from langchain_openai import ChatOpenAI
from pydantic import SecretStr
from langchain.tools import tool
from langchain_qdrant import QdrantVectorStore
from qdrant_client import QdrantClient
from qdrant_client.http.models import Distance, VectorParams
from langchain_ollama import OllamaEmbeddings
from langchain_core.documents import Document
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain.agents import create_agent
import os

# ---------- LLM ----------
llm = ChatOpenAI(
    model="openai/gpt-oss-20b",
    base_url='https://platform.brojs.ru/jrnl-bh/api/inference/v1',
    api_key=SecretStr("jrnl_30283ab953615cbb6846ff9940a1eedce0b76d7b2f59a2394f29e74643e6a90d"),
    temperature=0.7,
)

# ---------- Vector Store ----------
client = QdrantClient(":memory:")
client.create_collection(
    collection_name="knowledge_base",
    vectors_config=VectorParams(size=1024, distance=Distance.COSINE),
)
embeddings = OllamaEmbeddings(model="nomic-embed-text")
vector_store = QdrantVectorStore(
    client=client,
    collection_name="knowledge_base",
    embedding=embeddings,
)

# ---------- Text Splitter ----------
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)

# ---------- Tools ----------
@tool
def add_to_knowledge_base(content: str, title: str) -> str:
    """Add a document to the knowledge base."""
    docs = splitter.split_text(content)
    documents = [Document(page_content=c, metadata={"title": title}) for c in docs]
    vector_store.add_documents(documents)
    return f"Added {len(docs)} chunks under title '{title}'."

@tool
def search_knowledge_base(query: str, max_results: int = 5) -> str:
    """Search the knowledge base for relevant information."""
    results = vector_store.similarity_search_with_score(query, k=max_results)
    if not results:
        return "No relevant documents found."
    out_lines = []
    for doc, score in results:
        title = doc.metadata.get("title", "Untitled")
        out_lines.append(f"[{score:.2f}] {title}: {doc.page_content[:200]}...")
    return "\n".join(out_lines)

# ---------- Agent ----------
system_prompt = """
You are an assistant with access to a knowledge base.
Use the tools `add_to_knowledge_base` and `search_knowledge_base` as needed.
Respond concisely. If you need more info, ask the user.
"""
agent = create_agent(
    model=llm,
    tools=[add_to_knowledge_base, search_knowledge_base],
    system_prompt=system_prompt,
)

# ---------- CLI ----------
def main():
    print("RAG Agent CLI. Commands: /add <title> <content>, /search <query>, /quit")
    while True:
        try:
            inp = input("> ").strip()
        except EOFError:
            break
        if not inp:
            continue
        if inp.lower() in ("exit", "quit", "/quit"):
            print("Bye!")
            break
        if inp.startswith("/add "):
            parts = inp[5:].split(None, 1)
            if len(parts) != 2:
                print("Usage: /add <title> <content>")
                continue
            title, content = parts
            res = add_to_knowledge_base(content=content, title=title)
            print(res)
        elif inp.startswith("/search "):
            query = inp[8:].strip()
            if not query:
                print("Usage: /search <query>")
                continue
            res = search_knowledge_base(query=query, max_results=5)
            print(res)
        else:
            # Regular chat with agent
            result = agent.invoke({"messages": [{"role": "human", "content": inp}]})
            ai_msg = result["messages"][-1]
            print(ai_msg.content)

if __name__ == "__main__":
    main()