import streamlit as st
import pandas as pd
import requests
import os

# ----------------------------
# Configuration
# ----------------------------
st.set_page_config(page_title="Instrumentation Standards AI Assistant", layout="wide")

# ----------------------------
# Load Data
# ----------------------------
@st.cache_data
def load_data():
    df = pd.read_excel("data.xlsx", engine='openpyxl')
    return df

df = load_data()

# ----------------------------
# GROQ API Summarizer
# ----------------------------
GROQ_API_KEY = os.getenv("Standardssearch")  # Set this in your environment or .streamlit/secrets.toml

def summarize_with_groq(text, model="llama3-70b-8192"):
    url = "https://api.groq.com/openai/v1/chat/completions"
    headers = {
        "Authorization": f"Bearer {GROQ_API_KEY}",
        "Content-Type": "application/json"
    }
    payload = {
        "model": model,
        "messages": [
            {"role": "system", "content": "You are a helpful assistant summarizing industrial standards for instrumentation and control engineers."},
            {"role": "user", "content": f"Summarize the following technical standard for a professional audience:\n\n{text}"}
        ],
        "temperature": 0.5
    }

    response = requests.post(url, headers=headers, json=payload)
    if response.status_code == 200:
        return response.json()["choices"][0]["message"]["content"]
    else:
        return f"Error: {response.status_code} - {response.text}"

# ----------------------------
# UI Layout
# ----------------------------

st.title("📘 Instrumentation & Control Engineering Standards")
st.markdown("This app helps Instrumentation & Control Engineers search and understand global engineering standards.")
st.markdown("**Standards included:** ANSI, API, ASME, ASTM, BS, IEC, ISA, ISO, MSS, NACE, NAMUR, NFPA, PIP, EN, and more.")

# Search Filters
st.subheader("🔍 Search for Industrial Standards")
col1, col2 = st.columns(2)

with col1:
    entity_input = st.text_input("Enter Standard Entity").strip()

with col2:
    name_input = st.text_input("Enter Standard Name").strip()

# Filtered Results
if entity_input or name_input:
    filtered_df = df[
        df["Standards Entity"].str.contains(entity_input, case=False, na=False) &
        df["Standard Name"].str.contains(name_input, case=False, na=False)
    ]
    
    if not filtered_df.empty:
        st.success(f"✅ Found {len(filtered_df)} matching standard(s). Click to expand.")

        for index, row in filtered_df.iterrows():
            with st.expander(f"📘 {row['Standard Name']}"):
                st.markdown(f"**Entity:** {row['Standards Entity']}")
                st.markdown(f"**Description:** {row['Description']}")
                
                if st.button("Summarize with AI", key=f"summarize_{index}"):
                    with st.spinner("Querying GROQ model for summary..."):
                        summary = summarize_with_groq(row["Description"])
                        st.markdown("### ✨ AI Summary")
                        st.write(summary)
    else:
        st.warning("No matching standards found.")
else:
    st.info("Enter at least one search term to begin.")

# Footer
st.markdown("---")
st.caption("Built with ❤️ for Control & Instrumentation Engineers | Powered by GROQ + Streamlit")