トークンと AWS CDK - AWS Cloud Development Kit (AWS CDK) v2

これは AWS CDK v2 開発者ガイドです。旧版の CDK v1 は 2022 年 6 月 1 日にメンテナンスを開始し、2023 年 6 月 1 日にサポートを終了しました。

トークンと AWS CDK

AWS Cloud Development Kit (AWS CDK) では、トークンは、コンストラクトの定義やスタックの合成時にわからない値のプレースホルダーです。実際のインフラストラクチャが作成されると、これらの値はデプロイ時に完全に解決されます。AWS CDK アプリケーションを開発するとき、トークンを使用してアプリケーション全体でこれらの値を管理します。

トークンの例

次の内容は、Amazon Simple Storage Service (Amazon S3) バケットのコンストラクトを定義する CDK スタックの例です。バケットの名前はまだ不明なため、bucketName の値はトークンとして保存されます。

TypeScript
import * as cdk from 'aws-cdk-lib'; import { Construct } from 'constructs'; import * as s3 from 'aws-cdk-lib/aws-s3'; export class CdkDemoAppStack extends cdk.Stack { constructor(scope: Construct, id: string, props?: cdk.StackProps) { super(scope, id, props); // Define an S3 bucket const myBucket = new s3.Bucket(this, 'myBucket'); // Store value of the S3 bucket name const myBucketName = myBucket.bucketName; // Print the current value for the S3 bucket name at synthesis console.log("myBucketName: " + bucketName); } }
JavaScript
const { Stack, Duration } = require('aws-cdk-lib'); const s3 = require('aws-cdk-lib/aws-s3'); class CdkDemoAppStack extends Stack { constructor(scope, id, props) { super(scope, id, props); // Define an S3 bucket const myBucket = new s3.Bucket(this, 'myBucket'); // Store value of the S3 bucket name const myBucketName = myBucket.bucketName; // Print the current value for the S3 bucket name at synthesis console.log("myBucketName: " + myBucketName); } } module.exports = { CdkDemoAppStack }
Python
from aws_cdk import ( Stack ) from constructs import Construct from aws_cdk import aws_s3 as s3 class CdkDemoAppStack(Stack): def __init__(self, scope: Construct, construct_id: str, **kwargs) -> None: super().__init__(scope, construct_id, **kwargs) # Define an S3 bucket my_bucket = s3.Bucket(self, "myBucket") # Store the value of the S3 bucket name my_bucket_name = my_bucket.bucket_name # Print the current value for the S3 bucket name at synthesis print(f"myBucketName: {my_bucket_name}")
Java
package com.myorg; import software.constructs.Construct; import software.amazon.awscdk.Stack; import software.amazon.awscdk.StackProps; import software.amazon.awscdk.services.s3.Bucket; import java.util.Map; public class CdkDemoAppStack extends Stack { public CdkDemoAppStack(final Construct scope, final String id) { this(scope, id, null); } public CdkDemoAppStack(final Construct scope, final String id, final StackProps props) { super(scope, id, props); // Define an S3 bucket Bucket myBucket = Bucket.Builder.create(this, "myBucket") .build(); // Store the token for the bucket name String myBucketName = myBucket.getBucketName(); // Print the token at synthesis System.out.println("myBucketName: " + myBucketName); } }
C#
using Amazon.CDK; using Constructs; using Amazon.CDK.AWS.S3; namespace CdkDemoApp { public class CdkDemoAppStack : Stack { internal CdkDemoAppStack(Construct scope, string id, IStackProps props = null) : base(scope, id, props) { // Define an S3 bucket var myBucket = new Bucket(this, "myBucket"); // Store the token for the bucket name var myBucketName = myBucket.BucketName; // Print the token at synthesis System.Console.WriteLine($"myBucketName: {myBucketName}"); } } }
Go
package main import ( "fmt" "github.com/aws/aws-cdk-go/awscdk/v2" "github.com/aws/aws-cdk-go/awscdk/v2/awss3" "github.com/aws/constructs-go/constructs/v10" "github.com/aws/jsii-runtime-go" ) type CdkDemoAppStackProps struct { awscdk.StackProps } func NewCdkDemoAppStack(scope constructs.Construct, id string, props *CdkDemoAppStackProps) awscdk.Stack { var sprops awscdk.StackProps if props != nil { sprops = props.StackProps } stack := awscdk.NewStack(scope, &id, &sprops) // Define an S3 bucket myBucket := awss3.NewBucket(stack, jsii.String("myBucket"), &awss3.BucketProps{}) // Store the token for the bucket name myBucketName := myBucket.BucketName() // Print the token at synthesis fmt.Println("myBucketName: ", *myBucketName) return stack } // ...

スタックを合成するために cdk synth を実行するとき、myBucketName の値は ${Token[TOKEN.1234]} のトークン形式で表示されます。このトークン形式は、AWS CDK がトークンをエンコードする方法の結果です。この例では、トークンは文字列としてエンコードされます。

$ cdk synth --quiet myBucketName: ${Token[TOKEN.21]}

バケット名の値は合成時にわからないため、トークンは myBucket<unique-hash> としてレンダリングされます。AWS CloudFormation テンプレートは Ref 組み込み関数を使用してその値を参照し、デプロイ時に判明されます。

Resources: myBucket5AF9C99B: # ... Outputs: bucketNameOutput: Description: The name of the S3 bucket Value: Ref: myBucket5AF9C99B

