Skip to main content

Aurora DSQL 聊天内存

¥Aurora DSQL Chat Memory

为了在聊天会话中实现更长期的持久化，你可以将默认的内存中 chatHistory 替换为与 PostgreSQL 兼容的无服务器 Amazon Aurora DSQL 数据库。

¥For longer-term persistence across chat sessions, you can swap out the default in-memory chatHistory for the serverless PostgreSQL-compatible Amazon Aurora DSQL Database.

这与 PostgreSQL 集成非常相似，但有一些区别，以使其与 DSQL 兼容：

¥This is very similar to the PostgreSQL integration with a few differences to make it compatible with DSQL:

  1. PostgreSQL 中的 id 列是 SERIAL 自增类型，DSQL 是使用数据库函数 gen_random_uuid 的 UUID 格式。

    ¥The id column in PostgreSQL is SERIAL auto-incrementent, and DSQL is UUID using the database function gen_random_uuid.

  2. 创建 created_at 列来跟踪消息的顺序和历史记录。

    ¥A created_at column is created to track the order and history of the messages.

  3. PostgreSQL 中的 message 列是 JSONB 格式，DSQL 是经过 JavaScript 解析处理的 TEXT 格式。

    ¥The message column in PostgreSQL is JSONB, and DSQL is TEXT with Javascript parsing handling

设置​

¥Setup

转到你的 AWS 控制台并创建一个 Aurora DSQL 集群 https://console.aws.amazon.com/dsql/clusters

¥Go to you AWS Console and create an Aurora DSQL Cluster, https://console.aws.amazon.com/dsql/clusters

tip

请参阅 此部分包含有关安装集成包的一般说明。

¥See this section for general instructions on installing integration packages.

npm install @langchain/openai @langchain/community @langchain/core pg @aws-sdk/dsql-signer

用法​

¥Usage

每个聊天历史记录会话都存储在 Aurora DSQL（兼容 Postgres）数据库中，并且需要一个会话 ID。

¥Each chat history session is stored in a Aurora DSQL (Postgres-compatible) database and requires a session id.

与 Aurora DSQL 的连接通过 PostgreSQL 池处理。你可以通过 pool 参数传递一个池实例，也可以通过 poolConfig 参数传递一个池配置。有关更多信息，请参阅 pg-node 池文档。提供的池优先，因此，如果同时传递了池实例和池配置，则只会使用池。

¥The connection to Aurora DSQL is handled through a PostgreSQL pool. You can either pass an instance of a pool via the pool parameter or pass a pool config via the poolConfig parameter. See pg-node docs on pools for more information. A provided pool takes precedence, thus if both a pool instance and a pool config are passed, only the pool will be used.

有关常规安全最佳实践的更多信息，请参阅我们的 https://docs.aws.amazon.com/aurora-dsql/latest/userguide/authentication-authorization.html。

¥For options on how to do the authentication and authorization for DSQL please check https://docs.aws.amazon.com/aurora-dsql/latest/userguide/authentication-authorization.html.

以下示例使用 AWS-SDK 生成传递给池配置的身份验证令牌：

¥The following example uses the AWS-SDK to generate an authentication token that is passed to the pool configuration:

import pg from "pg";

import { DsqlSigner } from "@aws-sdk/dsql-signer";
import { AuroraDsqlChatMessageHistory } from "@langchain/community/stores/message/aurora_dsql";
import { ChatOpenAI } from "@langchain/openai";
import { RunnableWithMessageHistory } from "@langchain/core/runnables";

import {
  ChatPromptTemplate,
  MessagesPlaceholder,
} from "@langchain/core/prompts";
import { StringOutputParser } from "@langchain/core/output_parsers";

async function getPostgresqlPool() {
  const signer = new DsqlSigner({
    hostname: process.env.DSQL_ENDPOINT!,
  });

  const token = await signer.getDbConnectAdminAuthToken();

  if (!token) throw new Error("Auth token error for DSQL");

  const poolConfig: pg.PoolConfig = {
    host: process.env.DSQL_ENDPOINT,
    port: 5432,
    user: "admin",
    password: token,
    ssl: true,
    database: "postgres",
  };

  const pool = new pg.Pool(poolConfig);
  return pool;
}

const pool = await getPostgresqlPool();

const model = new ChatOpenAI();

const prompt = ChatPromptTemplate.fromMessages([
  [
    "system",
    "You are a helpful assistant. Answer all questions to the best of your ability.",
  ],
  new MessagesPlaceholder("chat_history"),
  ["human", "{input}"],
]);

const chain = prompt.pipe(model).pipe(new StringOutputParser());

const chainWithHistory = new RunnableWithMessageHistory({
  runnable: chain,
  inputMessagesKey: "input",
  historyMessagesKey: "chat_history",
  getMessageHistory: async (sessionId) => {
    const chatHistory = new AuroraDsqlChatMessageHistory({
      sessionId,
      pool,
      // Can also pass `poolConfig` to initialize the pool internally,
      // but easier to call `.end()` at the end later.
    });
    return chatHistory;
  },
});

const res1 = await chainWithHistory.invoke(
  {
    input: "Hi! I'm MJDeligan.",
  },
  { configurable: { sessionId: "langchain-test-session" } }
);
console.log(res1);
/*
  "Hello MJDeligan! It's nice to meet you. My name is AI. How may I assist you today?"
*/

const res2 = await chainWithHistory.invoke(
  { input: "What did I just say my name was?" },
  { configurable: { sessionId: "langchain-test-session" } }
);
console.log(res2);

/*
  "You said your name was MJDeligan."
*/

// If you provided a pool config you should close the created pool when you are done
await pool.end();

API Reference: