_tpl_code(db=db, business=business)
-            for tpl_path, code in tpl_code_map.items():
-                code_filepath = gen_template.get_code_gen_path(tpl_path, business)
+            init_files = gen_template.get_init_files(business)
+            for init_filepath, init_content in init_files.items():
+                zf.writestr(init_filepath, init_content)

-                # 写入 init 文件
-                code_dir = os.path.dirname(code_filepath)
-                init_filepath = os.path.join(code_dir, '__init__.py')
-                if 'model' not in code_filepath.split('/'):
-                    zf.writestr(init_filepath, gen_template.init_content)
-                else:
-                    zf.writestr(
-                        init_filepath,
-                        f'{gen_template.init_content}'
-                        f'from backend.app.{business.app_name}.model.{business.table_name} '
-                        f'import {to_pascal(business.table_name)}\n',
-                    )
-
-                # api __init__.py
-                if 'api' in code_dir:
-                    api_init_filepath = os.path.join(os.path.dirname(code_dir), '__init__.py')
-                    zf.writestr(api_init_filepath, gen_template.init_content)
-
-                # app __init__.py
-                if 'service' in code_dir:
-                    app_init_filepath = os.path.join(os.path.dirname(code_dir), '__init__.py')
-                    zf.writestr(app_init_filepath, gen_template.init_content)
-
-                # 写入代码文件
+            rendered_codes = await self._render_tpl_code(db=db, business=business)
+            for code_filepath, code in rendered_codes.items():
                zf.writestr(code_filepath, code)

        bio.seek(0)
--- a/backend/plugin/code_generator/sql/mysql/init.sql
+++ b/backend/plugin/code_generator/sql/mysql/init.sql
@@ -1,4 +1,4 @@
-insert into gen_business (id, app_name, table_name, doc_comment, table_comment, class_name, schema_name, filename, default_datetime_column, api_version, gen_path, remark, created_time, updated_time)
+insert into gen_business (id, app_name, table_name, doc_comment, table_comment, class_name, schema_name, filename, datetime_mixin, api_version, gen_path, remark, created_time, updated_time)
 values (1, 'test', 'sys_opera_log', '操作日志表', '操作日志表', 'SysOperaLog', 'SysOperaLog', 'sys_opera_log', true, 'v1', null, null, '2025-12-15 15:30:33', null);

 insert into gen_column (id, name, comment, type, pd_type, `default`, sort, `length`, is_pk, is_nullable, gen_business_id)
--- a/backend/plugin/code_generator/sql/mysql/init_snowflake.sql
+++ b/backend/plugin/code_generator/sql/mysql/init_snowflake.sql
@@ -1,4 +1,4 @@
-insert into gen_business (id, app_name, table_name, doc_comment, table_comment, class_name, schema_name, filename, default_datetime_column, api_version, gen_path, remark, created_time, updated_time)
+insert into gen_business (id, app_name, table_name, doc_comment, table_comment, class_name, schema_name, filename, datetime_mixin, api_version, gen_path, remark, created_time, updated_time)
 values (2112248797819043840, 'test', 'sys_opera_log', '操作日志表', '操作日志表', 'SysOperaLog', 'SysOperaLog', 'sys_opera_log', true, 'v1', null, null, '2025-12-15 15:30:33', null);

 insert into gen_column (id, name, comment, type, pd_type, `default`, sort, `length`, is_pk, is_nullable, gen_business_id)
--- a/backend/plugin/code_generator/sql/postgresql/init.sql
+++ b/backend/plugin/code_generator/sql/postgresql/init.sql
@@ -1,4 +1,4 @@
-insert into gen_business (id, app_name, table_name, doc_comment, table_comment, class_name, schema_name, filename, default_datetime_column, api_version, gen_path, remark, created_time, updated_time)
+insert into gen_business (id, app_name, table_name, doc_comment, table_comment, class_name, schema_name, filename, datetime_mixin, api_version, gen_path, remark, created_time, updated_time)
 values (1, 'test', 'sys_opera_log', '操作日志表', '操作日志表', 'SysOperaLog', 'SysOperaLog', 'sys_opera_log', true, 'v1', null, null, '2025-12-15 15:30:33', null);

