import os
from langchain_openai import ChatOpenAI
from langchain_core.messages import HumanMessage
from langchain.tools import tool
from langgraph.checkpoint.memory import MemorySaver
from langgraph.types import Command

# LLM setup
llm = ChatOpenAI(
    model="openai/gpt-oss-20b:free",
    base_url="https://openrouter.ai/api/v1",
    api_key=os.getenv("OPENAI_API_KEY"),
    temperature=0.0,
)

# Dummy tool: get_price
@tool
def get_price(product: str, city: str) -> str:
    """Return a markdown table with price for a product in a city."""
    # In a real scenario, fetch from API. Here we return a static example.
    return f"| Продукт | Цена (руб.) | Магазин |\n| {product} | 89 | Магнит |"

# Agent definition
from langchain.agents import create_agent

agent = create_agent(
    llm=llm,
    tools=[get_price],
    system_prompt="You are a helpful assistant that can call get_price to provide price information.",
)

# Stream execution
stream = agent.stream(
    {
        "messages": [HumanMessage(content="Покажи цену молока и хлеба в Казани.")]
    },
    stream_mode=["messages", "updates"],
)

step = 1

def format_chunk_message(chunk):
    message, meta = chunk
    global step
    if meta.get("langgraph_step") != step:
        step = meta.get("langgraph_step")
        print("\n --- --- --- \n")
    if message.content:
        print(message.content, end="", flush=True)

def format_message(message):
    if message.content:
        return message.content
    return f"{message.tool_calls[0]['name']}({message.tool_calls[0]['args']})"

for chunk in stream:
    chunk_type, chunk_data = chunk
    if chunk_type == "messages":
        format_chunk_message(chunk_data)
    elif chunk_type == "updates":
        if chunk_data.get("model"):
            last_message = chunk_data["model"]["messages"][-1]
            print(format_message(last_message))

print("\n--- Завершено ---")