一意のハッシュの生成方法の詳細については、「AWS CloudFormation テンプレートで生成された論理 ID 」を参照してください。

トークンを渡す

トークンが表す実際の値であるかのように、渡すことができます。次の内容は、バケット名のトークンを AWS Lambda 関数の コンストラクトに渡す例です。

TypeScript

import * as cdk from 'aws-cdk-lib'; import { Construct } from 'constructs'; import * as s3 from 'aws-cdk-lib/aws-s3'; import * as lambda from 'aws-cdk-lib/aws-lambda'; export class CdkDemoAppStack extends cdk.Stack { constructor(scope: Construct, id: string, props?: cdk.StackProps) { super(scope, id, props); // Define an S3 bucket const myBucket = new s3.Bucket(this, 'myBucket'); // ... // Define a Lambda function const myFunction = new lambda.Function(this, "myFunction", { runtime: lambda.Runtime.NODEJS_20_X, handler: "index.handler", code: lambda.Code.fromInline(` exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; `), functionName: myBucketName + "Function", // Pass token for the S3 bucket name environment: { BUCKET_NAME: myBucketName, // Pass token for the S3 bucket name } }); } }
JavaScript
const { Stack, Duration } = require('aws-cdk-lib'); const s3 = require('aws-cdk-lib/aws-s3'); const lambda = require('aws-cdk-lib/aws-lambda'); class CdkDemoAppStack extends Stack { constructor(scope, id, props) { super(scope, id, props); // Define an S3 bucket const myBucket = new s3.Bucket(this, 'myBucket'); // ... // Define a Lambda function const myFunction = new lambda.Function(this, 'myFunction', { runtime: lambda.Runtime.NODEJS_20_X, handler: 'index.handler', code: lambda.Code.fromInline(` exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; `), functionName: myBucketName + 'Function', // Pass token for the S3 bucket name environment: { BUCKET_NAME: myBucketName, // Pass token for the S3 bucket name } }); } } module.exports = { CdkDemoAppStack }
Python
from aws_cdk import ( Stack ) from constructs import Construct from aws_cdk import aws_s3 as s3 from aws_cdk import aws_lambda as _lambda class CdkDemoAppStack(Stack): def __init__(self, scope: Construct, construct_id: str, **kwargs) -> None: super().__init__(scope, construct_id, **kwargs) # Define an S3 bucket my_bucket = s3.Bucket(self, "myBucket") # ... # Define a Lambda function my_function = _lambda.Function(self, "myFunction", runtime=_lambda.Runtime.NODEJS_20_X, handler="index.handler", code=_lambda.Code.from_inline(""" exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; """), function_name=f"{my_bucket_name}Function", # Pass token for the S3 bucket name environment={ "BUCKET_NAME": my_bucket_name # Pass token for the S3 bucket name } )
Java
package com.myorg; import software.constructs.Construct; import software.amazon.awscdk.Stack; import software.amazon.awscdk.StackProps; import software.amazon.awscdk.services.s3.Bucket; import software.amazon.awscdk.services.lambda.Code; import software.amazon.awscdk.services.lambda.Function; import software.amazon.awscdk.services.lambda.Runtime; import java.util.Map; public class CdkDemoAppStack extends Stack { public CdkDemoAppStack(final Construct scope, final String id) { this(scope, id, null); } public CdkDemoAppStack(final Construct scope, final String id, final StackProps props) { super(scope, id, props); // Define an S3 bucket Bucket myBucket = Bucket.Builder.create(this, "myBucket") .build(); // ... // Define a Lambda function Function myFunction = Function.Builder.create(this, "myFunction") .runtime(Runtime.NODEJS_20_X) .handler("index.handler") .code(Code.fromInline( "exports.handler = async function(event) {" + "return {" + "statusCode: 200," + "body: JSON.stringify('Hello World!')," + "};" + "};" )) .functionName(myBucketName + "Function") // Pass the token for the s3 bucket to the function construct .environment(Map.of("BUCKET_NAME", myBucketName)) // Pass the bucket name as environment variable .build(); } }
C#
using Amazon.CDK; using Constructs; using Amazon.CDK.AWS.S3; using Amazon.CDK.AWS.Lambda; using System; using System.Collections.Generic; namespace CdkDemoApp { public class CdkDemoAppStack : Stack { internal CdkDemoAppStack(Construct scope, string id, IStackProps props = null) : base(scope, id, props) { // Define an S3 bucket var myBucket = new Bucket(this, "myBucket"); // ... // Define a Lambda function var myFunction = new Function(this, "myFunction", new FunctionProps { Runtime = Runtime.NODEJS_20_X, Handler = "index.handler", Code = Code.FromInline(@" exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; "), // Pass the token for the S3 bucket name Environment = new Dictionary<string, string> { { "BUCKET_NAME", myBucketName } }, FunctionName = $"{myBucketName}Function" // Pass the token for the s3 bucket to the function construct }); } } }
Go
package main import ( "fmt" "github.com/aws/aws-cdk-go/awscdk/v2" "github.com/aws/aws-cdk-go/awscdk/v2/awslambda" "github.com/aws/aws-cdk-go/awscdk/v2/awss3" "github.com/aws/constructs-go/constructs/v10" "github.com/aws/jsii-runtime-go" ) type CdkDemoAppStackProps struct { awscdk.StackProps } func NewCdkDemoAppStack(scope constructs.Construct, id string, props *CdkDemoAppStackProps) awscdk.Stack { var sprops awscdk.StackProps if props != nil { sprops = props.StackProps } stack := awscdk.NewStack(scope, &id, &sprops) // Define an S3 bucket myBucket := awss3.NewBucket(stack, jsii.String("myBucket"), &awss3.BucketProps{}) // ... // Define a Lambda function myFunction := awslambda.NewFunction(stack, jsii.String("myFunction"), &awslambda.FunctionProps{ Runtime: awslambda.Runtime_NODEJS_20_X(), Handler: jsii.String("index.handler"), Code: awslambda.Code_FromInline(jsii.String(` exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; `)), FunctionName: jsii.String(fmt.Sprintf("%sFunction", *myBucketName)), // Pass the token for the S3 bucket to the function name Environment: &map[string]*string{ "BUCKET_NAME": myBucketName, }, }) return stack } // ...