 insert into gen_column (id, name, comment, type, pd_type, "default", sort, "length", is_pk, is_nullable, gen_business_id)
--- a/backend/plugin/code_generator/sql/postgresql/init_snowflake.sql
+++ b/backend/plugin/code_generator/sql/postgresql/init_snowflake.sql
@@ -1,4 +1,4 @@
-insert into gen_business (id, app_name, table_name, doc_comment, table_comment, class_name, schema_name, filename, default_datetime_column, api_version, gen_path, remark, created_time, updated_time)
+insert into gen_business (id, app_name, table_name, doc_comment, table_comment, class_name, schema_name, filename, datetime_mixin, api_version, gen_path, remark, created_time, updated_time)
 values (2112248797819043840, 'test', 'sys_opera_log', '操作日志表', '操作日志表', 'SysOperaLog', 'SysOperaLog', 'sys_opera_log', true, 'v1', null, null, '2025-12-15 15:30:33', null);

 insert into gen_column (id, name, comment, type, pd_type, "default", sort, "length", is_pk, is_nullable, gen_business_id)
--- a/backend/plugin/code_generator/templates/python/crud.jinja
+++ b/backend/plugin/code_generator/templates/python/crud.jinja
@@ -8,7 +8,7 @@ from backend.app.{{ app_name }}.model import {{ class_name }}
 from backend.app.{{ app_name }}.schema.{{ table_name }} import Create{{ schema_name }}Param, Update{{ schema_name }}Param


-class CRUD{{ class_name }}(CRUDPlus[{{ schema_name }}]):
+class CRUD{{ class_name }}(CRUDPlus[{{ class_name }}]):
    async def get(self, db: AsyncSession, pk: int) -> {{ class_name }} | None:
        """
        获取{{ doc_comment }}
--- a/backend/plugin/code_generator/templates/python/model.jinja
+++ b/backend/plugin/code_generator/templates/python/model.jinja
@@ -5,29 +5,48 @@
 'HSTORE', 'INET', 'INT4MULTIRANGE', 'INT4RANGE', 'INT8MULTIRANGE', 'INT8RANGE', 'INTERVAL', 'JSONB', 'JSONPATH',
 'MACADDR', 'MACADDR8', 'MONEY', 'NUMMULTIRANGE', 'NUMRANGE', 'OID', 'REGCLASS', 'REGCONFIG', 'TSMULTIRANGE', 'TSQUERY',
 'TSRANGE', 'TSTZMULTIRANGE', 'TSTZRANGE', 'TSVECTOR'] %}
-{% if default_datetime_column %}
-from datetime import datetime
+{% set FACTORY_TYPES = ['dict', 'list', 'list[str]', 'list[int]'] %}
+{% set pd_types = models|map(attribute='pd_type')|list %}
+{% set NEED_MYSQL_DIALECT = database_type == 'mysql' and model_types|select('in', MYSQL_TYPES)|list|length > 0 %}
+{% set NEED_PGSQL_DIALECT = database_type == 'postgresql' and model_types|select('in', POSTGRESQL_TYPES)|list|length > 0 %}
+{% set NEED_DATE = 'date' in pd_types %}
+{% set NEED_DATETIME = datetime_mixin or 'datetime' in pd_types %}
+{% set NEED_UNIVERSAL_TEXT = 'TEXT' in model_types or 'Text' in model_types or 'LONGTEXT' in model_types %}
+{% set NEED_TIMEZONE = 'TIMESTAMP' in model_types or 'DateTime' in model_types or 'TIMESTAMP WITHOUT TIME ZONE' in model_types or 'TIMESTAMP WITH TIME ZONE' in model_types %}
+{% if NEED_DATETIME or NEED_DATE %}
+from datetime import {% if NEED_DATETIME %}datetime{% endif %}{% if NEED_DATE %}{% if NEED_DATETIME %}, {% endif %}date{% endif %}
+
 {% endif %}
 {% if model_types|select('in', DECIMAL_TYPES)|first %}
 from decimal import Decimal
+
 {% endif %}
 {% if 'Uuid' in model_types or 'UUID' in model_types %}
 from uuid import UUID
-{% endif %}

+{% endif %}
 import sqlalchemy as sa

-{% if database_type == 'mysql' -%}
+{% if NEED_MYSQL_DIALECT %}
 from sqlalchemy.dialects import mysql
-{% else -%}
+{% endif %}
+{% if NEED_PGSQL_DIALECT %}
 from sqlalchemy.dialects import postgresql
-{% endif -%}
+{% endif %}
 from sqlalchemy.orm import Mapped, mapped_column

-from backend.common.model import {% if default_datetime_column %}Base{% else %}DataClassBase{% endif %}, id_key
+from backend.common.model import {% if datetime_mixin %}Base{% else %}DataClassBase{% endif %}, id_key
+{%- if NEED_UNIVERSAL_TEXT or NEED_TIMEZONE %}, {% endif %}
+{%- if NEED_UNIVERSAL_TEXT %}UniversalText{% endif %}
+{%- if NEED_UNIVERSAL_TEXT and NEED_TIMEZONE %}, {% endif %}
+{%- if NEED_TIMEZONE %}TimeZone{% endif %}
+
+{% if NEED_TIMEZONE %}
+from backend.utils.timezone import timezone
+{% endif %}


-class {{ class_name }}({% if default_datetime_column %}Base{% else %}DataClassBase{% endif %}):
+class {{ class_name }}({% if datetime_mixin %}Base{% else %}DataClassBase{% endif %}):
    """{{ table_comment }}"""

