# 旧版繁重写法 (不推荐)
from langchain.chains import LLMChain
from langchain.prompts import PromptTemplate
from langchain_community.llms import OpenAI

chain = LLMChain(
    llm=OpenAI(),
    prompt=PromptTemplate.from_template("将以下文本翻译为英文：{text}"),
)
res = chain.run(text="你好世界")

# 现代 LCEL 写法 (标准推荐)
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(model="gpt-4o-mini")
prompt = ChatPromptTemplate.from_messages([
    ("system", "你是一名资深翻译官，风格自然地道。"),
    ("human", "翻译成英文：{text}"),
])

# 纯声明式管道：Prompt -> LLM -> Parser
chain = prompt | llm | StrOutputParser()

# 统一支持 .invoke(), .stream(), .batch(), .ainvoke()
result = chain.invoke({"text": "纸上得来终觉浅，绝知此事要躬行。"})
print("翻译结果:", result)

from langchain_core.runnables import RunnableLambda, RunnableParallel

step_upper = RunnableLambda(lambda s: s.upper())
step_length = RunnableLambda(lambda s: len(s))

# 并行执行两个独立处理逻辑
parallel_chain = RunnableParallel(
    uppercase=step_upper,
    char_count=step_length,
)

print(parallel_chain.invoke("LangChain"))
# 输出: {'uppercase': 'LANGCHAIN', 'char_count': 9}

from langchain_community.document_loaders import TextLoader
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain_community.vectorstores import Chroma
from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.runnables import RunnablePassthrough
from langchain_core.output_parsers import StrOutputParser

# 1. 加载本地知识文档
docs = TextLoader("knowledge_base.txt", encoding="utf-8").load()

# 2. 递归语义分块
splitter = RecursiveCharacterTextSplitter(chunk_size=400, chunk_overlap=50)
chunks = splitter.split_documents(docs)

# 3. 灌入向量数据库构建检索器
vectorstore = Chroma.from_documents(chunks, OpenAIEmbeddings(model="text-embedding-3-small"))
retriever = vectorstore.as_retriever(search_kwargs={"k": 3})

# 4. 构建声明式 RAG 管道
prompt_template = ChatPromptTemplate.from_template("""
仅基于以下已知信息回答问题。如果信息不足，请明确说明无法回答：
【参考背景】：
{context}

【用户问题】：
{question}
""")

rag_pipeline = (
    {"context": retriever, "question": RunnablePassthrough()}
    | prompt_template
    | ChatOpenAI(model="gpt-4o-mini", temperature=0)
    | StrOutputParser()
)

# 执行问答
answer = rag_pipeline.invoke("公司差旅报销的标准是什么？")
print("RAG 检索回答:\n", answer)

from langchain_core.tools import tool
from langchain_core.prompts import ChatPromptTemplate
from langchain.agents import create_tool_calling_agent, AgentExecutor
from langchain_openai import ChatOpenAI

@tool
def calculate_salary_tax(monthly_income: float) -> str:
    """计算个人所得税预估金额"""
    taxable = max(0.0, monthly_income - 5000)
    tax = taxable * 0.1  # 简化计算
    return f"当月应缴个税预估为: {tax:.2f} 元"

tools = [calculate_salary_tax]
llm = ChatOpenAI(model="gpt-4o-mini")

prompt = ChatPromptTemplate.from_messages([
    ("system", "你是一名专业财务助理，具备财务工具调用能力。"),
    ("human", "{input}"),
    ("placeholder", "{agent_scratchpad}"),  # 预留给中间思考和工具结果的槽位
])

agent = create_tool_calling_agent(llm, tools, prompt)
executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

res = executor.invoke({"input": "我本月税前工资是 18000 元，帮我算一下个税是多少？"})
print("最终输出:", res["output"])

client.chat.completions.create(model="gpt-4o", messages=[{"role": "user", "content": "hi"}])