テンプレートを合成するときに Ref および Fn::Join の組み込み関数を使用して値を指定し、デプロイ時に判明されます。

Resources: myBucket5AF9C99B: Type: AWS::S3::Bucket # ... myFunction884E1557: Type: AWS::Lambda::Function Properties: # ... Environment: Variables: BUCKET_NAME: Ref: myBucket5AF9C99B FunctionName: Fn::Join: - "" - - Ref: myBucket5AF9C99B - Function # ...

トークンエンコーディングの仕組み

トークンは、IResolvable インターフェイスを実装するオブジェクトであり、1 つの resolve メソッドが含まれています。合成中、AWS CDK はこのメソッドを呼び出して CloudFormation テンプレートのトークンの最終値を生成します。

注記

IResolvable インターフェイスを直接操作することはめったにありません。ほとんどの場合、トークンの文字列エンコードされたバージョンのみが表示されます。

トークンエンコードのタイプ

トークンは合成プロセスに関与し、すべてのタイプにおける任意の値を生成します。通常、他の関数は stringnumber などの基本的なタイプの引数のみを受け入れます。このような場合にトークンを使用するには、cdk.Token クラスの静的メソッドを使用して 3 つのタイプのいずれかにトークンをエンコードできます。

  • Token.asString で文字列エンコーディング (またはトークンオブジェクトで .toString() を呼び出す) を生成します。

  • Token.asList でリストエンコーディングを生成します。

  • Token.asNumber で数値エンコーディングを生成します。

IResolvable の可能性がある任意の値を取得し、指定されたタイプのプリミティブ値にエンコードします。

重要

前のタイプのいずれかがエンコードされたトークンである可能性があるため、内容を解析または読み取ろうとするときは注意してください。例えば、文字列を解析してそこから値を抽出しようとし、文字列がエンコードされたトークンである場合、解析は失敗します。同様に、配列の長さをクエリしたり、数値で数学演算を実行したりした場合は、最初にエンコードされたトークンではないことを確認する必要があります。

アプリでトークンを確認する方法

値に未解決のトークンが含まれているかどうかを確認するには、Token.isUnresolved (Python: is_unresolved) メソッドを呼び出します。次の内容は、Amazon S3 バケット名の値がトークンであるかどうかをチェックする例です。トークンではない場合、バケット名の長さを検証します。

TypeScript

// ... export class CdkDemoAppStack extends cdk.Stack { constructor(scope: Construct, id: string, props?: cdk.StackProps) { super(scope, id, props); // Define an S3 bucket const myBucket = new s3.Bucket(this, 'myBucket'); // ... // Check if bucket name is a token. If not, check if length is less than 10 characters if (cdk.Token.isUnresolved(myBucketName)) { console.log("Token identified."); } else if (!cdk.Token.isUnresolved(myBucketName) && myBucketName.length > 10) { throw new Error('Maximum length for name is 10 characters.'); }; // ...
JavaScript
const { Stack, Duration, Token, CfnOutput } = require('aws-cdk-lib'); // ... class CdkDemoAppStack extends Stack { constructor(scope, id, props) { super(scope, id, props); // Define an S3 bucket const myBucket = new s3.Bucket(this, 'myBucket'); // ... // Check if bucket name is a token. If not, check if length is less than 10 characters if (Token.isUnresolved(myBucketName)) { console.log("Token identified."); } else if (!Token.isUnresolved(myBucketName) && myBucketName.length > 10) { throw new Error('Maximum length for name is 10 characters.'); }; // ...
Python

from aws_cdk import ( Stack, Token ) # ... class CdkDemoAppStack(Stack): def __init__(self, scope: Construct, construct_id: str, **kwargs) -> None: super().__init__(scope, construct_id, **kwargs) # Define an S3 bucket my_bucket = s3.Bucket(self, "myBucket") # ... # Check if bucket name is a token. If not, check if length is less than 10 characters if Token.is_unresolved(my_bucket_name): print("Token identified.") elif not Token.is_unresolved(my_bucket_name) and len(my_bucket_name) < 10: raise ValueError("Maximum length for name is 10 characters.") # ...
Java
// ... import software.amazon.awscdk.Token; // ... public class CdkDemoAppStack extends Stack { public CdkDemoAppStack(final Construct scope, final String id) { this(scope, id, null); } public CdkDemoAppStack(final Construct scope, final String id, final StackProps props) { super(scope, id, props); // Define an S3 bucket Bucket myBucket = Bucket.Builder.create(this, "myBucket") .build(); // ... // Check if the bucket name is a token. If not, check if length is less than 10 characters if (Token.isUnresolved(myBucketName)) { System.out.println("Token identified."); } else if (!Token.isUnresolved(myBucketName) && myBucketName.length() > 10) { throw new IllegalArgumentException("Maximum length for name is 10 characters."); } // ... } }
C#
using Amazon.CDK; using Constructs; using Amazon.CDK.AWS.S3; using Amazon.CDK.AWS.Lambda; using System; using System.Collections.Generic; namespace CdkDemoApp { public class CdkDemoAppStack : Stack { internal CdkDemoAppStack(Construct scope, string id, IStackProps props = null) : base(scope, id, props) { // Define an S3 bucket var myBucket = new Bucket(this, "myBucket"); // ... // Check if bucket name is a token. If not, check if length is less than 10 characters if (Token.IsUnresolved(myBucketName)) { System.Console.WriteLine("Token identified."); } else if (!Token.IsUnresolved(myBucketName) && myBucketName.Length > 10) { throw new System.Exception("Maximum length for name is 10 characters."); } // ...
Go
// ... func NewCdkDemoAppStack(scope constructs.Construct, id string, props *CdkDemoAppStackProps) awscdk.Stack { var sprops awscdk.StackProps if props != nil { sprops = props.StackProps } stack := awscdk.NewStack(scope, &id, &sprops) // Define an S3 bucket myBucket := awss3.NewBucket(stack, jsii.String("myBucket"), &awss3.BucketProps{}) // ... // Check if the bucket name is unresolved (a token) if tokenUnresolved := awscdk.Token_IsUnresolved(myBucketName); tokenUnresolved != nil && *tokenUnresolved { fmt.Println("Token identified.") } else if tokenUnresolved != nil && !*tokenUnresolved && len(*myBucketName) > 10 { panic("Maximum length for name is 10 characters.") } // ...

