Initial commit with production configuration and deployment scripts

sql/tools/.gitignore

@@ -0,0 +1,8 @@
+# 忽略python虚拟环境
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/

sql/tools/README.md

@@ -0,0 +1,130 @@
+## 0. 友情提示
+
+在 `sql/tools` 目录下，我们提供一些数据库相关的工具，包括测试数据库的快速启动、MySQL 转换其它数据库等等。
+
+注意！所有的操作，必须在 `sql/tools` 目录下执行。
+
+## 1. 测试数据库的快速启动
+
+基于 Docker Compose，快速启动 MySQL、Oracle、PostgreSQL、SQL Server 等数据库。
+
+注意！使用 Docker Compose 启动完测试数据后，因为会自动导入项目的 SQL 脚本，所以可能需要等待 1-2 分钟。
+
+### 1.1 MySQL
+
+```Bash
+docker compose up -d mysql
+```
+
+#### 1.2 Oracle
+
+```Bash
+## x86 版本
+docker compose up -d oracle
+
+## MacBook Apple Silicon
+docker compose up -d oracle_m1
+```
+
+> 注意：如果使用 MacBook Apple Silicon 版本，它的 ORACLE_SID 不是 XE，而是 FREE！！！
+
+### 1.3 PostgreSQL
+
+```Bash
+docker compose up -d postgres
+```
+
+### 1.4 SQL Server
+
+```Bash
+docker compose up -d sqlserver
+# 注意：启动完 sqlserver 后，需要手动再执行如下命令，因为 SQL Server 不支持初始化脚本
+docker compose exec sqlserver bash /tmp/create_schema.sh
+```
+
+### 1.5 DM 达梦
+
+① 下载达梦 Docker 镜像：<https://eco.dameng.com/download/> 地址，点击“Docker 镜像”选项，进行下载。
+
+② 加载镜像文件，在镜像 tar 文件所在目录运行：
+
+```Bash
+docker load -i dm8_20240715_x86_rh6_rq_single.tar
+```
+
+③ 在项目 `sql/tools` 目录下运行：
+
+```Bash
+docker compose up -d dm8
+# 注意：启动完 dm 后，需要手动再执行如下命令，因为 dm 不支持初始化脚本
+docker compose exec dm8 bash -c '/opt/dmdbms/bin/disql SYSDBA/SYSDBA001 \`/tmp/schema.sql'
+exit
+```
+
+### 1.6 KingbaseES 人大金仓
+
+① 下载人大金仓 Docker 镜像：
+
+* [x86_64 版本](https://kingbase.oss-cn-beijing.aliyuncs.com/KESV8R3/V009R001C001B0025-安装包-docker/x86_64/kdb_x86_64_V009R001C001B0025.tar) 【Windows 选择这个】
+* [aarch64 版本](https://kingbase.oss-cn-beijing.aliyuncs.com/KESV8R3/V009R001C001B0025-安装包-docker/aarch64/kdb_aarch64_V009R001C001B0025.tar) 【MacBook Apple Silicon 选择这个】
+
+② 加载镜像文件，在镜像 tar 文件所在目录运行：
+
+```Bash
+docker load -i kdb_x86_64_V009R001C001B0025.tar
+```
+
+③ 在项目 `sql/tools` 目录下运行：
+
+```Bash
+docker compose up -d kingbase
+# 注意：启动完 kingbase 后，需要手动再执行如下命令
+docker compose exec kingbase bash -c 'ksql -U $DB_USER -d test -f /tmp/schema.sql'
+```
+
+### 1.7 华为 OpenGauss
+
+```Bash
+docker compose up -d opengauss
+# 注意：启动完 opengauss 后，需要手动再执行如下命令
+docker compose exec opengauss bash -c '/usr/local/opengauss/bin/gsql -U $GS_USERNAME -W $GS_PASSWORD -d postgres -f /tmp/schema.sql'
+```
+
+## 1.X 容器的销毁重建
+
+开发测试过程中，有时候需要创建全新干净的数据库。由于测试数据 Docker 容器采用数据卷 Volume 挂载数据库实例的数据目录，因此销毁数据需要停止容器后，删除数据卷，然后再重新创建容器。
+
+以 postgres 为例，操作如下：
+
+```Bash
+docker compose down postgres
+docker volume rm ruoyi-vue-pro_postgres
+```
+
+## 2. MySQL 转换其它数据库
+
+项目提供了 `sql/tools/convertor.py` 脚本，支持将 MySQL 转换为 Oracle、PostgreSQL、SQL Server、达梦、人大金仓、OpenGauss 等数据库的脚本。
+
+### 2.1 实现原理
+
+通过读取 MySQL 的 `sql/mysql/ruoyi-vue-pro.sql` 数据库文件，转换成对应的数据库脚本。
+
+### 2.2 使用方法
+
+① 安装依赖库 `simple-ddl-parser`
+
+```bash
+pip install simple-ddl-parser
+# pip3 install simple-ddl-parser
+```
+
+② 在 `sql/tools/` 目录下，执行如下命令打印生成 postgres 的脚本内容，其他可选参数有：`oracle`、`sqlserver`、`dm8`、`kingbase`、`opengauss`：
+
+```Bash
+python3 convertor.py postgres
+# python3 convertor.py postgres > tmp.sql
+```
+
+程序将 SQL 脚本打印到终端，可以重定向到临时文件 `tmp.sql`。
+
+确认无误后，可以利用 IDEA 进行格式化。当然，也可以直接导入到数据库中。

sql/tools/convertor.py

@@ -0,0 +1,945 @@
+# encoding=utf8
+"""芋道系统数据库迁移工具
+
+Author: dhb52 (https://gitee.com/dhb52)
+
+pip install simple-ddl-parser
+
+or with uv
+uv run --with simple-ddl-parser convertor.py postgres > ../postgresql/ruoyi-vue-pro.sql                                         239ms  四  5/22 21:03:16 2025
+uv run --with simple-ddl-parser convertor.py sqlserver > ../sqlserver/ruoyi-vue-pro.sql
+uv run --with simple-ddl-parser convertor.py kingbase > ../kingbase/ruoyi-vue-pro.sql
+uv run --with simple-ddl-parser convertor.py opengauss > ../opengauss/ruoyi-vue-pro.sql
+uv run --with simple-ddl-parser convertor.py oracle > ../oracle/ruoyi-vue-pro.sql
+uv run --with simple-ddl-parser convertor.py dm8 > ../dm/ruoyi-vue-pro-dm8.sql
+"""
+
+import argparse
+import pathlib
+import re
+import sys
+import time
+from abc import ABC, abstractmethod
+from typing import Dict, Generator, Optional, Tuple, Union
+
+from simple_ddl_parser import DDLParser
+
+PREAMBLE = """/*
+ Aagro Database Transfer Tool
+
+ Source Server Type    : MySQL
+
+ Target Server Type    : {db_type}
+
+ Date: {date}
+*/
+
+"""
+
+
+def load_and_clean(sql_file: str) -> str:
+    """加载源 SQL 文件，并清理内容方便下一步 ddl 解析
+
+    Args:
+        sql_file (str): sql文件路径
+
+    Returns:
+        str: 清理后的sql文件内容
+    """
+    REPLACE_PAIR_LIST = (
+        (")\nVALUES ", ") VALUES "),
+        (" CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci ", " "),
+        (" KEY `", " INDEX `"),
+        ("UNIQUE INDEX", "UNIQUE KEY"),
