utils/rag_utils.py

from PyPDF2 import PdfReader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_groq import ChatGroq
from langchain.chains import RetrievalQA

def process_pdf(file_path: str):
    reader = PdfReader(file_path)
    text = "\n".join([page.extract_text() or "" for page in reader.pages])
    
    splitter = RecursiveCharacterTextSplitter(
        chunk_size=1000,
        chunk_overlap=200,
        length_function=len
    )
    return splitter.split_text(text)

def get_groq_response(query, vector_db, model_name="mixtral-8x7b-32768"):
    try:
        # Updated to use current LLaMA3 model (replace with latest from Groq docs)
        llm = ChatGroq(
            temperature=0.1,
            model_name="llama3-70b-8192",  # Current recommended model
            max_tokens=2048
        )
        
        qa = RetrievalQA.from_chain_type(
            llm=llm,
            chain_type="stuff",
            retriever=vector_db.as_retriever(search_kwargs={"k": 4})
        )
        return qa.run(query)
    except Exception as e:
        # Provide helpful error message
        error_msg = str(e)
        if "model_decommissioned" in error_msg:
            return "Error: Please update the model name in rag_utils.py - check Groq's latest docs"
        return f"Error processing request: {error_msg}"