import os
from langchain_openai import OpenAI
from langchain_openai import ChatOpenAI

os.environ["OPENAI_API_KEY"] = 'api-key'
os.environ["OPENAI_API_BASE"] = 'https://api.moonshot.cn/v1/'

api_key = os.getenv("OPENAI_API_KEY")
base_url = os.getenv("OPENAI_API_BASE")

print(api_key, base_url)

# 初始化模型(两种方式)：

llm = OpenAI(
    openai_api_base=base_url,
    openai_api_key=api_key,
    model_name="moonshot-v1-8k",
    temperature=0.7,
)

chat = ChatOpenAI(
    openai_api_base=base_url,
    openai_api_key=api_key,
    model_name="moonshot-v1-8k",
    temperature=0.7,
).bind(logprobs=True)

from langchain_core.messages import HumanMessage, SystemMessage

chat.invoke(
    [
        HumanMessage(
            content="Translate this sentence from English to French: I love programming."
        )
    ]
)

messages = [
    SystemMessage(
        content="You are a helpful assistant that translates English to French."
    ),
    HumanMessage(content="I love programming."),
]
chat(messages)

batch_messages = [
    [
        SystemMessage(
            content="You are a helpful assistant that translates English to chinese."
        ),
        HumanMessage(content="I love programming."),
    ],
    [
        SystemMessage(
            content="You are a helpful assistant that translates English to chinese."
        ),
        HumanMessage(content="I love artificial intelligence."),
    ],
]
result = chat.generate(batch_messages)
# 在Python中，当你想要在控制台中输出变量的值时，通常会使用 print 函数。但是，如果你在交互式环境（比如Python解释器或Jupyter Notebook）中执行这段代码，并且想要查看 result 的值，你可以直接输入变量名 result，而不需要使用 print 函数。
result

# Caching 缓存
# LangChain为聊天模型提供了一个可选的缓存层。这很有用，原因有两个：
# 如果您经常多次请求相同的完成，它可以通过减少您向LLM提供商进行的 API 调用次数来为您节省资金。它可以通过减少您对LLM提供程序进行的 API 调用次数来加快应用程序的速度。

# In Memory Cache 内存缓存中

# <!-- ruff: noqa: F821 -->
from langchain.globals import set_llm_cache

# %time
from langchain.cache import InMemoryCache

set_llm_cache(InMemoryCache())

# The first time, it is not yet in cache, so it should take longer
chat.invoke("Tell me a joke")

# %time
# The second time it is, so it goes faster
chat.invoke("Tell me a joke")

# SQLite Cache SQLite缓存
# We can do the same thing with a SQLite cache
from langchain.cache import SQLiteCache

set_llm_cache(SQLiteCache(database_path="langchain.db"))

# %time
# The first time, it is not yet in cache, so it should take longer
chat.invoke("Tell me a joke")

# %time
# The second time it is, so it goes faster
chat.invoke("Tell me a joke")