    __tablename__ = '{{ table_name }}'
@@ -38,41 +57,36 @@ class {{ class_name }}({% if default_datetime_column %}Base{% else %}DataClassBa
    {%- if model.is_nullable %} Mapped[{{ model.pd_type }} | None]
    {%- else %} Mapped[{{ model.pd_type }}]
    {%- endif %} = mapped_column(
-    {%- if model.type in ['NVARCHAR', 'String', 'Unicode', 'VARCHAR'] -%}
+    {%- if model.type in ['TEXT', 'Text', 'LONGTEXT'] -%}
+        UniversalText
+    {%- elif model.type in ['TIMESTAMP', 'DateTime', 'TIMESTAMP WITHOUT TIME ZONE', 'TIMESTAMP WITH TIME ZONE'] -%}
+        TimeZone
+    {%- elif model.type in ['NVARCHAR', 'String', 'Unicode', 'VARCHAR', 'CHARACTER VARYING'] -%}
        sa.String({{ model.length }})
    {%- elif database_type == 'mysql' and model.type in MYSQL_TYPES -%}
-        mysql.{{ model.type }}()
+        mysql.{{ model.type|sqlalchemy_type }}()
    {%- elif database_type == 'postgresql' and model.type in POSTGRESQL_TYPES -%}
-    {%- else -%}
-        sa.{{ model.type }}()
-    {%- endif -%}, default=
-    {%- if model.is_nullable and model.default == None -%}
-        None
-    {%- else -%}
-        {%- if model.default != None -%}
-            '{{ model.default }}'
+        {%- if model.type == 'ARRAY' -%}
+        postgresql.ARRAY(sa.String)
        {%- else -%}
-            {%- if model.pd_type == 'str' -%}
-                ''
-            {%- elif model.pd_type == 'int' -%}
-                0
-            {%- elif model.pd_type == 'bytes' -%}
-                b''
-            {%- elif model.pd_type == 'bool' -%}
-                True
-            {%- elif model.pd_type == 'float' -%}
-                0.0
-            {%- elif model.pd_type == 'dict' -%}
-                {}
-            {%- elif model.pd_type == 'date' or model.pd_type == 'datetime' -%}
-                timezone.now()
-            {%- elif model.pd_type == 'list[str]' -%}
-                ()
-            {%- else -%}
-                ''
-            {%- endif -%}
+        postgresql.{{ model.type|sqlalchemy_type }}()
        {%- endif -%}
-    {%- endif -%}{% if model.sort != 0 %}, sort_order={{ model.sort }}{% endif %}, comment=
+    {%- else -%}
+        sa.{{ model.type|sqlalchemy_type }}()
+    {%- endif -%},
+    {%- if model.is_nullable and model.default == None %} default=None
+    {%- elif model.default != None %} default='{{ model.default }}'
+    {%- elif model.pd_type in FACTORY_TYPES %} default_factory={{ model.pd_type.split('[')[0] }}
+    {%- elif model.pd_type == 'str' %} default=''
+    {%- elif model.pd_type == 'int' %} default=0
+    {%- elif model.pd_type == 'bytes' %} default=b''
+    {%- elif model.pd_type == 'bool' %} default=True
+    {%- elif model.pd_type == 'float' %} default=0.0
+    {%- elif model.pd_type == 'date' %} default_factory=date.today
+    {%- elif model.pd_type == 'datetime' and model.type in ['TIMESTAMP', 'DateTime', 'TIMESTAMP WITHOUT TIME ZONE', 'TIMESTAMP WITH TIME ZONE'] %} default_factory=timezone.now
+    {%- elif model.pd_type == 'datetime' %} default_factory=datetime.now
+    {%- else %} default=None
+    {%- endif -%}, comment=
    {%- if model.comment != None -%}
        '{{ model.comment }}')
    {% else -%}
--- a/backend/plugin/code_generator/templates/python/schema.jinja
+++ b/backend/plugin/code_generator/templates/python/schema.jinja
@@ -1,5 +1,20 @@
-{% if default_datetime_column %}
-from datetime import datetime
+{% set pd_types = models|map(attribute='pd_type')|list %}
+{% set NEED_DATETIME = datetime_mixin or 'datetime' in pd_types %}
+{% set NEED_DATE = 'date' in pd_types %}
+{% set NEED_TIME = 'time' in pd_types %}
+{% set NEED_TIMEDELTA = 'timedelta' in pd_types %}
+{% set NEED_DECIMAL = 'Decimal' in pd_types %}
+{% set NEED_UUID = 'UUID' in pd_types or 'str | UUID' in pd_types %}
+{% if NEED_DATETIME or NEED_DATE or NEED_TIME or NEED_TIMEDELTA %}
+from datetime import {% if NEED_DATETIME %}datetime{% endif %}{% if NEED_DATE %}{% if NEED_DATETIME %}, {% endif %}date{% endif %}{% if NEED_TIME %}{% if NEED_DATETIME or NEED_DATE %}, {% endif %}time{% endif %}{% if NEED_TIMEDELTA %}{% if NEED_DATETIME or NEED_DATE or NEED_TIME %}, {% endif %}timedelta{% endif %}
+
+{% endif %}
+{% if NEED_DECIMAL %}
+from decimal import Decimal
+
+{% endif %}
+{% if NEED_UUID %}
+from uuid import UUID

