Set up HTTP integrations in API Gateway - Amazon API Gateway
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Set up HTTP integrations in API Gateway

You can integrate an API method with an HTTP endpoint using the HTTP proxy integration or the HTTP custom integration.

API Gateway supports the following endpoint ports: 80, 443 and 1024-65535.

With proxy integration, setup is simple. You only need to set the HTTP method and the HTTP endpoint URI, according to the backend requirements, if you are not concerned with content encoding or caching.

With custom integration, setup is more involved. In addition to the proxy integration setup steps, you need to specify how the incoming request data is mapped to the integration request and how the resulting integration response data is mapped to the method response.

Set up HTTP proxy integrations in API Gateway

To set up a proxy resource with the HTTP proxy integration type, create an API resource with a greedy path parameter (for example, /parent/{proxy+}) and integrate this resource with an HTTP backend endpoint (for example, https://petstore-demo-endpoint.execute-api.com/petstore/{proxy}) on the ANY method. The greedy path parameter must be at the end of the resource path.

As with a non-proxy resource, you can set up a proxy resource with the HTTP proxy integration by using the API Gateway console, importing an OpenAPI definition file, or calling the API Gateway REST API directly. For detailed instructions about using the API Gateway console to configure a proxy resource with the HTTP integration, see Tutorial: Build a REST API with HTTP proxy integration.

The following OpenAPI definition file shows an example of an API with a proxy resource that is integrated with the PetStore website.

OpenAPI 3.0
{ "openapi": "3.0.0", "info": { "version": "2016-09-12T23:19:28Z", "title": "PetStoreWithProxyResource" }, "paths": { "/{proxy+}": { "x-amazon-apigateway-any-method": { "parameters": [ { "name": "proxy", "in": "path", "required": true, "schema": { "type": "string" } } ], "responses": {}, "x-amazon-apigateway-integration": { "responses": { "default": { "statusCode": "200" } }, "requestParameters": { "integration.request.path.proxy": "method.request.path.proxy" }, "uri": "http://petstore-demo-endpoint.execute-api.com/petstore/{proxy}", "passthroughBehavior": "when_no_match", "httpMethod": "ANY", "cacheNamespace": "rbftud", "cacheKeyParameters": [ "method.request.path.proxy" ], "type": "http_proxy" } } } }, "servers": [ { "url": "https://4z9giyi2c1.execute-api.us-east-1.amazonaws.com/{basePath}", "variables": { "basePath": { "default": "/test" } } } ] }
OpenAPI 2.0
{ "swagger": "2.0", "info": { "version": "2016-09-12T23:19:28Z", "title": "PetStoreWithProxyResource" }, "host": "4z9giyi2c1.execute-api.us-east-1.amazonaws.com", "basePath": "/test", "schemes": [ "https" ], "paths": { "/{proxy+}": { "x-amazon-apigateway-any-method": { "produces": [ "application/json" ], "parameters": [ { "name": "proxy", "in": "path", "required": true, "type": "string" } ], "responses": {}, "x-amazon-apigateway-integration": { "responses": { "default": { "statusCode": "200" } }, "requestParameters": { "integration.request.path.proxy": "method.request.path.proxy" }, "uri": "http://petstore-demo-endpoint.execute-api.com/petstore/{proxy}", "passthroughBehavior": "when_no_match", "httpMethod": "ANY", "cacheNamespace": "rbftud", "cacheKeyParameters": [ "method.request.path.proxy" ], "type": "http_proxy" } } } } }

In this example, a cache key is declared on the method.request.path.proxy path parameter of the proxy resource. This is the default setting when you create the API using the API Gateway console. The API's base path (/test, corresponding to a stage) is mapped to the website's PetStore page (/petstore). The single integration request mirrors the entire PetStore website using the API's greedy path variable and the catch-all ANY method. The following examples illustrate this mirroring.

  • Set ANY as GET and {proxy+} as pets

    Method request initiated from the frontend:

    GET https://4z9giyi2c1.execute-api.us-west-2.amazonaws.com/test/pets HTTP/1.1

    Integration request sent to the backend:

    GET http://petstore-demo-endpoint.execute-api.com/petstore/pets HTTP/1.1

    The run-time instances of the ANY method and proxy resource are both valid. The call returns a 200 OK response with the payload containing the first batch of pets, as returned from the backend.

