import os
import sys
from pathlib import Path
from dotenv import load_dotenv

from langchain_openai import ChatOpenAI
from langchain_ollama import OllamaEmbeddings
from langchain_chroma import Chroma
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain.tools import tool
from langchain.agents import AgentExecutor, create_openai_tools_agent
from langchain_community.tools.tavily import TavilySearchResults

# Load env variables
load_dotenv()

# ---------- LLM and embeddings ----------
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,
)
embeddings = OllamaEmbeddings(model="nomic-embed-text")

# ---------- Vectorstore ----------
PERSIST_DIR = Path("./chroma_db")
PERSIST_DIR.mkdir(parents=True, exist_ok=True)
vectorstore = Chroma(persist_directory=str(PERSIST_DIR), embedding_function=embeddings)

# Load documents if collection empty
if not vectorstore.get_collection().count():
    docs_dir = Path("./documents")
    if docs_dir.exists():
        splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
        for file in docs_dir.glob("**/*.*"):
            if file.suffix.lower() in {".txt", ".md"}:
                text = file.read_text(encoding="utf-8")
                chunks = splitter.split_text(text)
                vectorstore.add_texts(chunks, ids=[f"{file.name}_{i}" for i in range(len(chunks))])
    vectorstore.persist()

# ---------- Tools ----------
@tool("search_local_kb")
def search_local_kb(query: str, top_k: int = 3) -> str:
    """Semantic search in local ChromaDB knowledge base."""
    retriever = vectorstore.as_retriever(search_kwargs={"k": top_k})
    docs = retriever.get_relevant_documents(query)
    if not docs:
        return "No relevant documents found in local KB."
    return "\n---\n".join(doc.page_content for doc in docs)

@tool("web_search")
def web_search(query: str) -> str:
    """Web search using Tavily."""
    tavily = TavilySearchResults(max_results=3)
    results = tavily.run(query)
    if not results:
        return "No web results found."
    return "\n---\n".join(f"{r['title']}\n{r['url']}\n{r.get('content', '')}" for r in results)

# ---------- Agent ----------
SYSTEM_PROMPT = (
    "You are an AI assistant. For questions about local documents, use the tool 'search_local_kb'. "
    "For up-to-date information or news, use 'web_search'. Always indicate the source in your answer."
)

tools = [search_local_kb, web_search]
agent = create_openai_tools_agent(llm, tools, system_message=SYSTEM_PROMPT)
agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

# ---------- CLI ----------

def main():
    print("RAG Agent with ChromaDB and Tavily. Type 'exit' to quit.")
    while True:
        try:
            query = input("\nЗапрос: ")
        except (EOFError, KeyboardInterrupt):
            print("\nBye!")
            break
        if query.strip().lower() in {"exit", "quit"}:
            print("Bye!")
            break
        result = agent_executor.invoke({"input": query})
        # The tool name is stored in the tool_calls field of the result
        tool_name = result.get("tool_calls", [{}])[0].get("name", "unknown")
        source = "tavily" if tool_name == "web_search" else "chromadb"
        print(f"\n[{'Web Search' if source=='tavily' else 'Local KB'}] {result.get('output', '')}\n")
        print(f"Источник: {source}\n")

if __name__ == "__main__":
    main()