 {% endif %}
 from pydantic import ConfigDict, Field
@@ -35,7 +50,7 @@ class Get{{ schema_name }}Detail({{ schema_name }}SchemaBase):
    model_config = ConfigDict(from_attributes=True)

    id: int
-    {% if default_datetime_column %}
+    {% if datetime_mixin %}
    created_time: datetime
    updated_time: datetime | None = None
    {% endif %}
--- a/backend/plugin/code_generator/templates/sql/mysql/init.jinja
+++ b/backend/plugin/code_generator/templates/sql/mysql/init.jinja
@@ -0,0 +1,11 @@
+insert into sys_menu (title, name, path, sort, icon, type, component, perms, status, display, cache, link, remark, parent_id, created_time, updated_time)
+values ('{{ doc_comment }}', '{{ schema_name }}', '/{{ app_name }}/{{ table_name|replace("_", "-") }}', 0, 'tabler:list', 1, '/{{ app_name }}/{{ table_name }}/views/index', null, 1, 1, 1, '', null, null, now(), null);
+
+set @parent_menu_id = LAST_INSERT_ID();
+
+insert into sys_menu (title, name, path, sort, icon, type, component, perms, status, display, cache, link, remark, parent_id, created_time, updated_time)
+values
+('新增', 'Add{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:add', 1, 0, 1, '', null, @parent_menu_id, now(), null),
+('修改', 'Edit{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:edit', 1, 0, 1, '', null, @parent_menu_id, now(), null),
+('删除', 'Delete{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:del', 1, 0, 1, '', null, @parent_menu_id, now(), null),
+('查询', 'Get{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:get', 1, 0, 1, '', null, @parent_menu_id, now(), null);
--- a/backend/plugin/code_generator/templates/sql/mysql/init_snowflake.jinja
+++ b/backend/plugin/code_generator/templates/sql/mysql/init_snowflake.jinja
@@ -0,0 +1,9 @@
+insert into sys_menu (id, title, name, path, sort, icon, type, component, perms, status, display, cache, link, remark, parent_id, created_time, updated_time)
+values ({{ parent_menu_id }}, '{{ doc_comment }}', '{{ schema_name }}', '/{{ app_name }}/{{ table_name|replace("_", "-") }}', 0, 'tabler:list', 1, '/{{ app_name }}/{{ table_name }}/views/index', null, 1, 1, 1, '', null, null, now(), null);
+
+insert into sys_menu (id, title, name, path, sort, icon, type, component, perms, status, display, cache, link, remark, parent_id, created_time, updated_time)
+values
+({{ button_ids[0] }}, '新增', 'Add{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:add', 1, 0, 1, '', null, {{ parent_menu_id }}, now(), null),
+({{ button_ids[1] }}, '修改', 'Edit{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:edit', 1, 0, 1, '', null, {{ parent_menu_id }}, now(), null),
+({{ button_ids[2] }}, '删除', 'Delete{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:del', 1, 0, 1, '', null, {{ parent_menu_id }}, now(), null),
+({{ button_ids[3] }}, '查询', 'Get{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:get', 1, 0, 1, '', null, {{ parent_menu_id }}, now(), null);
--- a/backend/plugin/code_generator/templates/sql/postgresql/init.jinja
+++ b/backend/plugin/code_generator/templates/sql/postgresql/init.jinja
@@ -0,0 +1,17 @@
+do $$
+declare
+    parent_menu_id bigint;
+begin
+    insert into sys_menu (title, name, path, sort, icon, type, component, perms, status, display, cache, link, remark, parent_id, created_time, updated_time)
+    values ('{{ doc_comment }}', '{{ schema_name }}', '/{{ app_name }}/{{ table_name|replace("_", "-") }}', 0, 'tabler:list', 1, '/{{ app_name }}/{{ table_name }}/views/index', null, 1, 1, 1, '', null, null, now(), null)
+    returning id into parent_menu_id;
+
