paper-dynasty-card-creation/.claude/ops-rework/check_prod_missing_ratings.py

import asyncio
import aiohttp
import pandas as pd


AUTH_TOKEN = {'Authorization': f'Bearer Tp3aO3jhYve5NJF1IqOmJTmk'}
PROD_URL = 'https://pd.manticorum.com/api'


async def db_get_prod(endpoint: str, params: list = None, timeout: int = 30):
    """Get data from production API."""
    req_url = f'{PROD_URL}/v2/{endpoint}'

    if params:
        param_str = '&'.join([f'{k}={v}' for k, v in params])
        req_url += f'?{param_str}'

    async with aiohttp.ClientSession(headers=AUTH_TOKEN) as session:
        async with session.get(req_url, timeout=aiohttp.ClientTimeout(total=timeout)) as r:
            if r.status == 200:
                return await r.json()
            else:
                print(f"Error {r.status}: {await r.text()}")
                return None


async def main():
    """Check production database for missing ratings."""

    print("Checking PRODUCTION database")
    print("=" * 80)

    # Get cardset
    c_query = await db_get_prod('cardsets', params=[('name', '2025 Season')])
    if not c_query or c_query['count'] == 0:
        print("2025 Season cardset not found in production")
        return

    cardset_id = c_query['cardsets'][0]['id']
    print(f'Cardset: 2025 Season (ID: {cardset_id})\n')

    # Get batting cards
    bc_query = await db_get_prod('battingcards', params=[('cardset_id', cardset_id)])
    batting_cards = pd.DataFrame(bc_query['cards']) if bc_query and bc_query['count'] > 0 else pd.DataFrame()
    print(f"Total batting cards: {len(batting_cards)}")

    # Get batting ratings
    vl_query = await db_get_prod('battingcardratings', params=[
        ('cardset_id', cardset_id), ('vs_hand', 'L'),
        ('team_id', 31), ('ts', 's37136685556r6135248705')
    ], timeout=60)
    vr_query = await db_get_prod('battingcardratings', params=[
        ('cardset_id', cardset_id), ('vs_hand', 'R'),
        ('team_id', 31), ('ts', 's37136685556r6135248705')
    ], timeout=60)

    vl_ratings = pd.DataFrame(vl_query['ratings']) if vl_query and vl_query['count'] > 0 else pd.DataFrame()
    vr_ratings = pd.DataFrame(vr_query['ratings']) if vr_query and vr_query['count'] > 0 else pd.DataFrame()

    print(f"Total batting ratings (vL): {len(vl_ratings)}")
    print(f"Total batting ratings (vR): {len(vr_ratings)}")

    # Extract battingcard IDs
    if len(vl_ratings) > 0:
        if isinstance(vl_ratings['battingcard'].iloc[0], dict):
            vl_ratings['battingcard_id'] = vl_ratings['battingcard'].apply(lambda x: x['id'])
            vr_ratings['battingcard_id'] = vr_ratings['battingcard'].apply(lambda x: x['id'])
        else:
            vl_ratings['battingcard_id'] = vl_ratings['battingcard']
            vr_ratings['battingcard_id'] = vr_ratings['battingcard']

        # Find cards with both ratings
        cards_with_both = batting_cards[
            batting_cards['id'].isin(vl_ratings['battingcard_id']) &
            batting_cards['id'].isin(vr_ratings['battingcard_id'])
        ]
    else:
        cards_with_both = pd.DataFrame()

    cards_without_ratings = batting_cards[~batting_cards['id'].isin(cards_with_both['id'])] if len(cards_with_both) > 0 else batting_cards

    print(f"\nCards with BOTH ratings: {len(cards_with_both)}")
    print(f"Cards WITHOUT complete ratings: {len(cards_without_ratings)}")

    # Get player info for cards without ratings
    if len(cards_without_ratings) > 0:
        p_query = await db_get_prod('players', params=[('cardset_id', cardset_id)])
        players_df = pd.DataFrame(p_query['players']) if p_query else pd.DataFrame()