+        ("b'0'", "'0'"),
+        ("b'1'", "'1'"),
+    )
+
+    content = open(sql_file, encoding="utf-8").read()
+    for replace_pair in REPLACE_PAIR_LIST:
+        content = content.replace(*replace_pair)
+    content = re.sub(r"ENGINE.*COMMENT", "COMMENT", content)
+    content = re.sub(r"ENGINE.*;", ";", content)
+    return content
+
+
+class Convertor(ABC):
+    def __init__(self, src: str, db_type) -> None:
+        self.src = src
+        self.db_type = db_type
+        self.content = load_and_clean(self.src)
+        self.table_script_list = re.findall(r"CREATE TABLE [^;]*;", self.content)
+
+    @abstractmethod
+    def translate_type(self, type: str, size: Optional[Union[int, Tuple[int]]]) -> str:
+        """字段类型转换
+
+        Args:
+            type (str): 字段类型
+            size (Optional[Union[int, Tuple[int]]]): 字段长度描述, 如varchar(255), decimal(10,2)
+
+        Returns:
+            str: 类型定义
+        """
+        pass
+
+    @abstractmethod
+    def gen_create(self, table_ddl: Dict) -> str:
+        """生成 create 脚本
+
+        Args:
+            table_ddl (Dict): 表DDL
+
+        Returns:
+            str:  生成脚本
+        """
+        pass
+
+    @abstractmethod
+    def gen_pk(self, table_name: str) -> str:
+        """生成主键定义
+
+        Args:
+            table_name (str): 表名
+
+        Returns:
+            str: 生成脚本
+        """
+        pass
+
+    @abstractmethod
+    def gen_index(self, ddl: Dict) -> str:
+        """生成索引定义
+
+        Args:
+            table_ddl (Dict): 表DDL
+
+        Returns:
+            str: 生成脚本
+        """
+        pass
+
+    @abstractmethod
+    def gen_comment(self, table_ddl: Dict) -> str:
+        """生成字段/表注释
+
+        Args:
+            table_ddl (Dict): 表DDL
+
+        Returns:
+            str: 生成脚本
+        """
+        pass
+
+    @abstractmethod
+    def gen_uk(self, table_ddl: Dict) -> str:
+        """生成
+
+        Args:
+            table_ddl (Dict): 表DDL
+
+        Returns:
+            str: 生成脚本
+        """
+
+    @abstractmethod
+    def gen_insert(self, table_name: str) -> str:
+        """生成 insert 语句块
+
+        Args:
+            table_name (str): 表名
+
+        Returns:
+            str: 生成脚本
+        """
+        pass
+
+    def gen_dual(self) -> str:
+        """生成虚拟 dual 表
+
+        Returns:
+            str: 生成脚本, 默认返回空脚本, 表示当前数据库无需手工创建
+        """
+        return ""
+
+    @staticmethod
+    def inserts(table_name: str, script_content: str) -> Generator:
+        PREFIX = f"INSERT INTO `{table_name}`"
+
+        # 收集 `table_name` 对应的 insert 语句
+        for line in script_content.split("\n"):
+            if line.startswith(PREFIX):
+                head, tail = line.replace(PREFIX, "").split(" VALUES ", maxsplit=1)
+                head = head.strip().replace("`", "").lower()
+                tail = tail.strip().replace(r"\"", '"')
+                # tail = tail.replace("b'0'", "'0'").replace("b'1'", "'1'")
+                yield f"INSERT INTO {table_name.lower()} {head} VALUES {tail}"
+
+    @staticmethod
+    def index(ddl: Dict) -> Generator:
+        """生成索引定义
+
+        Args:
+            ddl (Dict): 表DDL
+
+        Yields:
+            Generator[str]: create index 语句
+        """
+
+        def generate_columns(columns):
+            keys = [
+                f"{col['name'].lower()}{' ' + col['order'].lower() if col['order'] != 'ASC' else ''}"
+                for col in columns[0]
+            ]
+            return ", ".join(keys)
+
+        for no, index in enumerate(ddl["index"], 1):
+            columns = generate_columns(index["columns"])
+            table_name = ddl["table_name"].lower()
+            yield f"CREATE INDEX idx_{table_name}_{no:02d} ON {table_name} ({columns})"
+
+    @staticmethod
+    def unique_index(ddl: Dict) -> Generator:
+        if "constraints" in ddl and "uniques" in ddl["constraints"]:
+            uk_list = ddl["constraints"]["uniques"]
+            for uk in uk_list:
+                table_name = ddl["table_name"]
+                uk_name = uk["constraint_name"]
+                uk_columns = uk["columns"]
+                yield table_name, uk_name, uk_columns
+
+    @staticmethod
+    def filed_comments(table_sql: str) -> Generator:
+        for line in table_sql.split("\n"):
+            match = re.match(r"^`([^`]+)`.* COMMENT '([^']+)'", line.strip())
+            if match:
+                field = match.group(1)
+                comment_string = match.group(2).replace("\\n", "\n")
+                yield field, comment_string
+
+    def table_comment(self, table_sql: str) -> str:
+        match = re.search(r"COMMENT \='([^']+)';", table_sql)
+        return match.group(1) if match else None