+    insert into sys_menu (title, name, path, sort, icon, type, component, perms, status, display, cache, link, remark, parent_id, created_time, updated_time)
+    values
+    ('新增', 'Add{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:add', 1, 0, 1, '', null, parent_menu_id, now(), null),
+    ('修改', 'Edit{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:edit', 1, 0, 1, '', null, parent_menu_id, now(), null),
+    ('删除', 'Delete{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:del', 1, 0, 1, '', null, parent_menu_id, now(), null),
+    ('查询', 'Get{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:get', 1, 0, 1, '', null, parent_menu_id, now(), null);
+end $$;
+
+select setval(pg_get_serial_sequence('sys_menu', 'id'),coalesce(max(id), 0) + 1, true) from sys_menu;
--- a/backend/plugin/code_generator/templates/sql/postgresql/init_snowflake.jinja
+++ b/backend/plugin/code_generator/templates/sql/postgresql/init_snowflake.jinja
@@ -0,0 +1,9 @@
+insert into sys_menu (id, title, name, path, sort, icon, type, component, perms, status, display, cache, link, remark, parent_id, created_time, updated_time)
+values ({{ parent_menu_id }}, '{{ doc_comment }}', '{{ schema_name }}', '/{{ app_name }}/{{ table_name|replace("_", "-") }}', 0, 'tabler:list', 1, '/{{ app_name }}/{{ table_name }}/views/index', null, 1, 1, 1, '', null, null, now(), null);
+
+insert into sys_menu (id, title, name, path, sort, icon, type, component, perms, status, display, cache, link, remark, parent_id, created_time, updated_time)
+values
+({{ button_ids[0] }}, '新增', 'Add{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:add', 1, 0, 1, '', null, {{ parent_menu_id }}, now(), null),
+({{ button_ids[1] }}, '修改', 'Edit{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:edit', 1, 0, 1, '', null, {{ parent_menu_id }}, now(), null),
+({{ button_ids[2] }}, '删除', 'Delete{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:del', 1, 0, 1, '', null, {{ parent_menu_id }}, now(), null),
+({{ button_ids[3] }}, '查询', 'Get{{ schema_name }}', null, 0, null, 2, null, '{{ permission }}:get', 1, 0, 1, '', null, {{ parent_menu_id }}, now(), null);
--- a/backend/plugin/code_generator/utils/format_code.py
+++ b/backend/plugin/code_generator/utils/format_code.py
@@ -0,0 +1,43 @@
+import asyncio
+import tempfile
+import uuid
+
+import anyio
+
+from anyio import open_file
+
+
+async def format_python_code(code: str) -> str:
+    """
+    使用 ruff 格式化 Python 代码
+
+    :param code: 原始代码
+    :return:
+    """
+    temp_dir = anyio.Path(tempfile.gettempdir())
+    temp_file = temp_dir / f'fba_codegen_{uuid.uuid4().hex}.py'
+
+    try:
+        async with await open_file(temp_file, 'w', encoding='utf-8') as f:
+            await f.write(code)
+
+        process = await asyncio.create_subprocess_exec(
+            'ruff',
+            'format',
+            str(temp_file),
+            stdout=asyncio.subprocess.PIPE,
+            stderr=asyncio.subprocess.PIPE,
+        )
+        await process.communicate()
+
+        if process.returncode == 0:
+            async with await open_file(temp_file, encoding='utf-8') as f:
+                formatted_code = await f.read()
+        else:
+            formatted_code = code
+    except (FileNotFoundError, OSError):
+        return code
+    finally:
+        await temp_file.unlink(missing_ok=True)
+
+    return formatted_code
--- a/backend/plugin/code_generator/utils/gen_template.py
+++ b/backend/plugin/code_generator/utils/gen_template.py
@@ -1,10 +1,14 @@
 from collections.abc import Sequence

