import fitz  # PyMuPDF
import numpy as np
import faiss
import requests
import os
from sentence_transformers import SentenceTransformer
from langchain.text_splitter import RecursiveCharacterTextSplitter
import gradio as gr

# 🔹 Step 1: PDF File Path from Hugging Face local space
pdf_path = "our_philosophy-_falsafatuna (1).pdf"  # Must be uploaded to "Files and versions" in your Space

# 🔹 Step 2: Extract Text from PDF
doc = fitz.open(pdf_path)
text = ""
for page in doc:
    text += page.get_text()

# 🔹 Step 3: Split Text into Chunks
splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
chunks = splitter.split_text(text)

# 🔹 Step 4: Create Embeddings and FAISS Index
model = SentenceTransformer("all-MiniLM-L6-v2")
embeddings = model.encode(chunks)

dimension = embeddings.shape[1]
index = faiss.IndexFlatL2(dimension)
index.add(np.array(embeddings))

chunk_list = chunks  # Used for retrieval

# 🔹 Step 5: RAG Query
def query_rag(question, k=3):
    question_embedding = model.encode([question])
    D, I = index.search(np.array(question_embedding), k)
    retrieved_chunks = [chunk_list[i] for i in I[0]]
    context = "\n".join(retrieved_chunks)

    prompt = f"Answer the question based on the following context:\n{context}\n\nQuestion: {question}\nAnswer:"
    return prompt

# 🔹 Step 6: Generate answer from Groq API using environment variable for key
def generate_answer(prompt):
    GROQ_API_KEY = os.environ["GROQ_API_KEY"]  # Secure way to get API key
    url = "https://api.groq.com/openai/v1/chat/completions"
    headers = {
        "Authorization": f"Bearer {GROQ_API_KEY}",
        "Content-Type": "application/json"
    }
    data = {
        "model": "llama3-8b-8192",
        "messages": [{"role": "user", "content": prompt}],
        "temperature": 0.3
    }
    response = requests.post(url, headers=headers, json=data)
    return response.json()['choices'][0]['message']['content']

# 🔹 Step 7: Full RAG Pipeline
def rag_pipeline(question):
    prompt = query_rag(question)
    answer = generate_answer(prompt)
    return answer

# 🔹 Step 8: Gradio Interface
interface = gr.Interface(
    fn=rag_pipeline,
    inputs=gr.Textbox(lines=2, placeholder="Ask any question from Falsafatuna..."),
    outputs="text",
    title="📘 Read ❤️Falsafatuna❤️ (Our Philosophy) by Allama Muhammad Baqir as-Sadr",
    description="Developed by Najaf Ali Sharqi — Educator, researcher and advocate of AI for Education. This app allows you to ask any question from the book *Falsafatuna* and receive intelligent responses using Groq + LLaMA3."
)

interface.launch()