        # Extract player ID
        if isinstance(cards_without_ratings['player'].iloc[0], dict):
            cards_without_ratings['player_id_extract'] = cards_without_ratings['player'].apply(
                lambda x: x['player_id'] if isinstance(x, dict) else x
            )
        else:
            cards_without_ratings['player_id_extract'] = cards_without_ratings['player']

        cards_without = pd.merge(
            cards_without_ratings,
            players_df[['player_id', 'p_name', 'cost', 'description']],
            left_on='player_id_extract',
            right_on='player_id',
            how='left'
        )

        print(f"\nSample of players without ratings (first 20):")
        print(f"{'Player Name':<30} {'Card ID':<10} {'Cost':<10} {'Description'}")
        print("-" * 80)
        for _, row in cards_without.head(20).iterrows():
            name = row.get('p_name', 'Unknown')
            card_id = row.get('id', 'N/A')
            cost = row.get('cost', 'N/A')
            desc = row.get('description', 'N/A')
            print(f"{name:<30} {card_id:<10} {cost:<10} {desc}")

        # Count new vs existing
        new_players = cards_without[cards_without['cost'] == 99999]
        old_players = cards_without[cards_without['cost'] != 99999]

        print(f"\nNew players (cost=99999): {len(new_players)}")
        print(f"Existing players: {len(old_players)}")

        # Check rarity distribution for cards without ratings
        if 'rarity' in players_df.columns:
            players_without_ratings = players_df[players_df['player_id'].isin(cards_without['player_id'])]

            if isinstance(players_without_ratings['rarity'].iloc[0], dict):
                players_without_ratings['rarity_id'] = players_without_ratings['rarity'].apply(
                    lambda x: x['id'] if isinstance(x, dict) else x
                )
            else:
                players_without_ratings['rarity_id'] = players_without_ratings['rarity']

            rarity_counts = players_without_ratings['rarity_id'].value_counts().sort_index()

            rarity_names = {99: 'Hall of Fame', 1: 'Diamond', 2: 'Gold', 3: 'Silver', 4: 'Bronze', 5: 'Common'}

            print(f"\nRarity distribution for players WITHOUT ratings:")
            print(f"  {'Rarity':<15} {'Count':<10}")
            print("  " + "-" * 30)
            for rarity_id, count in rarity_counts.items():
                name = rarity_names.get(rarity_id, f'Unknown ({rarity_id})')
                print(f"  {name:<15} {count:<10}")

    # Check pitching cards too
    print("\n" + "=" * 80)
    print("Checking PITCHING cards:")
    print("=" * 80)

    pc_query = await db_get_prod('pitchingcards', params=[('cardset_id', cardset_id)])
    pitching_cards = pd.DataFrame(pc_query['cards']) if pc_query and pc_query['count'] > 0 else pd.DataFrame()
    print(f"\nTotal pitching cards: {len(pitching_cards)}")

    pvl_query = await db_get_prod('pitchingcardratings', params=[
        ('cardset_id', cardset_id), ('vs_hand', 'L'), ('short_output', True)
    ], timeout=60)
    pvr_query = await db_get_prod('pitchingcardratings', params=[
        ('cardset_id', cardset_id), ('vs_hand', 'R'), ('short_output', True)
    ], timeout=60)

    pvl_ratings = pd.DataFrame(pvl_query['ratings']) if pvl_query and pvl_query['count'] > 0 else pd.DataFrame()
    pvr_ratings = pd.DataFrame(pvr_query['ratings']) if pvr_query and pvr_query['count'] > 0 else pd.DataFrame()

    print(f"Total pitching ratings (vL): {len(pvl_ratings)}")
    print(f"Total pitching ratings (vR): {len(pvr_ratings)}")

    if len(pvl_ratings) > 0:
        # Check for missing pitching ratings
        pvl_ratings['pitchingcard_id'] = pvl_ratings['pitchingcard']
        pvr_ratings['pitchingcard_id'] = pvr_ratings['pitchingcard']

        pitchers_with_both = pitching_cards[
            pitching_cards['id'].isin(pvl_ratings['pitchingcard_id']) &
            pitching_cards['id'].isin(pvr_ratings['pitchingcard_id'])
        ]
        pitchers_without = pitching_cards[~pitching_cards['id'].isin(pitchers_with_both['id'])]

        print(f"\nPitchers with BOTH ratings: {len(pitchers_with_both)}")
        print(f"Pitchers WITHOUT complete ratings: {len(pitchers_without)}")


if __name__ == '__main__':
    asyncio.run(main())