cdk synth を実行するとき、myBucketName はトークンとして識別されます。

$ cdk synth --quiet Token identified.
注記

トークンエンコーディングを使用し、タイプシステムからエスケープできます。例えば、合成時に数値を生成するトークンを文字列エンコードできます。これらの関数を使用する場合、合成後にテンプレートが使用可能な状態に解決されていることを確認するのがユーザーの責任です。

文字列エンコードされたトークンの使用

文字列エンコードされたトークンは次のようになります。

${TOKEN[Bucket.Name.1234]}

次の例で示すように、通常の文字列のように渡して連結できます。

TypeScript
const functionName = bucket.bucketName + 'Function';
JavaScript
const functionName = bucket.bucketName + 'Function';
Python
function_name = bucket.bucket_name + "Function"
Java
String functionName = bucket.getBucketName().concat("Function");
C#
string functionName = bucket.BucketName + "Function";
Go
functionName := *bucket.BucketName() + "Function"

次の例で示すように、言語が文字列補間をサポートしている場合、文字列補間も使用できます。

TypeScript
const functionName = `${bucket.bucketName}Function`;
JavaScript
const functionName = `${bucket.bucketName}Function`;
Python
function_name = f"{bucket.bucket_name}Function"
Java
String functionName = String.format("%sFunction". bucket.getBucketName());
C#
string functionName = $"${bucket.bucketName}Function";
Go

同様の機能には fmt.Sprintf を使用します。

functionName := fmt.Sprintf("%sFunction", *bucket.BucketName())

他の方法で文字列を操作しないでください。例えば、文字列の部分文字列を取ると、文字列トークンが壊れる可能性があります。

リストエンコードされたトークンの使用

リストエンコードされたトークンは次のようになります。

["#{TOKEN[Stack.NotificationArns.1234]}"]

これらのリストにする唯一の安全なことは、他のコンストラクトに直接渡すことです。文字列リスト形式のトークンは連結することも、トークンから要素を取得することもできません。安全に操作する唯一の方法は、Fn.select などの AWS CloudFormation 組み込み関数を使用することです。