 from jinja2 import Environment, FileSystemLoader, Template, select_autoescape
+from pydantic.alias_generators import to_pascal

 from backend.core.conf import settings
 from backend.plugin.code_generator.model import GenBusiness, GenColumn
 from backend.plugin.code_generator.path_conf import JINJA2_TEMPLATE_DIR
+from backend.plugin.code_generator.utils.type_conversion import sql_type_to_sqlalchemy_name
+from backend.utils.snowflake import snowflake
+from backend.utils.timezone import timezone


 class GenTemplate:
@@ -18,60 +22,64 @@ class GenTemplate:
            keep_trailing_newline=True,
            enable_async=True,
        )
+        self.env.filters['sqlalchemy_type'] = sql_type_to_sqlalchemy_name
        self.init_content = ''

    def get_template(self, jinja_file: str) -> Template:
        """
-        获取模板文件
+        获取 Jinja2 模板对象

-        :param jinja_file: Jinja2 模板文件
-        :return:
+        :param jinja_file: Jinja2 模板文件路径
+        :return: Template 对象
        """
        return self.env.get_template(jinja_file)

    @staticmethod
-    def get_template_files() -> list[str]:
+    def get_template_path_mapping(business: GenBusiness) -> dict[str, str]:
        """
-        获取模板文件列表
-
-        :return:
-        """
-        return [
-            'python/api.jinja',
-            'python/crud.jinja',
-            'python/model.jinja',
-            'python/schema.jinja',
-            'python/service.jinja',
-        ]
-
-    @staticmethod
-    def get_code_gen_paths(business: GenBusiness) -> list[str]:
-        """
-        获取代码生成路径列表
+        获取模板文件到生成文件的路径映射

        :param business: 代码生成业务对象
-        :return:
+        :return: {模板路径: 生成文件路径}
        """
        app_name = business.app_name
        filename = business.filename
-        return [
-            f'{app_name}/api/{business.api_version}/{filename}.py',
-            f'{app_name}/crud/crud_{filename}.py',
-            f'{app_name}/model/{filename}.py',
-            f'{app_name}/schema/{filename}.py',
-            f'{app_name}/service/{filename}_service.py',
-        ]
+        api_version = business.api_version
+        pk_suffix = '_snowflake' if settings.DATABASE_PK_MODE == 'snowflake' else ''

-    def get_code_gen_path(self, tpl_path: str, business: GenBusiness) -> str:
-        """
-        获取代码生成路径
+        return {
+            'python/api.jinja': f'{app_name}/api/{api_version}/{filename}.py',
+            'python/crud.jinja': f'{app_name}/crud/crud_{filename}.py',
+            'python/model.jinja': f'{app_name}/model/{filename}.py',
+            'python/schema.jinja': f'{app_name}/schema/{filename}.py',
+            'python/service.jinja': f'{app_name}/service/{filename}_service.py',
+            f'sql/mysql/init{pk_suffix}.jinja': f'{app_name}/sql/mysql/init{pk_suffix}.sql',
+            f'sql/postgresql/init{pk_suffix}.jinja': f'{app_name}/sql/postgresql/init{pk_suffix}.sql',
+        }

-        :param tpl_path: 模板文件路径
-        :param business: 代码生成业务对象
-        :return:
+    def get_init_files(self, business: GenBusiness) -> dict[str, str]:
        """
-        code_gen_path_mapping = dict(zip(self.get_template_files(), self.get_code_gen_paths(business)))
-        return code_gen_path_mapping[tpl_path]
+        获取需要生成的 __init__.py 文件及其内容
+
+        :param business: 业务对象
+        :return: {相对路径: 文件内容}
+        """
+        app_name = business.app_name
+        table_name = business.table_name
+        class_name = business.class_name or to_pascal(table_name)
+
+        return {
+            f'{app_name}/__init__.py': self.init_content,
+            f'{app_name}/api/__init__.py': self.init_content,
+            f'{app_name}/api/{business.api_version}/__init__.py': self.init_content,
+            f'{app_name}/crud/__init__.py': self.init_content,
+            f'{app_name}/model/__init__.py': (
+                f'{self.init_content}'
+                f'from backend.app.{app_name}.model.{table_name} import {class_name} as {class_name}\n'
+            ),
+            f'{app_name}/schema/__init__.py': self.init_content,
+            f'{app_name}/service/__init__.py': self.init_content,
+        }