+
+    def print(self):
+        """打印转换后的sql脚本到终端"""
+        print(
+            PREAMBLE.format(
+                db_type=self.db_type,
+                date=time.strftime("%Y-%m-%d %H:%M:%S"),
+            )
+        )
+
+        dual = self.gen_dual()
+        if dual:
+            print(
+                f"""-- ----------------------------
+-- Table structure for dual
+-- ----------------------------
+{dual}
+"""
+            )
+
+        error_scripts = []
+        for table_sql in self.table_script_list:
+            ddl = DDLParser(table_sql.replace("`", "")).run()
+
+            # 如果parse失败, 需要跟进
+            if len(ddl) == 0:
+                error_scripts.append(table_sql)
+                continue
+
+            table_ddl = ddl[0]
+            table_name = table_ddl["table_name"]
+
+            # 忽略 quartz 的内容
+            if table_name.lower().startswith("qrtz"):
+                continue
+
+            # 解析注释
+            for column in table_ddl["columns"]:
+                column["comment"] = bytes(column["comment"], "utf-8").decode(
+                    "unicode_escape"
+                )[1:-1]
+            table_ddl["comment"] = bytes(table_ddl["comment"], "utf-8").decode(
+                "unicode_escape"
+            )[1:-1]
+
+            # 为每个表生成个6个基本部分
+            create = self.gen_create(table_ddl)
+            pk = self.gen_pk(table_name)
+            uk = self.gen_uk(table_ddl)
+            index = self.gen_index(table_ddl)
+            comment = self.gen_comment(table_ddl)
+            inserts = self.gen_insert(table_name)
+
+            # 组合当前表的DDL脚本
+            script = f"""{create}
+
+{pk}
+
+{uk}
+
+{index}
+
+{comment}
+
+{inserts}
+"""
+
+            # 清理
+            script = re.sub("\n{3,}", "\n\n", script).strip() + "\n"
+
+            print(script)
+
+        # 将parse失败的脚本打印出来
+        if error_scripts:
+            print("!!! 以下内容无法正常解析", file=sys.stderr)
+            for script in error_scripts:
+                # print to stderr
+                print(script, file=sys.stderr)
+
+
+class PostgreSQLConvertor(Convertor):
+    def __init__(self, src):
+        super().__init__(src, "PostgreSQL")
+
+    def translate_type(self, type: str, size: Optional[Union[int, Tuple[int]]]):
+        """类型转换"""
+
+        type = type.lower()
+
+        if type == "varchar":
+            return f"varchar({size})"
+        if type in ("int", "int unsigned"):
+            return "int4"
+        if type in ("bigint", "bigint unsigned"):
+            return "int8"
+        if type == "datetime":
+            return "timestamp"
+        if type == "timestamp":
+            return f"timestamp({size})"
+        if type == "bit":
+            return "bool"
+        if type in ("tinyint", "smallint"):
+            return "int2"
+        if type in ("text", "longtext"):
+            return "text"
+        if type in ("blob", "mediumblob"):
+            return "bytea"
+        if type == "decimal":
+            return (
+                f"numeric({','.join(str(s) for s in size)})" if len(size) else "numeric"
+            )
+
+    def gen_create(self, ddl: Dict) -> str:
+        """生成 create"""
+
+        def _generate_column(col):
+            name = col["name"].lower()
+            if name == "deleted":
+                return "deleted int2 NOT NULL DEFAULT 0"
+
+            type = col["type"].lower()
+            full_type = self.translate_type(type, col["size"])
+            nullable = "NULL" if col["nullable"] else "NOT NULL"
+            default = f"DEFAULT {col['default']}" if col["default"] is not None else ""
+            return f"{name} {full_type} {nullable} {default}"
+
+        table_name = ddl["table_name"].lower()
+        columns = [f"{_generate_column(col).strip()}" for col in ddl["columns"]]
+        filed_def_list = ",\n  ".join(columns)
+        script = f"""-- ----------------------------
+-- Table structure for {table_name}
+-- ----------------------------
+DROP TABLE IF EXISTS {table_name};
+CREATE TABLE {table_name} (
+    {filed_def_list}
+);"""
+
+        return script
+
+    def gen_index(self, ddl: Dict) -> str:
+        return "\n".join(f"{script};" for script in self.index(ddl))
+
+    def gen_comment(self, table_ddl: Dict) -> str:
+        """生成字段及表的注释"""
+
+        script = ""
+        for column in table_ddl["columns"]:
+            table_comment = column["comment"]
+            script += (
+                f"COMMENT ON COLUMN {table_ddl['table_name']}.{column['name']} IS '{table_comment}';"
+                + "\n"
+            )
+
+        table_comment = table_ddl["comment"]
+        if table_comment:
+            script += (
+                f"COMMENT ON TABLE {table_ddl['table_name']} IS '{table_comment}';\n"
+            )
+
+        return script
+
+    def gen_pk(self, table_name) -> str:
+        """生成主键定义"""
+        return f"ALTER TABLE {table_name} ADD CONSTRAINT pk_{table_name} PRIMARY KEY (id);\n"
+
+    def gen_uk(self, table_ddl: Dict) -> str:
+        script = ""
+        uk_list = list(Convertor.unique_index(table_ddl))
+        for idx, (table_name, _, uk_columns) in enumerate(uk_list, 1):
+            uk_name = f"uk_{table_name}_{idx:02d}"
+            script += f"CREATE UNIQUE INDEX {uk_name} ON {table_name} ({', '.join(uk_columns)});\n"
+
+        return script
+
+    def gen_insert(self, table_name: str) -> str:
+        """生成 insert 语句，以及根据最后的 insert id+1 生成 Sequence"""
+
+        inserts = list(Convertor.inserts(table_name, self.content))
+        ## 生成 insert 脚本
+        script = ""
+        last_id = 0
+        if inserts:
+            inserts_lines = "\n".join(inserts)
+            script += f"""\n\n-- ----------------------------
+-- Records of {table_name.lower()}
+-- ----------------------------
+-- @formatter:off
+BEGIN;
+{inserts_lines}
+COMMIT;
+-- @formatter:on"""
+            match = re.search(r"VALUES \((\d+),", inserts[-1])
+            if match:
+                last_id = int(match.group(1))
+
+        # 生成 Sequence
+        script += (
+            "\n\n"
+            + f"""DROP SEQUENCE IF EXISTS {table_name}_seq;
+CREATE SEQUENCE {table_name}_seq
+    START {last_id + 1};"""
+        )
+
+        return script
+
+    def gen_dual(self) -> str:
+        return """DROP TABLE IF EXISTS dual;
+CREATE TABLE dual
+(
+    id int2
+);
+
+COMMENT ON TABLE dual IS '数据库连接的表';
+
+-- ----------------------------
+-- Records of dual
+-- ----------------------------
+-- @formatter:off
+INSERT INTO dual VALUES (1);
+-- @formatter:on"""
+
+
+class OracleConvertor(Convertor):
+    def __init__(self, src):
+        super().__init__(src, "Oracle")
+
+    def translate_type(self, type: str, size: Optional[Union[int, Tuple[int]]]):
+        """类型转换"""
+        type = type.lower()
+
+        if type == "varchar":
+            return f"varchar2({size if size < 4000 else 4000})"
+        if type in ("int", "int unsigned"):
+            return "number"
+        if type == "bigint" or type == "bigint unsigned":
+            return "number"
+        if type == "datetime":
+            return "date"
+        if type == "timestamp":
+            return f"timestamp({size})"
+        if type == "bit":
+            return "number(1,0)"
+        if type in ("tinyint", "smallint"):
+            return "smallint"
+        if type in ("text", "longtext"):
+            return "clob"
+        if type in ("blob", "mediumblob"):
+            return "blob"
+        if type == "decimal":
+            return (
+                f"number({','.join(str(s) for s in size)})" if len(size) else "number"
+            )
+
+    def gen_create(self, ddl) -> str:
+        """生成 CREATE 语句"""
+
+        def generate_column(col):
+            name = col["name"].lower()
+            if name == "deleted":
+                return "deleted number(1,0) DEFAULT 0 NOT NULL"
+
+            type = col["type"].lower()
+            full_type = self.translate_type(type, col["size"])
+            nullable = "NULL" if col["nullable"] else "NOT NULL"
+            # Oracle的 INSERT '' 不能通过NOT NULL校验，因此对文字类型字段覆写为 NULL
+            nullable = "NULL" if type in ("varchar", "text", "longtext") else nullable
+            default = f"DEFAULT {col['default']}" if col["default"] is not None else ""
+            # Oracle 中 size 不能作为字段名
+            field_name = '"size"' if name == "size" else name
+            # Oracle DEFAULT 定义在 NULLABLE 之前
+            return f"{field_name} {full_type} {default} {nullable}"
+
+        table_name = ddl["table_name"].lower()
+        columns = [f"{generate_column(col).strip()}" for col in ddl["columns"]]
+        field_def_list = ",\n    ".join(columns)
+        script = f"""-- ----------------------------
+-- Table structure for {table_name}
+-- ----------------------------
+CREATE TABLE {table_name} (
+    {field_def_list}
+);"""
+
+        # oracle INSERT '' 不能通过 NOT NULL 校验
+        script = script.replace("DEFAULT '' NOT NULL", "DEFAULT '' NULL")
+
+        return script
+
+    def gen_index(self, ddl: Dict) -> str:
+        return "\n".join(f"{script};" for script in self.index(ddl))
+
+    def gen_comment(self, table_ddl: Dict) -> str:
+        script = ""
+        for column in table_ddl["columns"]:
+            table_comment = column["comment"]
+            script += (
