如何从 node.js 对话框流包中设置对话框流上下文答案

【问题标题】：How to set dialogflow context from node.js dialogflow package如何从 node.js 对话框流包中设置对话框流上下文
【发布时间】：2020-10-31 15:45:10
【问题描述】：

我正在使用sessionClient.detectIntent() 将文本发送到 dialogflow，它工作正常，现在我也想发送 context 文本。我该怎么做？

【问题讨论】：

标签： node.js dialogflow-es dialogflow-es-fulfillment dialogflow-cx

【解决方案1】：

我通常通过以下方式创建上下文：

exports.createContext = async function(contextId, parameters, sessionPath, sessionId, lifespan = 333) {
    let contextPath;
    try {
        contextPath = contextClient.contextPath(projectId, sessionId, contextId);

    } catch (e) {
        console.error("Error => ");
        console.error(e)
    }
    const request = {
        parent: sessionPath,
        context: {
            name: contextPath,
            parameters: struct.encode(parameters),
            lifespanCount: lifespan
        }
    };
    contextClient.createContext(request)
};

当我必须在调用 detectIntent 方法之前创建上下文时，只需调用此方法：

bot.createContext('CONTEXT_NAME_GOES_HERE', '{PARAMETER: VALUE}', sessionPath, session_id, lifespan = 1);

【讨论】：

【解决方案2】：

您可以在detectIntent 上将contexts 与查询文本一起发送，方法是将其添加到queryParams 下的上下文数组字段中。请注意，这种通过 detectIntent 发送上下文的方法将在执行查询之前创建（如果未创建）并激活上下文。

可以参考下面的代码sn-p：

const dialogflow = require('@google-cloud/dialogflow');

/**
 * Send a query and a context to the Dialogflow agent, and return the query result.
 * @param {string} projectId The project to be used
 */
async function detectIntent(projectId, sessionId, text) {

  // Create a new session
  const sessionClient = new dialogflow.SessionsClient();
  const sessionPath = sessionClient.projectAgentSessionPath(projectId, sessionId);

  // The text query request.
  const request = {
    session: sessionPath,
    queryParams:{
        //List of context to be sent and activated before the query is executed
        contexts:[
            {
                // The context to be sent and activated or overrated in the session 
                name: `projects/${projectId}/agent/sessions/${sessionId}/contexts/CONTEXT_NAME`,
                // The lifespan of the context
                lifespanCount: 8
              }
        ]
    },
    queryInput: {
      text: {
        // The query to send to the dialogflow agent
        text: text,
        // The language used by the client (en-US)
        languageCode: 'en-US',
      },
    },
  };

  // Send request and log result
  const responses = await sessionClient.detectIntent(request);
  console.log('Detected intent');
  const result = responses[0].queryResult;
  console.log(`  Query: ${result.queryText}`);
  console.log(`  Response: ${result.fulfillmentText}`);
  console.log(`  Output Contexts: ${JSON.stringify(result.outputContexts)}`)
}

detectIntent("PROJECT_ID","SESSION_ID","TEXT_QUERY");

输出：

Detected intent
  Query: Hi
  Response: Hello! How can I help you?
  Output Contexts: [{"name":"projects/PROJECT_ID/agent/sessions/SESSION_ID/contexts/CONTEXT_NAME","lifespanCount":7,"parameters":null}]

【讨论】：