from langchain_openai import ChatOpenAI, OpenAIEmbeddings
from langchain_community.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.vectorstores import Chroma
from langchain.chains import RetrievalQA

# 1. Load document
loader = PyPDFLoader("company_manual.pdf")
documents = loader.load()

# 2. Split text
splitter = RecursiveCharacterTextSplitter(
    chunk_size=1000, chunk_overlap=200
)
chunks = splitter.split_documents(documents)

# 3. Create vector database
embeddings = OpenAIEmbeddings()
vectorstore = Chroma.from_documents(chunks, embeddings)

# 4. Build a question and answer chain
qa_chain = RetrievalQA.from_chain_type(
    llm=ChatOpenAI(model="gpt-4o"),
    retriever=vectorstore.as_retriever(search_kwargs={"k": 3}),
    return_source_documents=True
)

# 5. Ask a question
result = qa_chain.invoke({"query": "What is the company's annual leave policy?"})
print(result["result"])
print("source:", [doc.metadata for doc in result["source_documents"]])

from langchain_community.document_loaders import DirectoryLoader

# Load documents from the entire directory
loader = DirectoryLoader(
    "./docs",
    glob="**/*.pdf",
    loader_cls=PyPDFLoader
)
documents = loader.load()
print(f"{len(documents)} document fragments loaded")

# The subsequent steps are the same as for single document
# Vector database will automatically handle multiple documents

from langchain_openai import ChatOpenAI
from langchain.agents import create_openai_tools_agent, AgentExecutor
from langchain_community.tools import DuckDuckGoSearchRun
from langchain.tools import tool
from langchain import hub

#define tools
search = DuckDuckGoSearchRun()

@tool
def calculate(expression: str) -> str:
    """Compute mathematical expressions, such as '2 + 2 * 3'"""
    return str(eval(expression))

tools = [search, calculate]

#Create Agent
llm = ChatOpenAI(model="gpt-4o")
prompt = hub.pull("hwchase17/openai-tools-agent")
agent = create_openai_tools_agent(llm, tools, prompt)
agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

# Run Agent
result = agent_executor.invoke({
    "input": "Search the price of Bitcoin today and calculate how much it would be worth if I had 0.5 Bitcoins"
})
print(result["output"])

import os




# Set LangSmith environment variables


os.environ["LANGCHAIN_TRACING_V2"] = "true"


os.environ["LANGCHAIN_API_KEY"] = "your-langsmith-api-key"


os.environ["LANGCHAIN_PROJECT"] = "my-rag-project"




# All subsequent LangChain calls will be automatically tracked


# Can be viewed at smith.langchain.com:


# - Input and output for each call


# - Token usage and cost


# - Delay and error messages


# - Complete call link