import warnings
warnings.filterwarnings('ignore')
import pandas as pd
import yfinance as yf
import numpy as np
import time
import datetime

# Download multi-ticker OHLCV data with yfinance
tickers = ['AAPL', 'MSFT', 'GOOGL', 'AMZN', 'TSLA']
ohlcv = yf.download(tickers, period='1y', auto_adjust=True, progress=False)
ohlcv = ohlcv.stack(future_stack=True).rename_axis(['Date', 'Ticker']).reset_index()
ohlcv.columns.name = None
print('Shape of dataset:', ohlcv.shape)
print(ohlcv.head(3))

def fetch_latest_ohlcv():
    tickers = ['AAPL', 'MSFT', 'GOOGL', 'AMZN', 'TSLA']
    df = yf.download(tickers, period='5d', auto_adjust=True, progress=False)
    df = df.stack(future_stack=True).rename_axis(['Date', 'Ticker']).reset_index()
    df.columns.name = None
    return df

print('Fetching latest 5 days of OHLCV data:')
temp_data = fetch_latest_ohlcv()
print(temp_data.head())

# Example: Manual refresh using datetime
now = datetime.datetime.now()
print('Current date and time:', now.strftime('%Y-%m-%d %H:%M:%S'))

# Beginner: Always update on demand (no automation)
latest_data = fetch_latest_ohlcv()
print('Manually updated! Latest shape:', latest_data.shape)

# Beginner: Save latest OHLCV data to CSV with a timestamp
timestamp = datetime.datetime.now().strftime('%Y%m%d_%H%M%S')
csv_filename = f'ohlcv_latest_{timestamp}.csv'
latest_data.to_csv(csv_filename, index=False)
print('Data saved to', csv_filename)

# Beginner: Refresh every minute (simulate with loop and time.sleep)
for i in range(2):
    print(f'Refresh {i+1}:')
    df = fetch_latest_ohlcv()
    print(df.head(1))
    time.sleep(2)  # Wait for two seconds (in real scheduling, use 60)
print('Simulated periodic update finished.')

# Intermediate: Save only today's new rows if not already present
latest_day = latest_data['Date'].max()
print('Most recent date in our dataset:', latest_day)
existing = pd.read_csv(csv_filename)
if latest_day in existing['Date'].values:
    print('Today is already in fileno new rows needed.')
else:
    combined = pd.concat([existing, latest_data[latest_data['Date'] == latest_day]], ignore_index=True)
    combined.to_csv(csv_filename, index=False)
    print('Added todays update to', csv_filename)

# Intermediate: Schedule updates by checking market close time
now = datetime.datetime.now()
if now.hour >= 16:
    print('Market is closed! Downloading fresh end-of-day prices...')
    df = fetch_latest_ohlcv()
    print(df.tail(3))
else:
    print('Market is open or not yet closed. Wait before final update.')

# Intermediate: Random seed for reproducibility
np.random.seed(42)
random_value = np.random.randint(1, 100)
print('Random number generated with seed 42:', random_value)

# Advanced: Modular update function with simple error logging
def safe_update(tickers, outfile):
    try:
        print('Updating...')
        start = time.time()
        df = yf.download(tickers, period='5d', auto_adjust=True, progress=False)
        df = df.stack(future_stack=True).rename_axis(['Date', 'Ticker']).reset_index()
        df.columns.name = None
        df.to_csv(outfile, index=False)
        elapsed = time.time() - start
        print('Update successful! Elapsed seconds:', round(elapsed, 2))
        return True
    except Exception as ex:
        print('Update failed:', str(ex))
        return False

success = safe_update(['AAPL', 'MSFT', 'GOOGL'], 'update_demo.csv')

# Advanced: Detect and handle missing data issues
def check_missing_dates(df):
    all_days = pd.date_range(df['Date'].min(), df['Date'].max())
    missing = set(all_days.date) - set(pd.to_datetime(df['Date']).dt.date)
    if missing:
        print('Missing trading days detected:', missing)
    else:
        print('No missing trading days.')

check_missing_dates(latest_data)

# Advanced: Combine update and gap-detection for robust pipelines
def reliable_ohlcv_update(tickers, outfile):
    updated = safe_update(tickers, outfile)
    if updated:
        df = pd.read_csv(outfile)
        if 'Date' in df.columns:
            check_missing_dates(df)
        else:
            print('Warning: Output file missing Date column!')
    else:
        print('Update failedno data checked.')

reliable_ohlcv_update(['AAPL', 'MSFT'], 'smart_ohlcv.csv')

# Error Handling: Catch network or API failures during update
def fetch_with_retry(tickers, max_attempts=3):
    for attempt in range(1, max_attempts + 1):
        try:
            print(f'Attempt {attempt} to fetch data...')
            df = yf.download(tickers, period='5d', auto_adjust=True, progress=False)
            df = df.stack(future_stack=True).rename_axis(['Date', 'Ticker']).reset_index()
            df.columns.name = None
            print('Success!')
            return df
        except Exception as e:
            print('Error:', e)
            if attempt == max_attempts:
                print('All attempts failed. Returning empty DataFrame.')
                return pd.DataFrame()
            time.sleep(2)

df_retry = fetch_with_retry(['AAPL', 'MSFT'])

# Best Practice: Dynamic filename and logging after update
def datestamp_filename(basename):
    ts = datetime.datetime.now().strftime('%Y%m%d_%H%M%S')
    return f'{basename}_{ts}.csv'

fname = datestamp_filename('scheduled_ohlcv')
result = safe_update(['AAPL'], fname)
print('Saved as:', fname, '| Success:', result)

# End-to-End Demo: Automated update and verification
def end_to_end_update_and_check():
    filename = datestamp_filename('end2end_ohlcv')
    print('[1] Automated update for MSFT and TSLA...')
    ok = safe_update(['MSFT', 'TSLA'], filename)
    if ok:
        df = pd.read_csv(filename)
        print('[2] Checking file:', filename)
        check_missing_dates(df)
        print('All steps completed!')
    else:
        print('Update failedskipping validation.')
    return filename

final_file = end_to_end_update_and_check()