番号エンコードされたトークンの使用

数値エンコードされたトークンは、次のような小さな負の浮動小数点数のセットです。

-1.8881545897087626e+289

リストトークンと同様に、数値トークンが壊れる可能性があるため、数値は変更できません。

次の内容は、数値としてエンコードされたトークンを含むコンストラクトの例です。

TypeScript
import { Stack, Duration, StackProps } from 'aws-cdk-lib'; import { Construct } from 'constructs'; import * as rds from 'aws-cdk-lib/aws-rds'; import * as ec2 from 'aws-cdk-lib/aws-ec2'; export class CdkDemoAppStack extends Stack { constructor(scope: Construct, id: string, props?: StackProps) { super(scope, id, props); // Define a new VPC const vpc = new ec2.Vpc(this, 'MyVpc', { maxAzs: 3, // Maximum number of availability zones to use }); // Define an RDS database cluster const dbCluster = new rds.DatabaseCluster(this, 'MyRDSCluster', { engine: rds.DatabaseClusterEngine.AURORA, instanceProps: { vpc, }, }); // Get the port token (this is a token encoded as a number) const portToken = dbCluster.clusterEndpoint.port; // Print the value for our token at synthesis console.log("portToken: " + portToken); } }
JavaScript
const { Stack, Duration } = require('aws-cdk-lib'); const lambda = require('aws-cdk-lib/aws-lambda'); const rds = require('aws-cdk-lib/aws-rds'); const ec2 = require('aws-cdk-lib/aws-ec2'); class CdkDemoAppStack extends Stack { constructor(scope, id, props) { super(scope, id, props); // Define a new VPC const vpc = new ec2.Vpc(this, 'MyVpc', { maxAzs: 3, // Maximum number of availability zones to use }); // Define an RDS database cluster const dbCluster = new rds.DatabaseCluster(this, 'MyRDSCluster', { engine: rds.DatabaseClusterEngine.AURORA, instanceProps: { vpc, }, }); // Get the port token (this is a token encoded as a number) const portToken = dbCluster.clusterEndpoint.port; // Print the value for our token at synthesis console.log("portToken: " + portToken); } } module.exports = { CdkDemoAppStack }
Python
from aws_cdk import ( Duration, Stack, ) from aws_cdk import aws_rds as rds from aws_cdk import aws_ec2 as ec2 from constructs import Construct class CdkDemoAppStack(Stack): def __init__(self, scope: Construct, construct_id: str, **kwargs) -> None: super().__init__(scope, construct_id, **kwargs) # Define a new VPC vpc = ec2.Vpc(self, 'MyVpc', max_azs=3 # Maximum number of availability zones to use ) # Define an RDS database cluster db_cluster = rds.DatabaseCluster(self, 'MyRDSCluster', engine=rds.DatabaseClusterEngine.AURORA, instance_props=rds.InstanceProps( vpc=vpc ) ) # Get the port token (this is a token encoded as a number) port_token = db_cluster.cluster_endpoint.port # Print the value for our token at synthesis print(f"portToken: {port_token}")
Java
package com.myorg; import software.constructs.Construct; import software.amazon.awscdk.Stack; import software.amazon.awscdk.StackProps; import software.amazon.awscdk.services.ec2.Vpc; import software.amazon.awscdk.services.rds.DatabaseCluster; import software.amazon.awscdk.services.rds.DatabaseClusterEngine; import software.amazon.awscdk.services.rds.InstanceProps; public class CdkDemoAppStack extends Stack { public CdkDemoAppStack(final Construct scope, final String id) { this(scope, id, null); } public CdkDemoAppStack(final Construct scope, final String id, final StackProps props) { super(scope, id, props); // Define a new VPC Vpc vpc = Vpc.Builder.create(this, "MyVpc") .maxAzs(3) // Maximum number of availability zones to use .build(); // Define an RDS database cluster DatabaseCluster dbCluster = DatabaseCluster.Builder.create(this, "MyRDSCluster") .engine(DatabaseClusterEngine.AURORA) .instanceProps(InstanceProps.builder() .vpc(vpc) .build()) .build(); // Get the port token (this is a token encoded as a number) Number portToken = dbCluster.getClusterEndpoint().getPort(); // Print the value for our token at synthesis System.out.println("portToken: " + portToken); } }
C#
using Amazon.CDK; using Constructs; using Amazon.CDK.AWS.EC2; using Amazon.CDK.AWS.RDS; using System; using System.Collections.Generic; namespace CdkDemoApp { public class CdkDemoAppStack : Stack { internal CdkDemoAppStack(Construct scope, string id, IStackProps props = null) : base(scope, id, props) { // Define a new VPC var vpc = new Vpc(this, "MyVpc", new VpcProps { MaxAzs = 3 // Maximum number of availability zones to use }); // Define an RDS database cluster var dbCluster = new DatabaseCluster(this, "MyRDSCluster", new DatabaseClusterProps { Engine = DatabaseClusterEngine.AURORA, // Remove parentheses InstanceProps = new Amazon.CDK.AWS.RDS.InstanceProps // Specify RDS InstanceProps { Vpc = vpc } }); // Get the port token (this is a token encoded as a number) var portToken = dbCluster.ClusterEndpoint.Port; // Print the value for our token at synthesis System.Console.WriteLine($"portToken: {portToken}"); } } }
Go
package main import ( "fmt" "github.com/aws/aws-cdk-go/awscdk/v2" "github.com/aws/aws-cdk-go/awscdk/v2/awsec2" "github.com/aws/aws-cdk-go/awscdk/v2/awsrds" "github.com/aws/constructs-go/constructs/v10" "github.com/aws/jsii-runtime-go" ) type CdkDemoAppStackProps struct { awscdk.StackProps } func NewCdkDemoAppStack(scope constructs.Construct, id string, props *CdkDemoAppStackProps) awscdk.Stack { var sprops awscdk.StackProps if props != nil { sprops = props.StackProps } stack := awscdk.NewStack(scope, &id, &sprops) // Define a new VPC vpc := awsec2.NewVpc(stack, jsii.String("MyVpc"), &awsec2.VpcProps{ MaxAzs: jsii.Number(3), // Maximum number of availability zones to use }) // Define an RDS database cluster dbCluster := awsrds.NewDatabaseCluster(stack, jsii.String("MyRDSCluster"), &awsrds.DatabaseClusterProps{ Engine: awsrds.DatabaseClusterEngine_AURORA(), InstanceProps: &awsrds.InstanceProps{ Vpc: vpc, }, }) // Get the port token (this is a token encoded as a number) portToken := dbCluster.ClusterEndpoint().Port() // Print the value for our token at synthesis fmt.Println("portToken: ", portToken) return stack } // ...

cdk synth を実行するとき、portToken の値が数値エンコードされたトークンとして表示されます。

$ cdk synth --quiet portToken: -1.8881545897087968e+289

数値エンコードされたトークンを渡す

数値エンコードされたトークンを他のコンストラクトに渡すとき、最初に文字列に変換するのが理にかなっている場合があります。例えば、数値エンコードされた文字列の値を連結文字列の一部として使用する場合、変換すると読みやすくなります。

次の例では、portTokenconnectionString の一部として Lambda 関数に渡す数値エンコードされたトークンです。

TypeScript
import { Stack, Duration, CfnOutput, StackProps } from 'aws-cdk-lib'; // ... import * as lambda from 'aws-cdk-lib/aws-lambda'; export class CdkDemoAppStack extends Stack { constructor(scope: Construct, id: string, props?: StackProps) { super(scope, id, props); // Define a new VPC // ... // Define an RDS database cluster // ... // Get the port token (this is a token encoded as a number) const portToken = dbCluster.clusterEndpoint.port; // ... // Example connection string with the port token as a number const connectionString = `jdbc:mysql://mydb.cluster.amazonaws.com:${portToken}/mydatabase`; // Use the connection string as an environment variable in a Lambda function const myFunction = new lambda.Function(this, 'MyLambdaFunction', { runtime: lambda.Runtime.NODEJS_20_X, handler: 'index.handler', code: lambda.Code.fromInline(` exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; `), environment: { DATABASE_CONNECTION_STRING: connectionString, // Using the port token as part of the string }, }); // Output the value of our connection string at synthesis console.log("connectionString: " + connectionString); // Output the connection string new CfnOutput(this, 'ConnectionString', { value: connectionString, }); } }
JavaScript
const { Stack, Duration, CfnOutput } = require('aws-cdk-lib'); // ... const lambda = require('aws-cdk-lib/aws-lambda'); class CdkDemoAppStack extends Stack { constructor(scope, id, props) { super(scope, id, props); // Define a new VPC // ... // Define an RDS database cluster // ... // Get the port token (this is a token encoded as a number) const portToken = dbCluster.clusterEndpoint.port; // ... // Example connection string with the port token as a number const connectionString = `jdbc:mysql://mydb.cluster.amazonaws.com:${portToken}/mydatabase`; // Use the connection string as an environment variable in a Lambda function const myFunction = new lambda.Function(this, 'MyLambdaFunction', { runtime: lambda.Runtime.NODEJS_20_X, handler: 'index.handler', code: lambda.Code.fromInline(` exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; `), environment: { DATABASE_CONNECTION_STRING: connectionString, // Using the port token as part of the string }, }); // Output the value of our connection string at synthesis console.log("connectionString: " + connectionString); // Output the connection string new CfnOutput(this, 'ConnectionString', { value: connectionString, }); } } module.exports = { CdkDemoAppStack }
Python
from aws_cdk import ( Duration, Stack, CfnOutput, ) from aws_cdk import aws_lambda as _lambda # ... class CdkDemoAppStack(Stack): def __init__(self, scope: Construct, construct_id: str, **kwargs) -> None: super().__init__(scope, construct_id, **kwargs) # Define a new VPC # ... # Define an RDS database cluster # ... # Get the port token (this is a token encoded as a number) port_token = db_cluster.cluster_endpoint.port # ... # Example connection string with the port token as a number connection_string = f"jdbc:mysql://mydb.cluster.amazonaws.com:{port_token}/mydatabase" # Use the connection string as an environment variable in a Lambda function my_function = _lambda.Function(self, 'MyLambdaFunction', runtime=_lambda.Runtime.NODEJS_20_X, handler='index.handler', code=_lambda.Code.from_inline(""" exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; """), environment={ 'DATABASE_CONNECTION_STRING': connection_string # Using the port token as part of the string } ) # Output the value of our connection string at synthesis print(f"connectionString: {connection_string}") # Output the connection string CfnOutput(self, 'ConnectionString', value=connection_string )
Java

// ... import software.amazon.awscdk.CfnOutput; import software.amazon.awscdk.services.lambda.Function; import software.amazon.awscdk.services.lambda.Runtime; import software.amazon.awscdk.services.lambda.Code; import java.util.Map; public class CdkDemoAppStack extends Stack { public CdkDemoAppStack(final Construct scope, final String id) { this(scope, id, null); } public CdkDemoAppStack(final Construct scope, final String id, final StackProps props) { super(scope, id, props); // Define a new VPC // ... // Define an RDS database cluster // ... // Get the port token (this is a token encoded as a number) Number portToken = dbCluster.getClusterEndpoint().getPort(); // ... // Example connection string with the port token as a number String connectionString = "jdbc:mysql://mydb.cluster.amazonaws.com:" + portToken + "/mydatabase"; // Use the connection string as an environment variable in a Lambda function Function myFunction = Function.Builder.create(this, "MyLambdaFunction") .runtime(Runtime.NODEJS_20_X) .handler("index.handler") .code(Code.fromInline( "exports.handler = async function(event) {\n" + " return {\n" + " statusCode: 200,\n" + " body: JSON.stringify('Hello World!'),\n" + " };\n" + "};")) .environment(Map.of( "DATABASE_CONNECTION_STRING", connectionString // Using the port token as part of the string )) .build(); // Output the value of our connection string at synthesis System.out.println("connectionString: " + connectionString); // Output the connection string CfnOutput.Builder.create(this, "ConnectionString") .value(connectionString) .build(); } }
C#
// ... using Amazon.CDK.AWS.Lambda; namespace CdkDemoApp { public class CdkDemoAppStack : Stack { internal CdkDemoAppStack(Construct scope, string id, IStackProps props = null) : base(scope, id, props) { // Define a new VPC // ... // Define an RDS database cluster var dbCluster = new DatabaseCluster(this, "MyRDSCluster", new DatabaseClusterProps // ... // Get the port token (this is a token encoded as a number) var portToken = dbCluster.ClusterEndpoint.Port; // ... // Example connection string with the port token as a number var connectionString = $"jdbc:mysql://mydb.cluster.amazonaws.com:{portToken}/mydatabase"; // Use the connection string as an environment variable in a Lambda function var myFunction = new Function(this, "MyLambdaFunction", new FunctionProps { Runtime = Runtime.NODEJS_20_X, Handler = "index.handler", Code = Code.FromInline(@" exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; "), Environment = new Dictionary<string, string> { { "DATABASE_CONNECTION_STRING", connectionString } // Using the port token as part of the string } }); // Output the value of our connection string at synthesis Console.WriteLine($"connectionString: {connectionString}"); // Output the connection string new CfnOutput(this, "ConnectionString", new CfnOutputProps { Value = connectionString }); } } }
Go
// ... "github.com/aws/aws-cdk-go/awscdk/v2/awslambda" ) type CdkDemoAppStackProps struct { awscdk.StackProps } func NewCdkDemoAppStack(scope constructs.Construct, id string, props *CdkDemoAppStackProps) awscdk.Stack { var sprops awscdk.StackProps if props != nil { sprops = props.StackProps } stack := awscdk.NewStack(scope, &id, &sprops) // Define a new VPC // ... // Define an RDS database cluster // ... // Get the port token (this is a token encoded as a number) portToken := dbCluster.ClusterEndpoint().Port() // ... // Example connection string with the port token as a number connectionString := fmt.Sprintf("jdbc:mysql://mydb.cluster.amazonaws.com:%s/mydatabase", portToken) // Use the connection string as an environment variable in a Lambda function myFunction := awslambda.NewFunction(stack, jsii.String("MyLambdaFunction"), &awslambda.FunctionProps{ Runtime: awslambda.Runtime_NODEJS_20_X(), Handler: jsii.String("index.handler"), Code: awslambda.Code_FromInline(jsii.String(` exports.handler = async function(event) { return { statusCode: 200, body: JSON.stringify('Hello World!'), }; }; `)), Environment: &map[string]*string{ "DATABASE_CONNECTION_STRING": jsii.String(connectionString), // Using the port token as part of the string }, }) // Output the value of our connection string at synthesis fmt.Println("connectionString: ", connectionString) // Output the connection string awscdk.NewCfnOutput(stack, jsii.String("ConnectionString"), &awscdk.CfnOutputProps{ Value: jsii.String(connectionString), }) return stack } // ...