    @staticmethod
    def get_vars(business: GenBusiness, models: Sequence[GenColumn]) -> dict[str, str | Sequence[GenColumn]]:
@@ -82,19 +90,26 @@ class GenTemplate:
        :param models: 代码生成模型对象列表
        :return:
        """
-        return {
+        vars_dict = {
            'app_name': business.app_name,
            'table_name': business.table_name,
            'doc_comment': business.doc_comment,
            'table_comment': business.table_comment,
            'class_name': business.class_name,
            'schema_name': business.schema_name,
-            'default_datetime_column': business.default_datetime_column,
-            'permission': str(business.table_name.replace('_', ':')),
+            'datetime_mixin': business.datetime_mixin,
+            'permission': business.table_name.replace('_', ':'),
            'database_type': settings.DATABASE_TYPE,
            'models': models,
            'model_types': [model.type for model in models],
+            'now': timezone.now(),
        }

+        if settings.DATABASE_PK_MODE == 'snowflake':
+            vars_dict['parent_menu_id'] = snowflake.generate()
+            vars_dict['button_ids'] = [snowflake.generate() for _ in range(4)]
+
+        return vars_dict
+

 gen_template: GenTemplate = GenTemplate()
--- a/backend/plugin/code_generator/utils/type_conversion.py
+++ b/backend/plugin/code_generator/utils/type_conversion.py
@@ -17,7 +17,8 @@ def sql_type_to_sqlalchemy(typing: str) -> str:
        if typing in GenMySQLColumnType.get_member_keys():
            return typing
    else:
-        if typing in GenPostgreSQLColumnType.get_member_keys():
+        normalized_type = typing.replace(' ', '_')
+        if normalized_type in GenPostgreSQLColumnType.get_member_keys():
            return typing
    return 'String'

@@ -33,8 +34,31 @@ def sql_type_to_pydantic(typing: str) -> str:
    try:
        if DataBaseType.mysql == settings.DATABASE_TYPE:
            return GenMySQLColumnType[typing].value
-        if typing == 'CHARACTER VARYING':  # postgresql 中 DDL VARCHAR 的别名
-            return 'str'
-        return GenPostgreSQLColumnType[typing].value
+        normalized_type = typing.replace(' ', '_')
+        return GenPostgreSQLColumnType[normalized_type].value
    except KeyError:
        return 'str'
+
+
+@lru_cache(maxsize=128)
+def sql_type_to_sqlalchemy_name(typing: str) -> str:
+    """
+    将 SQL 类型转换为有效的 SQLAlchemy 类型名称(用于代码生成)
+
+    :param typing: SQL 类型字符串
+    :return:
+    """
+    pg_type_mapping = {
+        'CHARACTER VARYING': 'String',
+        'CHARACTER': 'CHAR',
+        'TIMESTAMP WITHOUT TIME ZONE': 'TIMESTAMP',
+        'TIMESTAMP WITH TIME ZONE': 'TIMESTAMP',
+        'TIME WITHOUT TIME ZONE': 'TIME',
+        'TIME WITH TIME ZONE': 'TIME',
+        'DOUBLE PRECISION': 'Double',
+    }
+
+    if DataBaseType.postgresql == settings.DATABASE_TYPE and typing in pg_type_mapping:
+        return pg_type_mapping[typing]
+
+    return typing
--- a/backend/utils/pattern_validate.py
+++ b/backend/utils/pattern_validate.py
@@ -76,3 +76,14 @@ def is_has_special_char(value: str) -> re.Match[str]:
    """
    special_char_pattern = r'[!@#$%^&*()_+\-=\[\]{};:\'",.<>?/\\|`~]'
    return search_string(special_char_pattern, value)
+
+
+def is_english_identifier(value: str) -> re.Match[str]:
+    """
+    检查英文标识符
+
+    :param value: 待检查的值
+    :return:
+    """
+    identifier_pattern = r'^[a-zA-Z][a-zA-Z_]*$'
+    return match_string(identifier_pattern, value)