Redis 聊天消息历史

Redis (Remote Dictionary Server) 是一种开源的内存存储，用作分布式内存键值数据库、缓存和消息代理，并可选持久化。Redis 提供低延迟的读写操作。Redis 是最受欢迎的 NoSQL 数据库之一，也是整体最受欢迎的数据库之一。

本笔记本演示了如何使用 langchain-redis 包中的 RedisChatMessageHistory 类，通过 Redis 存储和管理聊天消息历史。

设置

首先，我们需要安装所需的依赖项，并确保我们有一个正在运行的 Redis 实例。

%pip install -qU langchain-redis langchain-openai redis

确保您运行着一个 Redis 服务器。您可以使用 Docker，通过以下命令启动一个：

docker run -d --name redis-stack -p 6379:6379 -p 8001:8001 redis/redis-stack:latest

或者，根据您操作系统的说明安装并在本地运行 Redis。

import os

# Use the environment variable if set, otherwise default to localhost
REDIS_URL = os.getenv("REDIS_URL", "redis://localhost:6379")
print(f"Connecting to Redis at: {REDIS_URL}")

Connecting to Redis at: redis://redis:6379

导入所需库

from langchain_core.chat_history import BaseChatMessageHistory
from langchain_core.messages import AIMessage, HumanMessage
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_core.runnables.history import RunnableWithMessageHistory
from langchain_openai import ChatOpenAI
from langchain_redis import RedisChatMessageHistory

API Reference:BaseChatMessageHistory | AIMessage | HumanMessage | ChatPromptTemplate | MessagesPlaceholder | RunnableWithMessageHistory | ChatOpenAI | RedisChatMessageHistory

RedisChatMessageHistory 的基本用法

This example demonstrates how to use RedisChatMessageHistory to store and

# Initialize RedisChatMessageHistory
history = RedisChatMessageHistory(session_id="user_123", redis_url=REDIS_URL)

# Add messages to the history
history.add_user_message("Hello, AI assistant!")
history.add_ai_message("Hello! How can I assist you today?")

# Retrieve messages
print("Chat History:")
for message in history.messages:
    print(f"{type(message).__name__}: {message.content}")

Chat History:
HumanMessage: Hello, AI assistant!
AIMessage: Hello! How can I assist you today?

将 RedisChatMessageHistory 与语言模型搭配使用

This guide shows you how to use RedisChatMessageHistory with your language models.

设置 OpenAI API 密钥

from getpass import getpass

# Check if OPENAI_API_KEY is already set in the environment
openai_api_key = os.getenv("OPENAI_API_KEY")

if not openai_api_key:
    print("OpenAI API key not found in environment variables.")
    openai_api_key = getpass("Please enter your OpenAI API key: ")

    # Set the API key for the current session
    os.environ["OPENAI_API_KEY"] = openai_api_key
    print("OpenAI API key has been set for this session.")
else:
    print("OpenAI API key found in environment variables.")

OpenAI API key not found in environment variables.
``````output
Please enter your OpenAI API key:  ········
``````output
OpenAI API key has been set for this session.

# Create a prompt template
prompt = ChatPromptTemplate.from_messages(
    [
        ("system", "You are a helpful AI assistant."),
        MessagesPlaceholder(variable_name="history"),
        ("human", "{input}"),
    ]
)

# Initialize the language model
llm = ChatOpenAI()

# Create the conversational chain
chain = prompt | llm


# Function to get or create a RedisChatMessageHistory instance
def get_redis_history(session_id: str) -> BaseChatMessageHistory:
    return RedisChatMessageHistory(session_id, redis_url=REDIS_URL)


# Create a runnable with message history
chain_with_history = RunnableWithMessageHistory(
    chain, get_redis_history, input_messages_key="input", history_messages_key="history"
)

# Use the chain in a conversation
response1 = chain_with_history.invoke(
    {"input": "Hi, my name is Alice."},
    config={"configurable": {"session_id": "alice_123"}},
)
print("AI Response 1:", response1.content)

response2 = chain_with_history.invoke(
    {"input": "What's my name?"}, config={"configurable": {"session_id": "alice_123"}}
)
print("AI Response 2:", response2.content)

AI Response 1: Hello Alice! How can I assist you today?
AI Response 2: Your name is Alice.

高级特性

自定义 Redis 配置

# Initialize with custom Redis configuration
custom_history = RedisChatMessageHistory(
    "user_456",
    redis_url=REDIS_URL,
    key_prefix="custom_prefix:",
    ttl=3600,  # Set TTL to 1 hour
    index_name="custom_index",
)

custom_history.add_user_message("This is a message with custom configuration.")
print("Custom History:", custom_history.messages)

Custom History: [HumanMessage(content='This is a message with custom configuration.')]

搜索消息

# Add more messages
history.add_user_message("Tell me about artificial intelligence.")
history.add_ai_message(
    "Artificial Intelligence (AI) is a branch of computer science..."
)

# Search for messages containing a specific term
search_results = history.search_messages("artificial intelligence")
print("Search Results:")
for result in search_results:
    print(f"{result['type']}: {result['content'][:50]}...")

Search Results:
human: Tell me about artificial intelligence....
ai: Artificial Intelligence (AI) is a branch of comput...

清除历史记录

# Clear the chat history
history.clear()
print("Messages after clearing:", history.messages)

Messages after clearing: []

结论

本笔记本演示了 langchain-redis 包中的 RedisChatMessageHistory 的关键功能。它展示了如何初始化和使用聊天记录，将其与语言模型集成，并利用自定义配置和消息搜索等高级功能。Redis 为 AI 应用程序中的聊天记录管理提供了一种快速且可扩展的解决方案。