+                f"COMMENT ON COLUMN {table_ddl['table_name']}.{column['name']} IS '{table_comment}';"
+                + "\n"
+            )
+
+        table_comment = table_ddl["comment"]
+        if table_comment:
+            script += (
+                f"COMMENT ON TABLE {table_ddl['table_name']} IS '{table_comment}';\n"
+            )
+
+        return script
+
+    def gen_pk(self, table_name: str) -> str:
+        """生成主键定义"""
+        return f"ALTER TABLE {table_name} ADD CONSTRAINT pk_{table_name} PRIMARY KEY (id);\n"
+
+    def gen_uk(self, table_ddl: Dict) -> str:
+        script = ""
+        uk_list = list(Convertor.unique_index(table_ddl))
+        for idx, (table_name, _, uk_columns) in enumerate(uk_list, 1):
+            uk_name = f"uk_{table_name}_{idx:02d}"
+            script += f"CREATE UNIQUE INDEX {uk_name} ON {table_name} ({', '.join(uk_columns)});\n"
+
+        return script
+
+    def gen_index(self, ddl: Dict) -> str:
+        return "\n".join(f"{script};" for script in self.index(ddl))
+
+    def gen_insert(self, table_name: str) -> str:
+        """拷贝 INSERT 语句"""
+        inserts = []
+        for insert_script in Convertor.inserts(table_name, self.content):
+            # 对日期数据添加 TO_DATE 转换
+            insert_script = re.sub(
+                r"('\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}')",
+                r"to_date(\g<1>, 'SYYYY-MM-DD HH24:MI:SS')",
+                insert_script,
+            )
+            inserts.append(insert_script)
+
+        ## 生成 insert 脚本
+        script = ""
+        last_id = 0
+        if inserts:
+            inserts_lines = "\n".join(inserts)
+            script += f"""\n\n-- ----------------------------
+-- Records of {table_name.lower()}
+-- ----------------------------
+-- @formatter:off
+{inserts_lines}
+COMMIT;
+-- @formatter:on"""
+            match = re.search(r"VALUES \((\d+),", inserts[-1])
+            if match:
+                last_id = int(match.group(1))
+
+        # 生成 Sequence
+        script += f"""
+
+CREATE SEQUENCE {table_name}_seq
+    START WITH {last_id + 1};"""
+
+        return script
+
+
+class SQLServerConvertor(Convertor):
+    """_summary_
+
+    Args:
+        Convertor (_type_): _description_
+    """
+
+    def __init__(self, src):
+        super().__init__(src, "Microsoft SQL Server")
+
+    def translate_type(self, type: str, size: Optional[Union[int, Tuple[int]]]):
+        """类型转换"""
+
+        type = type.lower()
+
+        if type == "varchar":
+            return f"nvarchar({size if size < 4000 else 4000})"
+        if type in ("int", "int unsigned"):
+            return "int"
+        if type in ("bigint", "bigint unsigned"):
+            return "bigint"
+        if type in ("datetime", "timestamp"):
+            return "datetime2"
+        if type == "bit":
+            return "varchar(1)"
+        if type in ("tinyint", "smallint"):
+            return "tinyint"
+        if type in ("text", "longtext"):
+            return "nvarchar(max)"
+        if type in ("blob", "mediumblob"):
+            return "varbinary(max)"
+        if type == "decimal":
+            return (
+                f"numeric({','.join(str(s) for s in size)})" if len(size) else "numeric"
+            )
+
+    def gen_create(self, ddl: Dict) -> str:
+        """生成 create"""
+
+        def _generate_column(col):
+            name = col["name"].lower()
+            if name == "id":
+                return "id bigint NOT NULL PRIMARY KEY IDENTITY"
+            if name == "deleted":
+                return "deleted bit DEFAULT 0 NOT NULL"
+
+            type = col["type"].lower()
+            full_type = self.translate_type(type, col["size"])
+            nullable = "NULL" if col["nullable"] else "NOT NULL"
+            default = f"DEFAULT {col['default']}" if col["default"] is not None else ""
+            return f"{name} {full_type} {default} {nullable}"
+
+        table_name = ddl["table_name"].lower()
+        columns = [f"{_generate_column(col).strip()}" for col in ddl["columns"]]
+        filed_def_list = ",\n    ".join(columns)
+        script = f"""-- ----------------------------
+-- Table structure for {table_name}
+-- ----------------------------
+DROP TABLE IF EXISTS {table_name}
+GO
+CREATE TABLE {table_name} (
+    {filed_def_list}
+)
+GO"""
+
+        return script
+
+    def gen_comment(self, table_ddl: Dict) -> str:
+        """生成字段及表的注释"""
+
+        script = ""
+        table_name = table_ddl["table_name"]
+
+        for column in table_ddl["columns"]:
+            column_comment = column["comment"]
+            field = column["name"]
+
+            script += f"""EXEC sp_addextendedproperty
+    'MS_Description', N'{column_comment}',
