Skip to main content

DynamoDB 支持的聊天内存

¥DynamoDB-Backed Chat Memory

为了在聊天会话中实现更长期的持久化，你可以将支持聊天内存类（如 BufferMemory）的默认内存 chatHistory 替换为 DynamoDB 实例。

¥For longer-term persistence across chat sessions, you can swap out the default in-memory chatHistory that backs chat memory classes like BufferMemory for a DynamoDB instance.

设置​

¥Setup

首先，在你的项目中安装 AWS DynamoDB 客户端：

¥First, install the AWS DynamoDB client in your project:

npm install @aws-sdk/client-dynamodb
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

接下来，登录你的 AWS 账户并创建一个 DynamoDB 表。将表命名为 langchain，并将分区键命名为 id。确保你的分区键是字符串。你可以保留排序键和其他设置。

¥Next, sign into your AWS account and create a DynamoDB table. Name the table langchain, and name your partition key id. Make sure your partition key is a string. You can leave sort key and the other settings alone.

你还需要检索有权访问该表的角色或用户的 AWS 访问密钥和密钥，并将其添加到你的环境变量中。

¥You'll also need to retrieve an AWS access key and secret key for a role or user that has access to the table and add them to your environment variables.

用法​

¥Usage

import { BufferMemory } from "langchain/memory";
import { DynamoDBChatMessageHistory } from "@langchain/community/stores/message/dynamodb";
import { ChatOpenAI } from "@langchain/openai";
import { ConversationChain } from "langchain/chains";

const memory = new BufferMemory({
  chatHistory: new DynamoDBChatMessageHistory({
    tableName: "langchain",
    partitionKey: "id",
    sessionId: new Date().toISOString(), // Or some other unique identifier for the conversation
    config: {
      region: "us-east-2",
      credentials: {
        accessKeyId: "<your AWS access key id>",
        secretAccessKey: "<your AWS secret access key>",
      },
    },
  }),
});

const model = new ChatOpenAI();
const chain = new ConversationChain({ llm: model, memory });

const res1 = await chain.invoke({ input: "Hi! I'm Jim." });
console.log({ res1 });
/*
{
  res1: {
    text: "Hello Jim! It's nice to meet you. My name is AI. How may I assist you today?"
  }
}
*/

const res2 = await chain.invoke({ input: "What did I just say my name was?" });
console.log({ res2 });

/*
{
  res1: {
    text: "You said your name was Jim."
  }
}
*/

API Reference: