# huaweicloud-RFS-ApplyExecutionPlanDemo-python

**Repository Path**: HuaweiCloudDeveloper/huaweicloud-rfs-apply-execution-plan-demo-python

## Basic Information

- **Project Name**: huaweicloud-RFS-ApplyExecutionPlanDemo-python
- **Description**: 提供通过执行计划创建和删除VPC资源的代码示例,说明如何使用执行计划部署资源,并在部署资源之前获得执行计划内容的详细信息。
- **Primary Language**: Unknown
- **License**: Apache-2.0
- **Default Branch**: master-dev
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2023-07-25
- **Last Updated**: 2025-06-16

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

## 1. 示例简介

华为云提供了AOS服务端SDK，您可以直接集成服务端SDK来调用资源编排的相关API，从而实现对资源的快速操作。 

该示例展示了如何通过Python语言版本AOS SDK使用资源栈快速创建您的资源。

## 2. 开发前准备

- 1.参考华为云开发工具包（SDK）引入AOS Python SDK
- 2.要使用华为云Python SDK，您需要拥有云账号以及该账号对应的Access Key（AK）和 Secret Access
  Key（SK）。请在华为云控制台“我的凭证-访问密钥”页面上创建和查看您的AK&SK。更多信息请查看[访问密钥](https://support.huaweicloud.com/usermanual-ca/zh-cn_topic_0046606340.html)。
    - Aos SDK期望用户主动传入project_id，若未传入project_id，Aos SDK将选择您对应region下默认项目的project_id。更多信息请查看[IAM基本概念](https://support.huaweicloud.com/productdesc-iam/iam_01_0023.html#section8)。
- 3.获取您期望使用的目标终端节点
- 4.预先准备您的模板文件（HCL语法文本描述文件，支持tf、tf.json、zip包文件格式，用于描述您的云资源）
    - 当前支持的provider版本请见[华为云官网](https://support.huaweicloud.com/productdesc-aos/rf_02_0006.html)
    - 模板书写说明详见[huaweicloud provider官网](https://registry.terraform.io/providers/huaweicloud/huaweicloud/latest)
- 5.华为云AOS Python SDK 需要运行在Python 3.3及以上版本。

## 3. 安装SDK

使用 pip 安装 SDK 依赖包。

```
# 安装aos服务包
pip install huaweicloudsdkaos
# 安装sdk核心包
pip install huaweicloudsdkcore
```

## 4. 代码示例

声明您的Access Key、Secret Key、VPC资源的HCL语法模板、资源栈名称以及执行计划名称。

```python
    # 认证用的ak和sk直接写到代码中有很大的安全风险，建议在配置文件或者环境变量中密文存放，使用时解密，确保安全。
    # 本示例以ak和sk保存在环境变量中来实现身份验证为例，运行本示例前请先在本地环境中设置环境变量HUAWEICLOUD_SDK_AK和HUAWEICLOUD_SDK_SK。
    ak = os.environ["HUAWEICLOUD_SDK_AK"]
    sk = os.environ["HUAWEICLOUD_SDK_SK"]
    # stack以project_id为单位进行管理
    project_id = "<YOUR PROJECT ID>"
    # HCL语法文本描述文件，支持tf、tf.json文件格式
    template_body = "<YOUR TEMPLATE>"
    # stack_name是唯一的，用于标识您的资源栈
    stack_name = "<YOUR STACK NAME>"
```

通过Aos SDK创建一个Client，资源编排服务为region级别服务，将在您选择的region下生成目标资源。

```python
    auth = BasicCredentials(
        project_id=project_id,
        ak=ak,
        sk=sk
    )
    
    # 资源栈及其资源将创建在此region下，此处以CN_NORTH_4为例
    client = AosClient.new_builder() \
        .with_credentials(credentials=auth) \
        .with_region(region=AosRegion.CN_NORTH_4) \
        .with_http_config(config) \
        .build()
```

首先创建一个空的资源栈，在该步骤中请不要传入templateBody或templateUri，否则将立即触发部署资源栈。

```python
    @staticmethod
    def create_stack(client, stack_name):
        create_stack_request = CreateStackRequest()
        create_stack_request.client_request_id = str(uuid.uuid4())
        create_stack_request_body = CreateStackRequestBody()
        create_stack_request_body.stack_name = stack_name
        create_stack_request.body = create_stack_request_body
        response = client.create_stack(create_stack_request)
        print("create stack return:" + str(response))
        return response.stack_id
```

在当前资源栈下调用CreateExecutionPlan，资源编排将为您异步处理解析模板，生成执行计划，此时不会直接部署资源。

```python
    @staticmethod
    def create_execution_plan(client, stack_name, stack_id, execution_plan_name, template_body):
        create_execution_plan_request = CreateExecutionPlanRequest()
        create_execution_plan_request.client_request_id = str(uuid.uuid4())
        create_execution_plan_request.stack_name = stack_name
        create_execution_plan_request_body = CreateExecutionPlanRequestBody()
        create_execution_plan_request_body.execution_plan_name = execution_plan_name
        create_execution_plan_request_body.stack_id = stack_id
        create_execution_plan_request_body.template_body = template_body
        create_execution_plan_request.body = create_execution_plan_request_body
        response = client.create_execution_plan(create_execution_plan_request)
        print("create execution plan return: " + str(response))
        return response.execution_plan_id
```

