import streamlit as st
import pandas as pd
import plotly.express as px
from datetime import datetime, timedelta
from simple_salesforce import Salesforce
from transformers import pipeline
from reportlab.lib.pagesizes import letter
from reportlab.platypus import SimpleDocTemplate, Table, TableStyle, Paragraph
from reportlab.lib import colors
from reportlab.lib.styles import getSampleStyleSheet
from utils import fetch_salesforce_data, detect_anomalies, generate_pdf_report

# Streamlit app configuration
st.set_page_config(page_title="LabOps Dashboard", layout="wide")

# Salesforce authentication (replace with your credentials)
sf = Salesforce(
    username=st.secrets["sf_username"],
    password=st.secrets["sf_password"],
    security_token=st.secrets["sf_security_token"]
)

# Initialize Hugging Face anomaly detection pipeline
anomaly_detector = pipeline("text-classification", model="bert-base-uncased", tokenizer="bert-base-uncased")

def main():
    st.title("Multi-Device LabOps Dashboard")
    
    # Filters
    col1, col2, col3 = st.columns(3)
    with col1:
        lab_site = st.selectbox("Select Lab Site", ["All", "Lab1", "Lab2", "Lab3"])
    with col2:
        equipment_type = st.selectbox("Equipment Type", ["All", "Cell Analyzer", "Weight Log", "UV Verification"])
    with col3:
        date_range = st.date_input("Date Range", [datetime.now() - timedelta(days=7), datetime.now()])
    
    # Fetch data from Salesforce
    query = f"""
        SELECT Equipment__c, Log_Timestamp__c, Status__c, Usage_Count__c
        FROM SmartLog__c
        WHERE Log_Timestamp__c >= {date_range[0].strftime('%Y-%m-%d')}
        AND Log_Timestamp__c <= {date_range[1].strftime('%Y-%m-%d')}
    """
    if lab_site != "All":
        query += f" AND Lab__c = '{lab_site}'"
    if equipment_type != "All":
        query += f" AND Equipment_Type__c = '{equipment_type}'"
    
    data = fetch_salesforce_data(sf, query)
    df = pd.DataFrame(data)
    
    if df.empty:
        st.warning("No data available for the selected filters.")
        return
    
    # Detect anomalies using Hugging Face
    df["Anomaly"] = df["Status__c"].apply(lambda x: detect_anomalies(x, anomaly_detector))
    
    # Device Cards
    st.subheader("Device Status")
    for equipment in df["Equipment__c"].unique():
        device_data = df[df["Equipment__c"] == equipment]
        latest_log = device_data.iloc[-1]
        anomaly = "⚠️ Anomaly" if latest_log["Anomaly"] == "POSITIVE" else "✅ Normal"
        st.markdown(f"""
            **{equipment}** | Health: {latest_log["Status__c"]} | Usage: {latest_log["Usage_Count__c"]} | Last Log: {latest_log["Log_Timestamp__c"]} | {anomaly}
        """)
    
    # Usage Chart
    st.subheader("Usage Trends")
    fig = px.line(df, x="Log_Timestamp__c", y="Usage_Count__c", color="Equipment__c", title="Daily Usage Trends")
    st.plotly_chart(fig, use_container_width=True)
    
    # Downtime Chart
    downtime_df = df[df["Status__c"] == "Down"]
    if not downtime_df.empty:
        fig_downtime = px.histogram(downtime_df, x="Log_Timestamp__c", color="Equipment__c", title="Downtime Patterns")
        st.plotly_chart(fig_downtime, use_container_width=True)
    
    # AMC Reminders
    st.subheader("AMC Reminders")
    amc_query = "SELECT Equipment__c, AMC_Expiry_Date__c FROM Equipment__c WHERE AMC_Expiry_Date__c <= NEXT_N_DAYS:14"
    amc_data = fetch_salesforce_data(sf, amc_query)
    for record in amc_data:
        st.write(f"Equipment {record['Equipment__c']} - AMC Expiry: {record['AMC_Expiry_Date__c']}")
    
    # Export PDF
    if st.button("Export PDF Report"):
        pdf_file = generate_pdf_report(df, lab_site, equipment_type, date_range)
        with open(pdf_file, "rb") as f:
            st.download_button("Download PDF", f, file_name="LabOps_Report.pdf")

if __name__ == "__main__":
    main()