data/scores.py

from __future__ import annotations

from datetime import datetime
from typing import Any

import pandas as pd
import requests

HEADERS = {
    "User-Agent": "Mozilla/5.0",
    "Accept-Language": "en-US,en;q=0.9",
}

SCORES_API_URL = "https://statsapi.mlb.com/api/v1/schedule"

SPORT_IDS = [1]  # MLB only

TEAM_NORMALIZATION = {
    "Chinese Taipei": "Chinese Taipei",
    "Czech Republic": "Czechia",
    "South Korea": "Korea",
    "USA": "United States",
    "U.S.A.": "United States",
}


def _normalize_team_name(name: Any) -> str:
    text = str(name or "").strip()
    if not text:
        return ""
    return TEAM_NORMALIZATION.get(text, text)


def _normalize_status(abstract_state: str, detailed_state: str, inning_state: str, current_inning: Any) -> str:
    abstract = str(abstract_state or "").strip().lower()
    detailed = str(detailed_state or "").strip()
    inning = str(inning_state or "").strip()
    inning_num = current_inning

    if abstract == "final":
        return "Final"

    if abstract == "live":
        if inning and inning_num not in (None, ""):
            return f"{inning} {inning_num}"
        return detailed or "Live"

    if abstract == "preview":
        return "Scheduled"

    if str(detailed).strip():
        return str(detailed).strip()

    return ""


def _safe_int(value: Any) -> int | None:
    try:
        if value is None:
            return None
        text = str(value).strip().lower()
        if text in {"", "nan", "none"}:
            return None
        return int(float(value))
    except Exception:
        return None


def _fetch_scores_for_sport_id(date_str: str, sport_id: int) -> pd.DataFrame:
    params = {
        "sportId": sport_id,
        "date": date_str,
        "hydrate": "linescore,broadcasts",
    }

    response = requests.get(SCORES_API_URL, headers=HEADERS, params=params, timeout=30)
    response.raise_for_status()
    payload = response.json()

    dates = payload.get("dates", []) or []
    rows: list[dict[str, Any]] = []

    for date_block in dates:
        games = date_block.get("games", []) or []

        for game in games:
            game_pk = game.get("gamePk")

            teams = game.get("teams", {}) or {}
            away = teams.get("away", {}) or {}
            home = teams.get("home", {}) or {}

            away_team_info = (away.get("team", {}) or {})
            home_team_info = (home.get("team", {}) or {})

            away_team = _normalize_team_name(away_team_info.get("name"))
            home_team = _normalize_team_name(home_team_info.get("name"))

            status_info = game.get("status", {}) or {}
            abstract_state = status_info.get("abstractGameState", "")
            detailed_state = status_info.get("detailedState", "")

            linescore = game.get("linescore", {}) or {}
            inning_state = linescore.get("inningState", "")
            current_inning = linescore.get("currentInning")

            status = _normalize_status(
                abstract_state=abstract_state,
                detailed_state=detailed_state,
                inning_state=inning_state,
                current_inning=current_inning,
            )

            away_score = _safe_int(away.get("score"))
            home_score = _safe_int(home.get("score"))

            linescore_teams = linescore.get("teams", {}) or {}
            away_ls = linescore_teams.get("away", {}) or {}
            home_ls = linescore_teams.get("home", {}) or {}

            if away_score is None:
                away_score = _safe_int(away_ls.get("runs"))
            if home_score is None:
                home_score = _safe_int(home_ls.get("runs"))

            away_hits = _safe_int(away_ls.get("hits"))
            home_hits = _safe_int(home_ls.get("hits"))
            away_errors = _safe_int(away_ls.get("errors"))
            home_errors = _safe_int(home_ls.get("errors"))

            game_datetime = game.get("gameDate", "")
            start_time_et = ""
            if game_datetime:
                try:
                    ts = pd.to_datetime(game_datetime, utc=True).tz_convert("America/New_York")
                    start_time_et = ts.strftime("%-I:%M %p ET")
                except Exception:
                    start_time_et = ""

            broadcasts = game.get("broadcasts", []) or []
            tv = ""
            if broadcasts:
                names = []
                for b in broadcasts:
                    name = str((b.get("name") or "")).strip()
                    if name and name not in names:
                        names.append(name)
                tv = ", ".join(names)

            if away_team and home_team:
                rows.append(
                    {
                        "score_fetch_time": datetime.utcnow(),
                        "game_date": date_str,
                        "game_pk": str(game_pk) if game_pk is not None else "",
                        "away_team": away_team,
                        "home_team": home_team,
                        "away_score": away_score,
                        "home_score": home_score,
                        "away_hits": away_hits,
                        "home_hits": home_hits,
                        "away_errors": away_errors,
                        "home_errors": home_errors,
                        "status": status,
                        "start_time_et": start_time_et,
                        "tv": tv,
                        "sport_id": sport_id,
                    }
                )

    if not rows:
        return pd.DataFrame()

    df = pd.DataFrame(rows)
    df = df.drop_duplicates(subset=["game_pk", "away_team", "home_team", "status"], keep="last")
    return df.reset_index(drop=True)


def fetch_scores_for_date(date_str: str) -> pd.DataFrame:
    parts: list[pd.DataFrame] = []

    for sport_id in SPORT_IDS:
        try:
            sport_df = _fetch_scores_for_sport_id(date_str, sport_id)
            if sport_df is not None and not sport_df.empty:
                parts.append(sport_df)
        except Exception:
            pass

    if not parts:
        return pd.DataFrame(
            columns=[
                "score_fetch_time",
                "game_date",
                "game_pk",
                "away_team",
                "home_team",
                "away_score",
                "home_score",
                "away_hits",
                "home_hits",
                "away_errors",
                "home_errors",
                "status",
                "start_time_et",
                "tv",
                "sport_id",
            ]
        )

    df = pd.concat(parts, ignore_index=True)
    df = df.drop_duplicates(subset=["game_pk", "away_team", "home_team"], keep="last")
    return df.reset_index(drop=True)