from flask import Flask, session, g, render_template_string
from flask_session import Session
from flask_oidc import OpenIDConnect
from werkzeug.middleware.proxy_fix import ProxyFix
from loguru import logger
import redis
import json
import os
import plotly.express as px
import pandas as pd
import psycopg
import sqlalchemy
import time

try:
  redis_url = os.environ['REDIS_URL']
  oidc_client_secrets = os.environ['OIDC_CLIENT_SECRETS']
  secret_key = os.environ['SECRET_KEY']
except KeyError as e:
  logger.error(f"Required environment variable not set ({e})")
  raise e
  

app = Flask(__name__)

app.config.update({
  'SECRET_KEY': secret_key,
  'SESSION_TYPE': 'redis',
  'SESSION_REDIS': redis.from_url(redis_url),
  'OIDC_CLIENT_SECRETS': json.loads(oidc_client_secrets),
  'OIDC_SCOPES': 'openid email',
  'OIDC_USER_INFO_ENABLED': True,
  'SESSION_USE_SIGNER': True,
})
Session(app)
oidc = OpenIDConnect(app)

@app.route('/token_debug', methods=['GET'])
@oidc.require_login
def token_debug():
    # Access Token vom Identity Provider abrufen
    access_token = oidc.get_access_token()
    return json.dumps({
        "access_token": access_token
    })

@app.route('/')
@oidc.require_login
def index():
    try:
        stepX_time = time.time()
        dbh = psycopg.connect()
        engine = sqlalchemy.create_engine("postgresql+psycopg://", creator=lambda: dbh)

        step0_time = time.time()
        df = pd.read_sql("SELECT month, cast(year AS varchar), current_energy AS value FROM pv_energy_by_month", con=engine)
        step1_time = time.time()
        duration1 = step1_time - step0_time
        logger.info(f"{duration1=}")
        fig_1 = px.bar(df, x='month', y='value', color='year', barmode='group')
        step2_time = time.time()
        duration2 = step2_time - step1_time
        logger.info(f"{duration2=}")
        fig_1.update_layout(
            title=f"Jahreswerte Exportierte Energie {duration1:.3f}, {duration2:.3f}",
            xaxis_title="",
            yaxis_title="",
            legend_title="Jahr",
            xaxis=dict(
                tickmode="array",
                tickvals=list(range(1, 13)),  # Monate 1–12
                ticktext=["Jan", "Feb", "Mär", "Apr", "Mai", "Jun", "Jul", "Aug", "Sep", "Okt", "Nov", "Dez"]
            ),
            yaxis=dict(ticksuffix=" kWh")
        )
        graph_html_1 = fig_1.to_html(full_html=False, default_height='30%')

        step3_time = time.time()
        df = pd.read_sql("SELECT time_bucket('5 minutes', time) AS bucket, AVG(power) AS avg_power FROM pv_power_v WHERE time >= date_trunc('day', now()) - '1 day'::interval AND time < date_trunc('day', now()) GROUP BY bucket ORDER BY bucket", con=engine)
        step4_time = time.time()
        duration3 = step4_time - step3_time
        logger.info(f"{duration3=}")
        fig_2 = px.line(df, x='bucket', y='avg_power')
        step5_time = time.time()
        duration4 = step5_time - step4_time
        logger.info(f"{duration4=}")
        fig_2.update_layout(
            xaxis_title="",
            yaxis_title="",
            title=f"Export gestern {duration3:.3f}, {duration4:.3f}",
            yaxis=dict(ticksuffix=" W")
        )
        graph_html_2 = fig_2.to_html(full_html=False, default_height='30%')

        step6_time = time.time()
        df = pd.read_sql("SELECT time_bucket('5 minutes', time) AS bucket, AVG(power) AS avg_power FROM pv_power_v WHERE time >= date_trunc('day', now()) AND time < date_trunc('day', now()) + '1 day'::interval GROUP BY bucket ORDER BY bucket", con=engine)
        step7_time = time.time()
        duration5 = step7_time - step6_time
        logger.info(f"{duration5=}")
        fig_3 = px.line(df, x='bucket', y='avg_power')
        step8_time = time.time()
        duration6 = step8_time - step7_time
        logger.info(f"{duration6=}")
        fig_3.update_layout(
            xaxis_title="",
            yaxis_title="",
            title=f"Export heute {duration5:.3f}, {duration6:.3f}",
            yaxis=dict(ticksuffix=" W")
        )
        graph_html_3 = fig_3.to_html(full_html=False, default_height='30%')
        stepZ_time = time.time()
        duration7 = stepZ_time - stepX_time
        logger.info(f"{duration7=}")

        return render_template_string(f"""
            <html>
                <head>
                    <title>Jahreswerte PV-Energie</title>
                </head>
                <body>
                    {graph_html_1}
                    {graph_html_2}
                    {graph_html_3}
                    <div style="height:9vh; background-color:lightgrey; font-family: Courier, Consolas, monospace;">
                      <table style="border-collapse: collapse;">
                        <style>table td {{ padding-right: 100px }}</style>
                        <tr>
                          <td>Query 1: {duration1:.3f} s</td><td>Graph 1: {duration2:.3f} s</td>
                        </tr><tr>
                          <td>Query 2: {duration3:.3f} s</td><td>Graph 2: {duration4:.3f} s</td>
                        </tr><tr>
                          <td>Query 3: {duration5:.3f} s</td><td>Graph 3: {duration6:.3f} s</td>
                        </tr><tr>
                          <td>Total: {duration7:.3f} s</td><td></td>
                        </tr>
                      </table>   
                    </div>
                </body>
            </html>
        """)
    except Exception as e:
        raise Exception(f"Error when querying energy export values: {e}")
    finally:
        if dbh is not None:
            dbh.close()




if __name__ == '__main__':
    app.run(port=8080)
else:
    exposed_app = ProxyFix(app, x_for=1, x_host=1)