+    'SCHEMA', N'dbo',
+    'TABLE', N'{table_name}',
+    'COLUMN', N'{field}'
+GO
+
+"""
+
+        table_comment = table_ddl["comment"]
+        if table_comment:
+            script += f"""EXEC sp_addextendedproperty
+    'MS_Description', N'{table_comment}',
+    'SCHEMA', N'dbo',
+    'TABLE', N'{table_name}'
+GO
+
+"""
+        return script
+
+    def gen_pk(self, table_name: str) -> str:
+        """生成主键定义"""
+        return ""
+
+    def gen_uk(self, table_ddl: Dict) -> str:
+        script = ""
+        uk_list = list(Convertor.unique_index(table_ddl))
+        for idx, (table_name, _, uk_columns) in enumerate(uk_list, 1):
+            uk_name = f"uk_{table_name}_{idx:02d}"
+            script += f"CREATE UNIQUE INDEX {uk_name} ON {table_name} ({', '.join(uk_columns)})\nGO"
+
+        return script
+
+    def gen_index(self, ddl: Dict) -> str:
+        """生成 index"""
+        return "\n".join(f"{script}\nGO" for script in self.index(ddl))
+
+    def gen_insert(self, table_name: str) -> str:
+        """生成 insert 语句"""
+
+        # 收集 `table_name` 对应的 insert 语句
+        inserts = []
+        for insert_script in Convertor.inserts(table_name, self.content):
+            # SQLServer: 字符串前加N，hack，是否存在替换字符串内容的风险
+            insert_script = insert_script.replace(", '", ", N'").replace(
+                "VALUES ('", "VALUES (N')"
+            )
+            # 删除 insert 的结尾分号
+            insert_script = re.sub(";$", r"\nGO", insert_script)
+            inserts.append(insert_script)
+
+        ## 生成 insert 脚本
+        script = ""
+        if inserts:
+            inserts_lines = "\n".join(inserts)
+            script += f"""\n\n-- ----------------------------
+-- Records of {table_name.lower()}
+-- ----------------------------
+-- @formatter:off
+BEGIN TRANSACTION
+GO
+SET IDENTITY_INSERT {table_name.lower()} ON
+GO
+{inserts_lines}
+SET IDENTITY_INSERT {table_name.lower()} OFF
+GO
+COMMIT
+GO
+-- @formatter:on"""
+
+        return script
+
+    def gen_dual(self) -> str:
+        return """DROP TABLE IF EXISTS dual
+GO
+CREATE TABLE dual
+(
+  id int
+)
+GO
+
+EXEC sp_addextendedproperty
+    'MS_Description', N'数据库连接的表',
+    'SCHEMA', N'dbo',
+    'TABLE', N'dual'
+GO
+
+-- ----------------------------
+-- Records of dual
+-- ----------------------------
+-- @formatter:off
+INSERT INTO dual VALUES (1)
+GO
+-- @formatter:on"""
+
+
+class DM8Convertor(Convertor):
+    def __init__(self, src):
+        super().__init__(src, "DM8")
+
+    def translate_type(self, type: str, size: Optional[Union[int, Tuple[int]]]):
+        """类型转换"""
+        type = type.lower()
+
+        if type == "varchar":
+            return f"varchar({size})"
+        if type in ("int", "int unsigned"):
+            return "int"
+        if type in ("bigint", "bigint unsigned"):
+            return "bigint"
+        if type == "datetime":
+            return "datetime"
+        if type == "timestamp":
+            return f"timestamp({size})"
+        if type == "bit":
+            return "bit"
+        if type in ("tinyint", "smallint"):
+            return "smallint"
+        if type in ("text", "longtext"):
+            return "text"
+        if type in ("blob", "mediumblob"):
+            return "blob"
+        if type == "decimal":
+            return (
+                f"decimal({','.join(str(s) for s in size)})" if len(size) else "decimal"
+            )
+
+    def gen_create(self, ddl) -> str:
+        """生成 CREATE 语句"""
+
+        def generate_column(col):
+            name = col["name"].lower()
+            if name == "id":
+                return "id bigint NOT NULL PRIMARY KEY IDENTITY"
+
+            type = col["type"].lower()
+            full_type = self.translate_type(type, col["size"])
+            nullable = "NULL" if col["nullable"] else "NOT NULL"
+            default = f"DEFAULT {col['default']}" if col["default"] is not None else ""
+            return f"{name} {full_type} {default} {nullable}"
+
+        table_name = ddl["table_name"].lower()
+        columns = [f"{generate_column(col).strip()}" for col in ddl["columns"]]
+        field_def_list = ",\n    ".join(columns)
+        script = f"""-- ----------------------------
+-- Table structure for {table_name}
+-- ----------------------------
+CREATE TABLE {table_name} (
+    {field_def_list}
+);"""
+
+        # oracle INSERT '' 不能通过 NOT NULL 校验
+        script = script.replace("DEFAULT '' NOT NULL", "DEFAULT '' NULL")
+
+        return script
+
+    def gen_comment(self, table_ddl: Dict) -> str:
+        script = ""
+        for column in table_ddl["columns"]:
+            table_comment = column["comment"]
