import streamlit as st
import pandas as pd
import numpy as np
import requests
import os
from io import BytesIO

# -------------------------------
# PAGE CONFIG
# -------------------------------
st.set_page_config(page_title="🚀 Data QA Demo", layout="wide")
st.title("🚀 AI‑Powered Data Quality Checker")
st.markdown("Validate, clean, and showcase datasets in real‑time.")

# Read Alpha Vantage key from Hugging Face Secrets
ALPHA_KEY = os.getenv("ALPHA_VANTAGE_API_KEY")

# -------------------------------
# UTILITIES
# -------------------------------
def qa_report(df: pd.DataFrame):
    report = {
        "Rows": df.shape[0],
        "Columns": df.shape[1],
        "Missing Values": int(df.isnull().sum().sum()),
        "Duplicate Rows": int(df.duplicated().sum()),
        "Negative Values": int((df.select_dtypes(include=[np.number]) < 0).sum().sum()),
    }
    return report

def fetch_alpha_vantage(symbol="BTC", market="USD"):
    if not ALPHA_KEY:
        return None, "ALPHA_VANTAGE_API_KEY not set. Add it in Settings → Secrets."
    url = (
        "https://www.alphavantage.co/query"
        f"?function=CURRENCY_EXCHANGE_RATE&from_currency={symbol}&to_currency={market}&apikey={ALPHA_KEY}"
    )
    try:
        r = requests.get(url, timeout=15)
        r.raise_for_status()
        data = r.json()
        rate = float(data["Realtime Currency Exchange Rate"]["5. Exchange Rate"])
        return rate, None
    except Exception as e:
        return None, f"API error: {e}"

def download_csv(df: pd.DataFrame, filename="cleaned.csv"):
    buffer = BytesIO()
    df.to_csv(buffer, index=False)
    st.download_button(
        label="📥 Download Cleaned CSV",
        data=buffer.getvalue(),
        file_name=filename,
        mime="text/csv",
        use_container_width=True,
    )

# -------------------------------
# LAYOUT
# -------------------------------
tab1, tab2, tab3 = st.tabs(["📂 Upload CSV", "📊 Live Market Data", "📥 Download Clean Data"])

# -------------------------------
# TAB 1: Upload CSV
# -------------------------------
with tab1:
    st.subheader("Upload your dataset")
    uploaded = st.file_uploader("Upload CSV file", type=["csv"])
    if uploaded:
        try:
            df = pd.read_csv(uploaded)
            st.success("File loaded successfully ✅")
            st.dataframe(df.head(), use_container_width=True)
            report = qa_report(df)

            c1, c2, c3, c4, c5 = st.columns(5)
            c1.metric("Rows", report["Rows"])
            c2.metric("Columns", report["Columns"])
            c3.metric("Missing Values", report["Missing Values"])
            c4.metric("Duplicate Rows", report["Duplicate Rows"])
            c5.metric("Negative Values", report["Negative Values"])

            st.markdown("#### Quick fixes")
            colA, colB, colC = st.columns(3)
            with colA:
                if st.button("Remove duplicate rows"):
                    df = df.drop_duplicates()
                    st.info("Duplicates removed.")
            with colB:
                if st.button("Fill missing numeric with 0"):
                    num_cols = df.select_dtypes(include=[np.number]).columns
                    df[num_cols] = df[num_cols].fillna(0)
                    st.info("Missing numeric values filled with 0.")
            with colC:
                if st.button("Clamp negatives to 0 (numeric)"):
                    num_cols = df.select_dtypes(include=[np.number]).columns
                    df[num_cols] = df[num_cols].clip(lower=0)
                    st.info("Negative numeric values clamped to 0.")

            st.markdown("#### Preview after fixes")
            st.dataframe(df.head(), use_container_width=True)
            download_csv(df, "cleaned_upload.csv")
        except Exception as e:
            st.error(f"Failed to read CSV: {e}")

# -------------------------------
# TAB 2: Live Market Data
# -------------------------------
with tab2:
    st.subheader("Fetch & Validate Live Crypto Data (Alpha Vantage)")
    symbol = st.text_input("Crypto symbol", "BTC")
    market = st.text_input("Market currency", "USD")
    if st.button("Fetch rate"):
        rate, err = fetch_alpha_vantage(symbol, market)
        if err:
            st.error(err)
        elif rate is None:
            st.error("No rate returned. Check symbol/market or API limits.")
        else:
            st.success("Live rate fetched ✅")
            col1, col2 = st.columns(2)
            col1.metric(f"{symbol}/{market}", f"{rate:.4f}")

            # Simple QA: non-negative and reasonable range
            is_valid = rate > 0 and rate < 1_000_000
            col2.metric("QA Valid", "Yes" if is_valid else "No")

            df_live = pd.DataFrame({"Symbol": [symbol], "Market": [market], "Rate": [rate]})
            st.dataframe(df_live, use_container_width=True)
            download_csv(df_live, "live_rate.csv")

# -------------------------------
# TAB 3: Download Clean Data (Sample)
# -------------------------------
with tab3:
    st.subheader("Demo cleaning on sample data")
    sample = pd.DataFrame({
        "id": [1, 2, 2, 3, 4],
        "value": [10, np.nan, 20, -5, 15],
        "note": ["ok", "missing", "ok", "neg", None],
    })
    st.write("Sample data (before):")
    st.dataframe(sample, use_container_width=True)

    cleaned = (
        sample
        .drop_duplicates()
        .assign(
            value=lambda d: d["value"].fillna(0).clip(lower=0),
            note=lambda d: d["note"].fillna(""),
        )
    )

    st.write("Cleaned data (after):")
    st.dataframe(cleaned, use_container_width=True)
    download_csv(cleaned, "cleaned_sample.csv")

# -------------------------------
# FOOTER
# -------------------------------
st.markdown("---")
st.caption("Made for event showcases • Streamlit on Hugging Face Spaces • Alpha Vantage integration")