Skip to main content

PlanetScale 聊天记忆

由于 PlanetScale 通过 REST API 工作，你可以将它与 Vercel Edge、Cloudflare Workers 和其他无服务器环境一起使用。

如果你希望在多个聊天会话之间实现更长期的持久化存储，可以用 PlanetScale Database 实例替换默认的内存型 chatHistory，例如用于 BufferMemory 类等聊天记忆类的数据存储。

安装设置​

你需要在项目中安装 @planetscale/database：

:::提示 请参阅安装集成包的一般说明部分。 :::

npm install @langchain/openai @planetscale/database @langchain/community @langchain/core

你还需要一个 PlanetScale 账号以及一个可以连接的数据库。关于如何创建 HTTP 客户端，请参阅 PlanetScale 文档。

使用方法​

每个存储在 PlanetScale 数据库中的聊天会话历史必须具有唯一的 ID。 config 参数会被直接传入 @planetscale/database 的 new Client() 构造函数中，并接受所有相同的参数。

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

const memory = new BufferMemory({
  chatHistory: new PlanetScaleChatMessageHistory({
    tableName: "stored_message",
    sessionId: "lc-example",
    config: {
      url: "ADD_YOURS_HERE", // Override with your own database instance's URL
    },
  }),
});

const model = new ChatOpenAI({
  model: "gpt-4o-mini",
});
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:

高级用法​

你也可以直接传入一个之前创建好的 @planetscale/database 客户端实例：

import { BufferMemory } from "langchain/memory";
import { PlanetScaleChatMessageHistory } from "@langchain/community/stores/message/planetscale";
import { ChatOpenAI } from "@langchain/openai";
import { ConversationChain } from "langchain/chains";
import { Client } from "@planetscale/database";

// Create your own Planetscale database client
const client = new Client({
  url: "ADD_YOURS_HERE", // Override with your own database instance's URL
});

const memory = new BufferMemory({
  chatHistory: new PlanetScaleChatMessageHistory({
    tableName: "stored_message",
    sessionId: "lc-example",
    client, // You can reuse your existing database client
  }),
});

const model = new ChatOpenAI({
  model: "gpt-4o-mini",
});
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:

Was this page helpful?


You can also leave detailed feedback on GitHub.