この値を connectionString に渡す場合、数値エンコードされた文字列により、cdk synth の実行時の出力値が混乱を招く場合があります。

$ cdk synth --quiet connectionString: jdbc:mysql://mydb.cluster.amazonaws.com:-1.888154589708796e+289/mydatabase

数値エンコードされたトークンを文字列に変換するには、cdk.Tokenization.stringifyNumber(token) を使用します。次の例では、接続文字列を定義する前に、数値エンコードされたトークンを文字列に変換します。

TypeScript
import { Stack, Duration, Tokenization, CfnOutput, StackProps } from 'aws-cdk-lib'; // ... export class CdkDemoAppStack extends Stack { constructor(scope: Construct, id: string, props?: StackProps) { super(scope, id, props); // Define a new VPC // ... // Define an RDS database cluster // ... // Get the port token (this is a token encoded as a number) const portToken = dbCluster.clusterEndpoint.port; // ... // Convert the encoded number to an encoded string for use in the connection string const portAsString = Tokenization.stringifyNumber(portToken); // Example connection string with the port token as a string const connectionString = `jdbc:mysql://mydb.cluster.amazonaws.com:${portAsString}/mydatabase`; // Use the connection string as an environment variable in a Lambda function const myFunction = new lambda.Function(this, 'MyLambdaFunction', { // ... environment: { DATABASE_CONNECTION_STRING: connectionString, // Using the port token as part of the string }, }); // Output the value of our connection string at synthesis console.log("connectionString: " + connectionString); // Output the connection string new CfnOutput(this, 'ConnectionString', { value: connectionString, }); } }
JavaScript
const { Stack, Duration, Tokenization, CfnOutput } = require('aws-cdk-lib'); // ... class CdkDemoAppStack extends Stack { constructor(scope, id, props) { super(scope, id, props); // Define a new VPC // ... // Define an RDS database cluster // ... // Get the port token (this is a token encoded as a number) const portToken = dbCluster.clusterEndpoint.port; // ... // Convert the encoded number to an encoded string for use in the connection string const portAsString = Tokenization.stringifyNumber(portToken); // Example connection string with the port token as a string const connectionString = `jdbc:mysql://mydb.cluster.amazonaws.com:${portAsString}/mydatabase`; // Use the connection string as an environment variable in a Lambda function const myFunction = new lambda.Function(this, 'MyLambdaFunction', { // ... environment: { DATABASE_CONNECTION_STRING: connectionString, // Using the port token as part of the string }, }); // Output the value of our connection string at synthesis console.log("connectionString: " + connectionString); // Output the connection string new CfnOutput(this, 'ConnectionString', { value: connectionString, }); } } module.exports = { CdkDemoAppStack }
Python
from aws_cdk import ( Duration, Stack, Tokenization, CfnOutput, ) # ... class CdkDemoAppStack(Stack): def __init__(self, scope: Construct, construct_id: str, **kwargs) -> None: super().__init__(scope, construct_id, **kwargs) # Define a new VPC # ... # Define an RDS database cluster # ... # Get the port token (this is a token encoded as a number) port_token = db_cluster.cluster_endpoint.port # Convert the encoded number to an encoded string for use in the connection string port_as_string = Tokenization.stringify_number(port_token) # Example connection string with the port token as a string connection_string = f"jdbc:mysql://mydb.cluster.amazonaws.com:{port_as_string}/mydatabase" # Use the connection string as an environment variable in a Lambda function my_function = _lambda.Function(self, 'MyLambdaFunction', # ... environment={ 'DATABASE_CONNECTION_STRING': connection_string # Using the port token as part of the string } ) # Output the value of our connection string at synthesis print(f"connectionString: {connection_string}") # Output the connection string CfnOutput(self, 'ConnectionString', value=connection_string )
Java
// ... import software.amazon.awscdk.Tokenization; public class CdkDemoAppStack extends Stack { public CdkDemoAppStack(final Construct scope, final String id) { this(scope, id, null); } public CdkDemoAppStack(final Construct scope, final String id, final StackProps props) { super(scope, id, props); // Define a new VPC // ... // Define an RDS database cluster // ... // Get the port token (this is a token encoded as a number) Number portToken = dbCluster.getClusterEndpoint().getPort(); // ... // Convert the encoded number to an encoded string for use in the connection string String portAsString = Tokenization.stringifyNumber(portToken); // Example connection string with the port token as a string String connectionString = "jdbc:mysql://mydb.cluster.amazonaws.com:" + portAsString + "/mydatabase"; // Use the connection string as an environment variable in a Lambda function Function myFunction = Function.Builder.create(this, "MyLambdaFunction") // ... .environment(Map.of( "DATABASE_CONNECTION_STRING", connectionString // Using the port token as part of the string )) .build(); // Output the value of our connection string at synthesis System.out.println("connectionString: " + connectionString); // Output the connection string CfnOutput.Builder.create(this, "ConnectionString") .value(connectionString) .build(); } }
C#
// ... namespace CdkDemoApp { public class CdkDemoAppStack : Stack { internal CdkDemoAppStack(Construct scope, string id, IStackProps props = null) : base(scope, id, props) { // Define a new VPC // ... // Define an RDS database cluster // ... // Get the port token (this is a token encoded as a number) var portToken = dbCluster.ClusterEndpoint.Port; // ... // Convert the encoded number to an encoded string for use in the connection string var portAsString = Tokenization.StringifyNumber(portToken); // Example connection string with the port token as a string var connectionString = $"jdbc:mysql://mydb.cluster.amazonaws.com:{portAsString}/mydatabase"; // Use the connection string as an environment variable in a Lambda function var myFunction = new Function(this, "MyLambdaFunction", new FunctionProps { // ... Environment = new Dictionary<string, string> { { "DATABASE_CONNECTION_STRING", connectionString } // Using the port token as part of the string } }); // Output the value of our connection string at synthesis Console.WriteLine($"connectionString: {connectionString}"); // Output the connection string new CfnOutput(this, "ConnectionString", new CfnOutputProps { Value = connectionString }); } } }
Go
// ... func NewCdkDemoAppStack(scope constructs.Construct, id string, props *CdkDemoAppStackProps) awscdk.Stack { var sprops awscdk.StackProps if props != nil { sprops = props.StackProps } stack := awscdk.NewStack(scope, &id, &sprops) // Define a new VPC // ... // Define an RDS database cluster // ... // Get the port token (this is a token encoded as a number) portToken := dbCluster.ClusterEndpoint().Port() // ... // Convert the encoded number to an encoded string for use in the connection string portAsString := awscdk.Tokenization_StringifyNumber(portToken) // Example connection string with the port token as a string connectionString := fmt.Sprintf("jdbc:mysql://mydb.cluster.amazonaws.com:%s/mydatabase", portAsString) // Use the connection string as an environment variable in a Lambda function myFunction := awslambda.NewFunction(stack, jsii.String("MyLambdaFunction"), &awslambda.FunctionProps{ // ... Environment: &map[string]*string{ "DATABASE_CONNECTION_STRING": jsii.String(connectionString), // Using the port token as part of the string }, }) // Output the value of our connection string at synthesis fmt.Println("connectionString: ", connectionString) // Output the connection string awscdk.NewCfnOutput(stack, jsii.String("ConnectionString"), &awscdk.CfnOutputProps{ Value: jsii.String(connectionString), }) fmt.Println(myFunction) return stack } // ...