您可通过GetExecutionPlan获取执行计划的详细信息。

```python
    @staticmethod
    def get_execution_plan(client, stack_name, execution_plan_name, stack_id, execution_plan_id):
        get_execution_plan_request = GetExecutionPlanRequest()
        get_execution_plan_request.client_request_id = str(uuid.uuid4())
        get_execution_plan_request.execution_plan_name = execution_plan_name
        get_execution_plan_request.execution_plan_id = execution_plan_id
        get_execution_plan_request.stack_name = stack_name
        get_execution_plan_request.stack_id = stack_id
        start_time = datetime.now().timestamp()
        while True:
            now_time = datetime.now().timestamp()
            if now_time - start_time > Example.__timeout_interval:
                print("create execution plan time out and will be exit")
                raise UnExpectedException("create execution plan time out")
            print("waiting for create execution plan ...")
            try:
                time.sleep(5)
            except Exception as e:
                print("waiting for create execution plan throw exception: " + str(e))
                raise e
            response = client.get_execution_plan(get_execution_plan_request)
            if response.status_code != 404:
                print("get execution plan return: " + str(response))
                break
            print("create execution plan in progress ...")
```

确认执行计划与目标资源状态一致后，可继续触发ApplyExecutionPlan，此时才会正式生成资源。

```python
    @staticmethod
    def apply_execution_plan(client, stack_name, execution_plan_name, stack_id, execution_plan_id):
        apply_execution_plan_request = ApplyExecutionPlanRequest()
        apply_execution_plan_request.client_request_id = str(uuid.uuid4())
        apply_execution_plan_request.execution_plan_name = execution_plan_name
        apply_execution_plan_request.stack_name = stack_name
        apply_execution_plan_request_body = ApplyExecutionPlanRequestBody()
        apply_execution_plan_request_body.execution_plan_id = execution_plan_id
        apply_execution_plan_request_body.stack_id = stack_id
        apply_execution_plan_request.body = apply_execution_plan_request_body
        response = client.apply_execution_plan(apply_execution_plan_request)
        print("apply execution plan return : " + str(response))

```

通过getExecutionPlanMetadata轮询获取执行计划的状态，等待部署完成。

```python
    @staticmethod
    def get_execution_plan_metadata(client, stack_name, execution_plan_name, stack_id, execution_plan_id):
        get_execution_plan_metadata_request = GetExecutionPlanMetadataRequest()
        get_execution_plan_metadata_request.client_request_id = str(uuid.uuid4())
        get_execution_plan_metadata_request.execution_plan_name = execution_plan_name
        get_execution_plan_metadata_request.stack_name = stack_name
        get_execution_plan_metadata_request.execution_plan_id = execution_plan_id
        get_execution_plan_metadata_request.stack_id = stack_id
        start_time = datetime.now().timestamp()
        while True:
            now_time = datetime.now().timestamp()
            if now_time - start_time > Example.__timeout_interval:
                print("apply execution plan time out and will be exit")
                raise UnExpectedException("apply execution plan time out")
            print("waiting for apply execution plan ...")
            try:
                time.sleep(5)
            except Exception as e:
                print("waiting for apply execution plan throw exception: " + str(e))
                raise e
            response = client.get_execution_plan_metadata(get_execution_plan_metadata_request)
            if response.status == "APPLIED":
                print("execution plan applied!")
                print("get execution plan metadata return: " + str(response))
                break
            print("apply execution plan in progress ...")
```

此时可通过ListStackResource来查看生成的资源。

```python
    @staticmethod
    def list_stack_resource(client, stack_id, stack_name):
        list_stack_resources_request = ListStackResourcesRequest()
        list_stack_resources_request.client_request_id = str(uuid.uuid4())
        list_stack_resources_request.stack_id = stack_id
        list_stack_resources_request.stack_name = stack_name
        response = client.list_stack_resources(list_stack_resources_request)
        print("list stack resources return: " + str(response))
```

最后，可以通过删除资源栈来删除资源栈、执行计划以及生成的资源。

```python
    @staticmethod
    def delete_stack(client, stack_id, stack_name):
        delete_stack_request = DeleteStackRequest()
        delete_stack_request.client_request_id = str(uuid.uuid4())
        delete_stack_request.stack_id = stack_id
        delete_stack_request.stack_name = stack_name
        client.delete_stack(delete_stack_request)
        get_stack_metadata_request = GetStackMetadataRequest()
        get_stack_metadata_request.client_request_id = str(uuid.uuid4())
        get_stack_metadata_request.stack_id = stack_id
        get_stack_metadata_request.stack_name = stack_name
        start_time = datetime.now().timestamp()
        while True:
            now_time = datetime.now().timestamp()
            if now_time - start_time > Example.__timeout_interval:
                print("delete stack time out and will be exit")
                raise UnExpectedException("delete stack time out")
            print("waiting for delete stack ...")
            try:
                time.sleep(5)
                client.get_stack_metadata(get_stack_metadata_request)
            except ClientRequestException as e:
                if e.status_code == 404:
                    print("delete complete!")
                    return
                print("delete stack throw exception: " + str(e))
            except Exception as e:
                print("waiting delete stack throw exception: " + str(e))
                raise e
```

## 5. 参考

更多信息请参考：[API Explorer](https://console.huaweicloud.com/apiexplorer/#/openapi/AOS/doc?api=CreateStack)

## 6. 修订记录

|    发布日期    | 文档版本 | 修订说明 |
|:----------:| :------: | :----------: |
| 2023-09-27 | 1.0 | 文档首次发布 |