+            script += (
+                f"COMMENT ON COLUMN {table_ddl['table_name']}.{column['name']} IS '{table_comment}';"
+                + "\n"
+            )
+
+        table_comment = table_ddl["comment"]
+        if table_comment:
+            script += (
+                f"COMMENT ON TABLE {table_ddl['table_name']} IS '{table_comment}';\n"
+            )
+
+        return script
+
+    def gen_pk(self, table_name: str) -> str:
+        """生成主键定义"""
+        return ""
+
+    def gen_uk(self, table_ddl: Dict) -> str:
+        script = ""
+        uk_list = list(Convertor.unique_index(table_ddl))
+        for idx, (table_name, _, uk_columns) in enumerate(uk_list, 1):
+            uk_name = f"uk_{table_name}_{idx:02d}"
+            script += f"CREATE UNIQUE INDEX {uk_name} ON {table_name} ({', '.join(uk_columns)});\n"
+
+        return script
+
+    def gen_index(self, ddl: Dict) -> str:
+        return "\n".join(f"{script};" for script in self.index(ddl))
+
+    def gen_insert(self, table_name: str) -> str:
+        """拷贝 INSERT 语句"""
+        inserts = list(Convertor.inserts(table_name, self.content))
+
+        ## 生成 insert 脚本
+        script = ""
+        if inserts:
+            inserts_lines = "\n".join(inserts)
+            script += f"""\n\n-- ----------------------------
+-- Records of {table_name.lower()}
+-- ----------------------------
+-- @formatter:off
+SET IDENTITY_INSERT {table_name.lower()} ON;
+{inserts_lines}
+COMMIT;
+SET IDENTITY_INSERT {table_name.lower()} OFF;
+-- @formatter:on"""
+
+        return script
+
+
+class KingbaseConvertor(PostgreSQLConvertor):
+    def __init__(self, src):
+        super().__init__(src)
+        self.db_type = "Kingbase"
+
+    def gen_create(self, ddl: Dict) -> str:
+        """生成 create"""
+
+        def _generate_column(col):
+            name = col["name"].lower()
+            if name == "deleted":
+                return "deleted int2 NOT NULL DEFAULT 0"
+
+            type = col["type"].lower()
+            full_type = self.translate_type(type, col["size"])
+            nullable = "NULL" if col["nullable"] else "NOT NULL"
+            if full_type == "text":
+                nullable = "NULL"
+            default = f"DEFAULT {col['default']}" if col["default"] is not None else ""
+            return f"{name} {full_type} {nullable} {default}"
+
+        table_name = ddl["table_name"].lower()
+        columns = [f"{_generate_column(col).strip()}" for col in ddl["columns"]]
+        filed_def_list = ",\n  ".join(columns)
+        script = f"""-- ----------------------------
+-- Table structure for {table_name}
+-- ----------------------------
+DROP TABLE IF EXISTS {table_name};
+CREATE TABLE {table_name} (
+    {filed_def_list}
+);"""
+
+        # Kingbase INSERT '' 不能通过 NOT NULL 校验
+        script = script.replace("NOT NULL DEFAULT ''", "NULL DEFAULT ''")
+
+        return script
+
+
+class OpengaussConvertor(KingbaseConvertor):
+    def __init__(self, src):
+        super().__init__(src)
+        self.db_type = "OpenGauss"
+
+
+def main():
+    parser = argparse.ArgumentParser(description="芋道系统数据库转换工具")
+    parser.add_argument(
+        "type",
+        type=str,
+        help="目标数据库类型",
+        choices=["postgres", "oracle", "sqlserver", "dm8", "kingbase", "opengauss"],
+    )
+    args = parser.parse_args()
+
+    sql_file = pathlib.Path("../mysql/ruoyi-vue-pro.sql").resolve().as_posix()
+    convertor = None
+    if args.type == "postgres":
+        convertor = PostgreSQLConvertor(sql_file)
+    elif args.type == "oracle":
+        convertor = OracleConvertor(sql_file)
+    elif args.type == "sqlserver":
+        convertor = SQLServerConvertor(sql_file)
+    elif args.type == "dm8":
+        convertor = DM8Convertor(sql_file)
+    elif args.type == "kingbase":
+        convertor = KingbaseConvertor(sql_file)
+    elif args.type == "opengauss":
+        convertor = OpengaussConvertor(sql_file)
+    else:
+        raise NotImplementedError(f"不支持目标数据库类型: {args.type}")
+
+    convertor.print()
+
+
+if __name__ == "__main__":
+    main()

sql/tools/docker-compose.yaml

@@ -0,0 +1,134 @@
+name: ruoyi-vue-pro
+
+volumes:
+    mysql: { }
+    postgres: { }
+    sqlserver: { }
+    dm8: { }
+    kingbase: { }
+    opengauss: { }
+
+services:
+    mysql:
+        image: mysql:8.0.33
+        restart: unless-stopped
+        environment:
+            TZ: Asia/Shanghai
+            MYSQL_ROOT_PASSWORD: 123456
+            MYSQL_DATABASE: ruoyi-vue-pro
+        ports:
+            - "3306:3306"
+        volumes:
+            - mysql:/var/lib/mysql/
+            # 注入初始化脚本
+            - ./mysql/ruoyi-vue-pro.sql:/docker-entrypoint-initdb.d/init.sql:ro