cdk synth を実行するとき、接続文字列の値はより簡潔で明確な形式で表されます。

$ cdk synth --quiet connectionString: jdbc:mysql://mydb.cluster.amazonaws.com:${Token[TOKEN.242]}/mydatabase

遅延値

AWS CloudFormation パラメータなどのデプロイ時間値を表すだけではなく、トークンは合成時間の遅延値を表すためにも一般的に使用されます。合成が完了する前に最終値が決定される値ですが、値が構築された時点では決定されません。トークンを使用してリテラル文字列または数値を別のコンストラクトに渡しますが、合成時の実際の値はまだ発生していない計算に依存する場合があります。

Lazy.stringLazy.number などの Lazy クラスの静的メソッドを使用して、合成時間の遅延値を表すトークンを構成できます。これらのメソッドは、produce プロパティがコンテキスト引数を受け入れる関数であるオブジェクトを受け入れ、呼び出されたときに最終値を返します。

次の例では Auto Scaling グループを作成し、作成後に容量が決定されます。

TypeScript
let actualValue: number; new AutoScalingGroup(this, 'Group', { desiredCapacity: Lazy.numberValue({ produce(context) { return actualValue; } }) }); // At some later point actualValue = 10;
JavaScript
let actualValue; new AutoScalingGroup(this, 'Group', { desiredCapacity: Lazy.numberValue({ produce(context) { return (actualValue); } }) }); // At some later point actualValue = 10;
Python
class Producer: def __init__(self, func): self.produce = func actual_value = None AutoScalingGroup(self, "Group", desired_capacity=Lazy.number_value(Producer(lambda context: actual_value)) ) # At some later point actual_value = 10
Java
double actualValue = 0; class ProduceActualValue implements INumberProducer { @Override public Number produce(IResolveContext context) { return actualValue; } } AutoScalingGroup.Builder.create(this, "Group") .desiredCapacity(Lazy.numberValue(new ProduceActualValue())).build(); // At some later point actualValue = 10;
C#
public class NumberProducer : INumberProducer { Func<Double> function; public NumberProducer(Func<Double> function) { this.function = function; } public Double Produce(IResolveContext context) { return function(); } } double actualValue = 0; new AutoScalingGroup(this, "Group", new AutoScalingGroupProps { DesiredCapacity = Lazy.NumberValue(new NumberProducer(() => actualValue)) }); // At some later point actualValue = 10;

JSON への変換

任意のデータの JSON 文字列を生成し、データにトークンが含まれているかどうかわからない場合があります。次の例で示すように、トークンが含まれているかどうかを問わず、データ構造を適切に JSON エンコードするには、stack.toJsonString メソッドを使用します。

TypeScript
const stack = Stack.of(this); const str = stack.toJsonString({ value: bucket.bucketName });
JavaScript
const stack = Stack.of(this); const str = stack.toJsonString({ value: bucket.bucketName });
Python
stack = Stack.of(self) string = stack.to_json_string(dict(value=bucket.bucket_name))
Java
Stack stack = Stack.of(this); String stringVal = stack.toJsonString(java.util.Map.of( // Map.of requires Java 9+ put("value", bucket.getBucketName())));
C#
var stack = Stack.Of(this); var stringVal = stack.ToJsonString(new Dictionary<string, string> { ["value"] = bucket.BucketName });