pages/sql.py

import streamlit as st
import os
from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
from langchain_core.messages import HumanMessage, AIMessage, SystemMessage



hf = os.getenv('hf')
os.environ['HUGGINGFACEHUB_API_TOKEN'] = hf
os.environ['HF_TOKEN'] = hf
# --- Config ---
st.set_page_config(page_title="AI Mentor Chat", layout="centered")
st.title("🤖 AI Mentor Chat")

# --- Sidebar for selections ---
st.sidebar.title("Mentor Preferences")

exp1 = ['<1', '1', '2', '3', '4', '5', '5+']
exp = st.sidebar.selectbox("Select experience:", exp1)

# Map experience to label
experience_map = {
    '<1': 'New bie mentor',
    '1': '1', '2': '2', '3': '3', '4': '4', '5': '5',
    '5+': 'Professional'
}
experience_label = experience_map[exp]

# --- Initialize Chat Model ---
deep_seek_skeleton = HuggingFaceEndpoint(
    repo_id='marin-community/marin-8b-instruct',
    provider='together',
    temperature=0.7,
    max_new_tokens=110,
    task='conversational'
)

deep_seek = ChatHuggingFace(
    llm=deep_seek_skeleton,
    repo_id='marin-community/marin-8b-instruct',
    provider='together',
    temperature=0.7,
    max_new_tokens=110,
    task='conversational'
)

# --- Session State ---
PAGE_KEY = "sql_chat_history" 
try:
    # --- Session State ---
    if PAGE_KEY not in st.session_state:
        st.session_state[PAGE_KEY] = []
    
    # --- Chat Form ---
    with st.form(key="chat_form"):
        user_input = st.text_input("Ask your question:")
        submit = st.form_submit_button("Send")
    
    # --- Chat Logic ---
    if submit and user_input:
        # Add system context
        system_prompt = f"""Act as a SQL mentor with {experience_label} years of experience. Teach in a friendly, approachable manner while following these strict rules:
        1. Only answer questions related to SQL (including frameworks)
        2. For any non-SQL query, respond with exactly: "I specialize only in SQL programming. This appears to be a non-SQL topic."
        3. Never suggest you can help with non-SQL topics
        4. Keep explanations clear, practical, and beginner-friendly when appropriate
        5. Include practical examples when explaining concepts
        6. For advanced topics, assume the student has basic SQL knowledge"""      
        # Create message list
        messages = [SystemMessage(content=system_prompt), HumanMessage(content=user_input)]
    
        # Get model response
        result = deep_seek.invoke(messages)
    
        # Append to history
        st.session_state[PAGE_KEY].append((user_input, result.content))
    
    # --- Display Chat History ---
    st.subheader("🗨️ Chat History")
    for user, bot in st.session_state[PAGE_KEY]:
        st.markdown(f"**You:** {user}")
        st.markdown(f"**Mentor:** {bot}")
        st.markdown("---")
except:
    st.warning('The token limit has reached please revisit in 24 hours!')