+        command:
+            --default-authentication-plugin=mysql_native_password
+            --character-set-server=utf8mb4
+            --collation-server=utf8mb4_general_ci
+            --explicit_defaults_for_timestamp=true
+            --lower_case_table_names=1
+
+    postgres:
+        image: postgres:14.2
+        restart: unless-stopped
+        environment:
+            POSTGRES_USER: root
+            POSTGRES_PASSWORD: 123456
+            POSTGRES_DB: ruoyi-vue-pro
+        ports:
+            - "5432:5432"
+        volumes:
+            - postgres:/var/lib/postgresql/data
+            # 注入初始化脚本
+            - ../postgresql/quartz.sql:/docker-entrypoint-initdb.d/quartz.sql:ro
+            - ../postgresql/ruoyi-vue-pro.sql:/docker-entrypoint-initdb.d/ruoyi-vue-pro.sql:ro
+
+    oracle:
+        image: gvenzl/oracle-xe:18-slim-faststart
+        restart: unless-stopped
+        environment:
+            ## 登录信息 SID: XE user: system password: oracle
+            ORACLE_PASSWORD: oracle
+        ports:
+            - "1521:1521"
+        volumes:
+            - ../oracle/ruoyi-vue-pro.sql:/tmp/schema.sql:ro
+            # 创建app用户: ROOT/123456@//localhost/XEPDB1
+            - ./oracle/1_create_user.sql:/docker-entrypoint-initdb.d/1_create_user.sql:ro
+            - ./oracle/2_create_schema.sh:/docker-entrypoint-initdb.d/2_create_schema.sh:ro
+
+    oracle_m1:
+      image: einslib/oracle-19c:19.3.0-ee-slim-faststart
+      restart: unless-stopped
+      environment:
+        ## 登录信息 SID: FREE user: system password: oracle
+        ORACLE_PASSWORD: oracle
+      ports:
+        - "1521:1521"
+      volumes:
+        - ../oracle/ruoyi-vue-pro.sql:/tmp/schema.sql:ro
+        # 创建app用户: ROOT/123456@//localhost/XEPDB1
+        - ./oracle/1_create_user.sql:/docker-entrypoint-initdb.d/1_create_user.sql:ro
+        - ./oracle/2_create_schema.sh:/docker-entrypoint-initdb.d/2_create_schema.sh:ro
+
+    sqlserver:
+        image: mcr.microsoft.com/mssql/server:2017-latest
+        restart: unless-stopped
+        environment:
+            TZ: Asia/Shanghai
+            ACCEPT_EULA: "Y"
+            SA_PASSWORD: "Aagro@2024"
+        ports:
+            - "1433:1433"
+        volumes:
+            - sqlserver:/var/opt/mssql
+            - ../sqlserver/ruoyi-vue-pro.sql:/tmp/schema.sql:ro
+            # docker compose exec sqlserver bash /tmp/create_schema.sh
+            - ./sqlserver/create_schema.sh:/tmp/create_schema.sh:ro
+
+    dm8:
+        # docker load -i dm8_20240715_x86_rh6_rq_single.tar
+        image: dm8_single:dm8_20240715_rev232765_x86_rh6_64
+        restart: unless-stopped
+        environment:
+            PAGE_SIZE: 16
+            LD_LIBRARY_PATH: /opt/dmdbms/bin
+            EXTENT_SIZE: 32
+            BLANK_PAD_MODE: 1
+            LOG_SIZE: 1024
+            UNICODE_FLAG: 1
+            LENGTH_IN_CHAR: 1
+            INSTANCE_NAME: dm8_test
+        ports:
+            - "5236:5236"
+        volumes:
+            - dm8:/opt/dmdbms/data
+            - ../dm/ruoyi-vue-pro-dm8.sql:/tmp/schema.sql:ro
+
+    kingbase:
+        image: kingbase_v009r001c001b0025_single_x86:v1
+#        image: kingbase_v009r001c001b0025_single_arm:v1
+        restart: unless-stopped
+        environment:
+            DB_USER: root
+            DB_PASSWORD: 123456
+        ports:
+            - "54321:54321"
+        volumes:
+            - kingbase:/home/kingbase/userdata
+            - ../kingbase/ruoyi-vue-pro.sql:/tmp/schema.sql:ro
+
+    opengauss:
+        image: opengauss/opengauss:5.0.0
+        restart: unless-stopped
+        environment:
+            GS_USERNAME: root
+            GS_PASSWORD: Aagro@2024
+            LD_LIBRARY_PATH: /usr/local/opengauss/lib:/usr/lib
+        ports:
+            - "5432:5432"
+        volumes:
+            - opengauss:/var/lib/opengauss
+            - ../opengauss/ruoyi-vue-pro.sql:/tmp/schema.sql:ro
+            # docker compose exec opengauss bash -c '/usr/local/opengauss/bin/gsql -U $GS_USERNAME -W $GS_PASSWORD -d postgres -f /tmp/schema.sql'

sql/tools/oracle/1_create_user.sql

@@ -0,0 +1,3 @@
+ALTER SESSION SET CONTAINER=XEPDB1;
+CREATE USER ROOT IDENTIFIED BY 123456 QUOTA UNLIMITED ON USERS;
+GRANT CONNECT, RESOURCE TO ROOT;

sql/tools/oracle/2_create_schema.sh

@@ -0,0 +1 @@
+sqlplus -s ROOT/123456@//localhost/XEPDB1 @/tmp/schema.sql

sql/tools/sqlserver/create_schema.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+
+/opt/mssql-tools/bin/sqlcmd -S localhost -U sa -P ${SA_PASSWORD} -Q "CREATE DATABASE [ruoyi-vue-pro];
+GO"
+/opt/mssql-tools/bin/sqlcmd -S localhost -U sa -P ${SA_PASSWORD} -d 'ruoyi-vue-pro' -i /tmp/schema.sql