import os
import asyncio
from langchain_ollama import ChatOllama, OllamaEmbeddings
from langchain_qdrant import QdrantVectorStore
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain_core.documents import Document
from langchain.tools import tool
from langchain_core.messages import HumanMessage
from deepagents import create_deep_agent
from deepagents.backends import CompositeBackend, LocalShellBackend, FilesystemBackend

# Инициализация эмбеддингов и LLM через Ollama
embeddings = OllamaEmbeddings(model="nomic-embed-text")
llm = ChatOllama(model="llama3")

# Инициализация QdrantVectorStore
vector_store = QdrantVectorStore(
    host="localhost",
    port=6333,
    collection_name="knowledge",
    embedding_function=embeddings,
)

# Чанкинг текста
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)

@tool
def search_knowledge_base(query: str, max_results: int = 3) -> str:
    """Поиск в базе знаний."""
    docs = vector_store.similarity_search(query, k=max_results)
    return "\n".join(d.page_content for d in docs) if docs else "No results."

@tool
def add_to_knowledge_base(content: str, title: str = "doc") -> str:
    """Добавление документа в базу знаний."""
    vector_store.add_documents([Document(page_content=content, metadata={"title": title})])
    return f"Added: {title}"

# Backend для deepagents
backend = CompositeBackend([
    LocalShellBackend(workspace_dir="./workspace"),
    FilesystemBackend(),
])

# Создание агента
agent = create_deep_agent(
    model=llm,
    tools=[search_knowledge_base, add_to_knowledge_base],
    backend=backend,
    system_prompt="You are a helpful assistant. Use the provided tools to search and add knowledge.",
)

def load_documents_from_dir(dir_path: str):
    """Загрузка всех .txt файлов из директории в векторную базу."""
    for root, _, files in os.walk(dir_path):
        for file in files:
            if file.lower().endswith(".txt"):
                file_path = os.path.join(root, file)
                with open(file_path, "r", encoding="utf-8") as f:
                    content = f.read()
                chunks = splitter.split_text(content)
                docs = [Document(page_content=chunk, metadata={"title": file}) for chunk in chunks]
                vector_store.add_documents(docs)

async def main():
    # Загрузка документов из папки data (если есть)
    data_dir = "./data"
    if os.path.isdir(data_dir):
        load_documents_from_dir(data_dir)

    print("RAG Agent ready. Commands:")
    print("/add <file_path>   - добавить документ")
    print("/search <query>    - поиск в базе")
    print("/quit              - выйти")

    while True:
        user_input = input("> ").strip()
        if not user_input:
            continue
        if user_input.startswith("/quit"):
            print("Goodbye.")
            break
        elif user_input.startswith("/add"):
            parts = user_input.split(maxsplit=1)
            if len(parts) < 2:
                print("Usage: /add <file_path>")
                continue
            file_path = parts[1]
            if not os.path.isfile(file_path):
                print(f"File not found: {file_path}")
                continue
            with open(file_path, "r", encoding="utf-8") as f:
                content = f.read()
            title = os.path.basename(file_path)
            message = f"Add document titled {title} with content: {content}"
            result = await agent.ainvoke(
                {"messages": [HumanMessage(content=message)]},
                {"configurable": {"thread_id": "session-1"}},
            )
            print(result["messages"][-1].content)
        elif user_input.startswith("/search"):
            parts = user_input.split(maxsplit=1)
            if len(parts) < 2:
                print("Usage: /search <query>")
                continue
            query = parts[1]
            message = f"Search the knowledge base for: {query}"
            result = await agent.ainvoke(
                {"messages": [HumanMessage(content=message)]},
                {"configurable": {"thread_id": "session-1"}},
            )
            print(result["messages"][-1].content)
        else:
            print("Unknown command. Use /add, /search, /quit.")

if __name__ == "__main__":
    asyncio.run(main())