# pip install sec-api pandas python-dotenv
from sec_api import QueryApi
import pandas as pd
from smtplib import SMTP
from email.mime.text import MIMEText
from email.mime.multipart import MIMEMultipart
from dotenv import load_dotenv
import os
import uuid

# Load environment variables
load_dotenv()

# Configuration from .env
EMAIL_SENDER = os.getenv('EMAIL_SENDER')
EMAIL_RECIPIENT = os.getenv('EMAIL_RECIPIENT')
EMAIL_PASSWORD = os.getenv('EMAIL_PASSWORD')
SMTP_SERVER = os.getenv('SMTP_SERVER', 'smtp.gmail.com')
SMTP_PORT = int(os.getenv('SMTP_PORT', 587))
SEC_API_KEY = os.getenv('SEC_API_KEY')

if not SEC_API_KEY:
    raise ValueError("SEC_API_KEY not set in .env file")

# Initialize SEC API
query_api = QueryApi(api_key=SEC_API_KEY)

# Point72 CIK
CIK = '0001603466'

# Query for latest 13F-HR filings
query = {
    "query": f'cik:{CIK} formType:"13F-HR"',
    "sort": [{"filedAt": {"order": "desc"}}],
    "size": 10  # Fetch more to ensure we get distinct periods
}
response = query_api.get_filings(query)
filings = response.get('filings', [])

# Filter unique periodOfReport and select the two most recent
filings = sorted(filings, key=lambda x: x['periodOfReport'], reverse=True)
unique_filings = []
seen_periods = set()
for filing in filings:
    period = filing['periodOfReport']
    if period not in seen_periods:
        unique_filings.append(filing)
        seen_periods.add(period)
    if len(unique_filings) == 2:
        break

if len(unique_filings) < 2:
    raise Exception(f"Not enough unique filings found: {len(unique_filings)} found")

latest = unique_filings[0]
prev = unique_filings[1]

# Fetch holdings
latest_holdings = latest.get('holdings', [])
prev_holdings = prev.get('holdings', [])

if not latest_holdings or not prev_holdings:
    raise Exception("No holdings data found in filings")

# Convert holdings to DataFrame
latest_df = pd.DataFrame(latest_holdings)
prev_df = pd.DataFrame(prev_holdings)

# Extract share amount from shrsOrPrnAmt dictionary
def extract_shares(row):
    if isinstance(row, dict):
        return row.get('sshPrnamt', 0)
    return row

if 'shrsOrPrnAmt' in latest_df.columns:
    latest_df['shrsOrPrnAmt'] = latest_df['shrsOrPrnAmt'].apply(extract_shares)
    prev_df['shrsOrPrnAmt'] = prev_df['shrsOrPrnAmt'].apply(extract_shares)

# Verify required columns
required_cols = ['cusip', 'nameOfIssuer', 'shrsOrPrnAmt', 'value']
for col in required_cols:
    if col not in latest_df.columns or col not in prev_df.columns:
        raise KeyError(f"Column {col} missing in holdings data")

# Format value with commas and dollar sign
latest_df['value_formatted'] = latest_df['value'].apply(lambda x: f"${x:,.0f}")
prev_df['value_formatted'] = prev_df['value'].apply(lambda x: f"${x:,.0f}")

# Set index for comparison
key_col = 'cusip'
latest_df = latest_df.set_index(key_col)
prev_df = prev_df.set_index(key_col)

# Additions
additions = latest_df[~latest_df.index.isin(prev_df.index)]

# Removals
removals = prev_df[~prev_df.index.isin(latest_df.index)]

# Changes >10%
both = latest_df.index.intersection(prev_df.index)
changes = latest_df.loc[both].join(prev_df.loc[both], lsuffix='_new', rsuffix='_old')
changes['shrsOrPrnAmt_new'] = pd.to_numeric(changes['shrsOrPrnAmt_new'], errors='coerce')
changes['shrsOrPrnAmt_old'] = pd.to_numeric(changes['shrsOrPrnAmt_old'], errors='coerce')
changes['share_change'] = changes['shrsOrPrnAmt_new'] - changes['shrsOrPrnAmt_old']
changes = changes[changes['share_change'].notna()]
changes = changes[abs(changes['share_change']) / changes['shrsOrPrnAmt_old'].replace(0, 1) > 0.1]
changes['share_change'] = changes['share_change'].apply(lambda x: f"{x:,.0f}")

# HTML table styling
html_style = """
<style>
    table {
        border-collapse: collapse;
        width: 100%;
        font-family: Arial, sans-serif;
    }
    th, td {
        border: 1px solid #ddd;
        padding: 8px;
        text-align: left;
    }
    th {
        background-color: #f2f2f2;
        font-weight: bold;
    }
    tr:nth-child(even) {background-color: #f9f9f9;}
    tr:hover {background-color: #f5f5f5;}
</style>
"""

# Convert DataFrames to HTML tables
def df_to_html(df, columns, title):
    if df.empty:
        return f"<h3>{title} (0)</h3><p>None</p>"
    df_subset = df[columns].reset_index(drop=True)
    return f"<h3>{title} ({len(df)})</h3>{df_subset.to_html(index=False, border=0, classes='table')}"

# Summary as HTML
summary = f"""
<html>
<head>{html_style}</head>
<body>
<h2>Point72 13F Changes {prev['periodOfReport']} to {latest['periodOfReport']}</h2>
{df_to_html(additions, ['nameOfIssuer', 'shrsOrPrnAmt', 'value_formatted'], 'Additions')}
{df_to_html(removals, ['nameOfIssuer', 'shrsOrPrnAmt', 'value_formatted'], 'Removals')}
{df_to_html(changes, ['nameOfIssuer_new', 'share_change'], 'Changes')}
</body>
</html>
"""

# Email
msg = MIMEMultipart()
msg['From'] = EMAIL_SENDER
msg['To'] = EMAIL_RECIPIENT
msg['Subject'] = f'Point72 13F Update {latest["periodOfReport"]}'
msg.attach(MIMEText(summary, 'html'))

server = SMTP(SMTP_SERVER, SMTP_PORT)
server.starttls()
server.login(EMAIL_SENDER, EMAIL_PASSWORD)
server.send_message(msg)
server.quit()

print("Email sent!")