#!/usr/bin/env python3
"""Parse Canalturf predictions and save to DB"""
import json
import sqlite3
from datetime import datetime

DB_PATH = "/home/h3r7/turf_scraper/turf.db"

# Load today's scraper output
with open('/home/h3r7/turf_scraper/v5_20260224_100313.json') as f:
    data = json.load(f)

# Find Canalturf quinté content
quinte_data = None
for page in data['pages']:
    if page['site'] == 'canalturf' and 'PRIX RAUBA CAPEU' in page['content']:
        quinte_data = page['content']
        break

if not quinte_data:
    print("No Quinté data found!")
    exit()

# Parse predictions from Canalturf
import re

# Extract horses from the content
horses = []

# Base (7 - I'M A BELIEVER)
base_match = re.search(r'Base\(s\)\s+(\d+)\s+([^\(]+)\s+\(([^)]+)\)', quinte_data)
if base_match:
    num, name, jockey = base_match.groups()
    horses.append((name.strip(), int(num), 90))  # 90% confidence for base

# Chance régulière
chance_matches = re.findall(r'Chance\(s\) régulière\(s\)\s+(\d+)\s+([^\(]+)\s+\(([^)]+)\)', quinte_data)
for num, name, jockey in chance_matches:
    horses.append((name.strip(), int(num), 70))

# Outsider
outsider_matches = re.findall(r'Outsider\(s\)\s+(\d+)\s+([^\(]+)\s+\(([^)]+)\)', quinte_data)
for num, name, jockey in outsider_matches:
    horses.append((name.strip(), int(num), 50))

print(f"Parsed {len(horses)} horses from Canalturf")

# Save to DB
conn = sqlite3.connect(DB_PATH)
c = conn.cursor()

today = "2026-02-24"

for name, num, conf in horses:
    c.execute("""
        INSERT INTO predictions (date, race_name, horse_number, horse_name, odds, prediction_rank, confidence, source)
        VALUES (?, ?, ?, ?, ?, ?, ?, ?)
    """, (today, "Quinté Cagnes-sur-Mer", num, name, 0, len(horses)+1, conf, "canalturf"))

conn.commit()
print(f"Saved {len(horses)} predictions!")

# Show what we saved
c.execute("SELECT horse_name, horse_number, confidence, source FROM predictions WHERE date = ?", (today,))
for row in c.fetchall():
    print(f"  {row[1]} - {row[0]} ({row[2]}%) - {row[3]}")

conn.close()