from __future__ import annotations

import re
import unicodedata


NAME_ALIASES = {
    # ---- Batters ----
    "shohei ohtani": "shohei ohtani",
    "shōhei ohtani": "shohei ohtani",
    "seiya suzuki": "seiya suzuki",
    "masataka yoshida": "masataka yoshida",
    "mookie betts": "mookie betts",
    "fernando tatis jr": "fernando tatis jr",
    "fernando tatis jr.": "fernando tatis jr",
    "vladimir guerrero jr": "vladimir guerrero jr",
    "vladimir guerrero jr.": "vladimir guerrero jr",
    "ronald acuña jr": "ronald acuna jr",
    "ronald acuna jr": "ronald acuna jr",
    "ronald acuña jr.": "ronald acuna jr",
    "ronald acuna jr.": "ronald acuna jr",
    "juan soto": "juan soto",
    "julio rodriguez": "julio rodriguez",
    "julio rodríguez": "julio rodriguez",
    "jose ramirez": "jose ramirez",
    "josé ramírez": "jose ramirez",
    "yordan alvarez": "yordan alvarez",
    "yordan álvarez": "yordan alvarez",
    "luis robert jr": "luis robert jr",
    "luis robert jr.": "luis robert jr",
    "bo bichette": "bo bichette",
    "manny machado": "manny machado",
    "xander bogaerts": "xander bogaerts",
    "rafael devers": "rafael devers",
    "ketel marte": "ketel marte",
    "isaac paredes": "isaac paredes",
    "andy pages": "andy pages",
    # ---- Pitchers ----
    # Jr./Sr. variants — canonical form keeps suffix when that is how statcast stores the name
    "nestor cortes jr": "nestor cortes jr",
    "nestor cortes": "nestor cortes jr",           # odds APIs sometimes omit Jr
    "néstor cortés jr": "nestor cortes jr",
    "néstor cortés": "nestor cortes jr",
    # International names with diacritics commonly mis-encoded by odds sources
    "framber valdéz": "framber valdez",
    "framber valdez": "framber valdez",
    "sandy alcántara": "sandy alcantara",
    "sandy alcantara": "sandy alcantara",
    "pablo lópez": "pablo lopez",
    "pablo lopez": "pablo lopez",
    "ranger suárez": "ranger suarez",
    "ranger suarez": "ranger suarez",
    "josé berríos": "jose berrios",
    "jose berrios": "jose berrios",
    "josé quintana": "jose quintana",
    "jose quintana": "jose quintana",
    "martín pérez": "martin perez",
    "martin perez": "martin perez",
    "eduardo rodríguez": "eduardo rodriguez",
    "eduardo rodriguez": "eduardo rodriguez",
    "cristopher sánchez": "cristopher sanchez",
    "cristopher sanchez": "cristopher sanchez",
    "adrián houser": "adrian houser",
    "adrian houser": "adrian houser",
    "julio urías": "julio urias",
    "julio urias": "julio urias",
    "yonny chirinos": "yonny chirinos",
    "yusei kikuchi": "yusei kikuchi",
    "yoshinobu yamamoto": "yoshinobu yamamoto",
    "kodai senga": "kodai senga",
    "shōta imanaga": "shota imanaga",
    "shota imanaga": "shota imanaga",
}


def _strip_accents(text: str) -> str:
    normalized = unicodedata.normalize("NFKD", text)
    return "".join(ch for ch in normalized if not unicodedata.combining(ch))


def normalize_player_name(name: str) -> str:
    text = str(name or "").strip().lower()
    text = _strip_accents(text)
    text = text.replace("’", "'")
    text = text.replace(".", "")
    text = re.sub(r"\s+", " ", text)

    # Remove common suffix punctuation inconsistencies
    text = text.replace(" jr ", " jr ")
    text = text.replace(" sr ", " sr ")

    return NAME_ALIASES.get(text, text)


def normalize_pitcher_name(name: str) -> str:
    """Canonical normalizer for pitcher names.

    Identical pipeline to normalize_player_name() — strips accents, removes
    punctuation, collapses whitespace, applies NAME_ALIASES — exposed as a
    separate entrypoint so pitcher-side imports are unambiguous and the alias
    table can be extended with pitcher-specific entries without touching the
    batter path.
    """
    return normalize_player_name(name)


def map_odds_name_to_model_name(name: str) -> str:
    return normalize_player_name(name)