Lambda 响应发送回 Lex 插槽？

Question

我已经编写了 5-6 个插槽并具有调用第三方天气 api 的 lambda 函数，并且我正在从 api 获得响应。现在我该如何处理该响应并将其与响应一起发送回 lex 插槽。

例如：在 Lex 中，槽 Country 我输入 India，然后 City 我输入 Hyderabad。在这里我调用 lambda 函数并希望响应应该来自带有温度详细信息的 lex 槽。

我正在使用 Lex 控制台和 Lambda 函数作为内联代码编辑器。

Answer 1

我将占用 2 个插槽，并在代码中处理空插槽（python）。

所以首先你必须在你的意图中定义2个插槽city和country，机器人将检查DialogCodeHook中的插槽值是否已填充，如果验证成功，它将调用FulfillmentCodeHook调用天气 api 将结果返回给用户。

注意：您必须检查Initialization and validation code hook，以便它转到DialogCodeHook

def build_response(message):
    return {
        "dialogAction":{
            "type":"Close",
            "fulfillmentState":"Fulfilled",
            "message":{
                "contentType":"PlainText",
                "content":message
            }
        }
    }

def elicit_slot(intent_name, slots, slot_to_elicit, message):
    return {
        'dialogAction': {
            'type': 'ElicitSlot',
            'intentName': intent_name,
            'slots': slots,
            'slotToElicit': slot_to_elicit,
            'message': message
        }
    }

def delegate(session_attributes, slots):
    return {
        'sessionAttributes': session_attributes,
        'dialogAction': {
            'type': 'Delegate',
            'slots': slots
        }
    }

def perform_action(intent_request):
    source = intent_request['invocationSource']   # DialogCodeHook or FulfillmentCodeHook
    slots = intent_request['currentIntent']['slots']  # your slots city and country
    if source == 'DialogCodeHook':
        # Perform basic validation on the supplied input slots.
        if slots['city'] is None:  # or any other validation that you want to perform
            return elicit_slot(
                intent_request['currentIntent']['name'], # current intent name
                slots, # current intent slots
                'city',  # slot name
                'Please enter city name'  # prompt the user to enter city name
            )
        if slots['country'] is None:
            return elicit_slot(
                intent_request['currentIntent']['name'], # current intent name
                slots, # current intent slots
                'country',  # slot name
                'Please enter country name' # prompt the user to enter country name
            )
        # delegate means all slot validation are done, we can move to Fulfillment
        return delegate(output_session_attributes, slots) 
    if source == 'FulfillmentCodeHook':
        result = your_api_call(slots['city'], slots['country'])
        return build_response(result)  # display the response back to user

def dispatch(intent_request):
    intent_name = intent_request['currentIntent']['name']
    # Dispatch to your bot's intent handlers
    if intent_name == 'GetWeather':
        return perform_action(intent_request)
    raise Exception('Intent with name ' + intent_name + ' not supported')

def lambda_handler(event, context):
    logger.debug(event)
    return dispatch(event)

希望对你有帮助。

Answer 2

对 Lex 的 Lambda 响应应具有以下结构，其中 dialogAction 参数是必需的

"dialogAction": 
   {
       "type": "Close",
        "fulfillmentState": "Fulfilled or Failed",
        "message": {
         "contentType": "PlainText or SSML or CustomPayload",
         "content": "Message to convey to the user. For example, Thanks, your pizza has been 
 ordered."
     },
      "responseCard": {
         "version": integer-value,
         "contentType": "application/vnd.amazonaws.card.generic",
         "genericAttachments": [
            {
             "title":"card-title",
             "subTitle":"card-sub-title",
             "imageUrl":"URL of the image to be shown",
             "attachmentLinkUrl":"URL of the attachment to be associated with the card",
             "buttons":[ 
                 {
                    "text":"button-text",
                    "value":"Value sent to server on button click"
                 }
              ]
           } 
       ] 
     }

您可以使用以下代码作为示例，在回调中创建 node.js 响应。这里使用 dialogAction 强制参数创建 JSON 对象并添加其他属性。将“要返回的数据”更改为您想要返回到 AWS Lex 的值。

var obj = {};
var dialogAction = {};
   dialogAction.type="Close";
   dialogAction.fulfillmentState="Fulfilled";

   var message = {};
   message.contentType="PlainText";
   message.content="Data to return";

   dialogAction.message = message;
   obj.dialogAction = dialogAction;
   connection.end(function (err) { callback(err, obj)});