  • Set ANY as GET and {proxy+} as pets?type=dog

    GET https://4z9giyi2c1.execute-api.us-west-2.amazonaws.com/test/pets?type=dog HTTP/1.1

    Integration request sent to the backend:

    GET http://petstore-demo-endpoint.execute-api.com/petstore/pets?type=dog HTTP/1.1

    The run-time instances of the ANY method and proxy resource are both valid. The call returns a 200 OK response with the payload containing the first batch of specified dogs, as returned from the backend.

  • Set ANY as GET and {proxy+} as pets/{petId}

    Method request initiated from the frontend:

    GET https://4z9giyi2c1.execute-api.us-west-2.amazonaws.com/test/pets/1 HTTP/1.1

    Integration request sent to the backend:

    GET http://petstore-demo-endpoint.execute-api.com/petstore/pets/1 HTTP/1.1

    The run-time instances of the ANY method and proxy resource are both valid. The call returns a 200 OK response with the payload containing the specified pet, as returned from the backend.

  • Set ANY as POST and {proxy+} as pets

    Method request initiated from the frontend:

    POST https://4z9giyi2c1.execute-api.us-west-2.amazonaws.com/test/pets HTTP/1.1 Content-Type: application/json Content-Length: ... { "type" : "dog", "price" : 1001.00 }

    Integration request sent to the backend:

    POST http://petstore-demo-endpoint.execute-api.com/petstore/pets HTTP/1.1 Content-Type: application/json Content-Length: ... { "type" : "dog", "price" : 1001.00 }

    The run-time instances of the ANY method and proxy resource are both valid. The call returns a 200 OK response with the payload containing the newly created pet, as returned from the backend.

  • Set ANY as GET and {proxy+} as pets/cat

    Method request initiated from the frontend:

    GET https://4z9giyi2c1.execute-api.us-west-2.amazonaws.com/test/pets/cat

    Integration request sent to the backend:

    GET http://petstore-demo-endpoint.execute-api.com/petstore/pets/cat

    The run-time instance of the proxy resource path does not correspond to a backend endpoint and the resulting request is invalid. As a result, a 400 Bad Request response is returned with the following error message.

    { "errors": [ { "key": "Pet2.type", "message": "Missing required field" }, { "key": "Pet2.price", "message": "Missing required field" } ] }
  • Set ANY as GET and {proxy+} as null

    Method request initiated from the frontend:

    GET https://4z9giyi2c1.execute-api.us-west-2.amazonaws.com/test

    Integration request sent to the backend:

    GET http://petstore-demo-endpoint.execute-api.com/petstore/pets

    The targeted resource is the parent of the proxy resource, but the run-time instance of the ANY method is not defined in the API on that resource. As a result, this GET request returns a 403 Forbidden response with the Missing Authentication Token error message as returned by API Gateway. If the API exposes the ANY or GET method on the parent resource (/), the call returns a 404 Not Found response with the Cannot GET /petstore message as returned from the backend.

For any client request, if the targeted endpoint URL is invalid or the HTTP verb is valid but not supported, the backend returns a 404 Not Found response. For an unsupported HTTP method, a 403 Forbidden response is returned.

Set up HTTP custom integrations in API Gateway

With the HTTP custom integration, you have more control of which data to pass between an API method and an API integration and how to pass the data. You do this using data mappings.

As part of the method request setup, you set the requestParameters property on a Method resource. This declares which method request parameters, which are provisioned from the client, are to be mapped to integration request parameters or applicable body properties before being dispatched to the backend. Then, as part of the integration request setup, you set the requestParameters property on the corresponding Integration resource to specify the parameter-to-parameter mappings. You also set the requestTemplates property to specify mapping templates, one for each supported content type. The mapping templates map method request parameters, or body, to the integration request body.

Similarly, as part of the method response setup, you set the responseParameters property on the MethodResponse resource. This declares which method response parameters, to be dispatched to the client, are to be mapped from integration response parameters or certain applicable body properties that were returned from the backend. Then, as part of the integration response setup, you set the responseParameters property on the corresponding IntegrationResponse resource to specify the parameter-to-parameter mappings. You also set the responseTemplates map to specify mapping templates, one for each supported content type. The mapping templates map integration response parameters, or integration response body properties, to the method response body.

For more information about setting up mapping templates, see Setting up data transformations for REST APIs.