"""Classify sounds into group / category and infer characters from Wwise names.

Scheme (documented in docs/categories.md):
  group     Speech | Music | Ambience | SFX | Unknown
  category  finer bucket derived from the name prefix (e.g. "Interview tapes", "Combat barks")
Characters: the authoritative speaker comes from the curated catalog (data/catalog/characters.csv,
built from the user's per-character lists). For other lines `infer_characters` returns the
characters/roles that the *name* mentions -- a hint, not a speaker label (conversations alternate
speakers between variations).
"""
import re

# (regex on the upper-case name, group, category) -- first match wins
_RULES = [
    (r"^(CHR_RADIOSCANNER|PSO_CARRADIO|SPEECH_RADIO|SPEECH_TV|SPEECH_CAMERAS|DIA_)", "Speech", "Radio & TV"),
    (r"^SPEECH_TAPE_", "Speech", "Interview tapes"),
    (r"^SPEECH_INTERROGATION_", "Speech", "Interrogations"),
    (r"^SPEECH_(CINEMATIC|NEW)_", "Speech", "Cinematics"),
    (r"^SPEECH_BOSS_", "Speech", "Boss fights"),
    (r"^SPEECH_RIDDLE(R|S)_", "Speech", "Riddler"),
    (r"^SPEECH_SIDE_", "Speech", "Side missions"),
    (r"^SPEECH_DLC_", "Speech", "DLC"),
    (r"^SPEECH_PRED_", "Speech", "Predator chatter"),
    (r"^SPEECH_(COMBAT|BULLY)_", "Speech", "Combat barks"),
    (r"^SPEECH_EMOTE_", "Speech", "Efforts & emotes"),
    (r"^SPEECH_REACT_", "Speech", "Reactions"),
    (r"^SPEECH_(STREETS|ZONES|OUTCAST|VERTICAL)_", "Speech", "Open-world chatter"),
    (r"^SPEECH_CHAR_", "Speech", "Character lines"),
    (r"^SPEECH_(CH|SEQUENCE|STEEL|MUSEUM|CHURCH|COURT|UNDER|GCPD|ZP1)_", "Speech", "Story missions"),
    (r"^SPEECH_(GAME|MAP|AUGMENTED|FORENSICS|TEMP)_", "Speech", "Gameplay & hints"),
    (r"^NPC_", "Ambience", "Crowds"),
    (r"^SPEECH_", "Speech", "Other speech"),
    (r"^MUSIC_PDLC", "Music", "Music (DLC)"),
    (r"^MUSIC", "Music", "Music"),
    (r"^(AMB|PSA)_", "Ambience", "Ambience"),
    (r"^WND_", "Ambience", "Wind"),
    (r"^P?VOL_", "Ambience", "Room tone"),
    (r"^SEQ_", "SFX", "Scripted sequences"),
    (r"^OBJ_", "SFX", "Objects"),
    (r"^WPN_", "SFX", "Weapons"),
    (r"^CHR_", "SFX", "Character foley"),
    (r"^GLOBAL", "SFX", "Global"),
]
_RULES = [(re.compile(p), g, c) for p, g, c in _RULES]

GROUPS = ("Speech", "Music", "Ambience", "SFX", "Unknown")


def classify(name: str, pack: str = None):
    """-> (group, category). `pack` = name of the stream pack a media belongs to, if any."""
    n = name.upper()
    if pack:
        p = pack.upper()
        if p.startswith("MUSIC"):
            return "Music", "Music (DLC)" if "PDLC" in p else "Music"
    for rx, g, c in _RULES:
        if rx.search(n):
            return g, c
    return "Unknown", "Unknown"


# --- characters -------------------------------------------------------------------------------
# display names for the labels used in the user's catalog files (data/catalog/source/ita/*.txt)
USER_LABELS = {
    "Alfred": "Alfred", "Azrael": "Azrael", "Bane": "Bane", "Batman": "Batman",
    "Calendar": "Calendar Man", "Catwoman": "Catwoman", "Croc": "Killer Croc", "Deadshot": "Deadshot",
    "Freeze": "Mr. Freeze", "Gordon": "Gordon", "Ivy": "Poison Ivy", "Joker": "Joker",
    "Madhatter": "Mad Hatter", "Oracle": "Oracle", "Penguin": "Penguin", "Quinn": "Harley Quinn",
    "Ras": "Ra's al Ghul", "Riddler": "Riddler", "Robin": "Robin", "Strange": "Hugo Strange",
    "Talia": "Talia al Ghul", "Twoface": "Two-Face", "Unknown": "Unknown", "Vicky": "Vicki Vale",
    "Zsazz": "Victor Zsasz",
}

# name tokens -> character / role  (matched on whole "_"-separated tokens, digits ignored)
_TOKENS = {
    "Batman": ("BATMAN", "BM"), "Catwoman": ("CATWOMAN", "CAT", "CW"), "Joker": ("JOKER",),
    "Penguin": ("PENGUIN", "PENG"), "Two-Face": ("TWOFACE", "TF"), "Riddler": ("RIDDLER",),
    "Mr. Freeze": ("FREEZE", "MRFREEZE", "FREEZESUIT"), "Poison Ivy": ("IVY",),
    "Hugo Strange": ("STRANGE", "HUGO"), "Robin": ("ROBIN",), "Oracle": ("ORACLE",),
    "Alfred": ("ALFRED", "ALF"), "Harley Quinn": ("QUINN", "HARLEY"), "Ra's al Ghul": ("RAS",),
    "Talia al Ghul": ("TALIA",), "Victor Zsasz": ("ZSASZ", "ZSAZZ"), "Bane": ("BANE",),
    "Deadshot": ("DEADSHOT",), "Mad Hatter": ("MADHATTER", "HATTER"), "Calendar Man": ("CALENDAR", "CM"),
    "Hush": ("HUSH",), "Clayface": ("CLAYFACE", "CLAY"), "Solomon Grundy": ("GRUNDY",),
    "Azrael": ("AZRAEL",), "Vicki Vale": ("VALE", "VICKI", "VICKY", "VIC"), "Gordon": ("GORDON",),
    "Nightwing": ("NIGHTWING", "NW"), "Killer Croc": ("CROC",),
    # roles
    "Thug": ("THUG", "THUGS", "GENTHUG", "JTHUG", "JTHUGS", "PTHUG", "PTHUGS", "BULLY"),
    "TYGER guard": ("TYGER",), "Ninja": ("NINJA", "ASSASSIN"), "Doctor": ("DOC", "DOCTOR"),
    "Cop": ("COP", "COPS", "GCPD"), "Political prisoner": ("HOBO", "OUTCAST", "OC"),
    "Hostage": ("HOST", "HOSTAGE"), "Guard": ("GUARD", "GUARDS"),
}
_TOK2CHAR = {t: c for c, toks in _TOKENS.items() for t in toks}
_STRIP_DIGITS = re.compile(r"\d+$")


def infer_characters(name: str):
    """Characters/roles mentioned by a sound name (ordered, unique). Empty if none."""
    n = name.upper()
    out = []
    if n.startswith("SPEECH_TAPE_"):
        out.append("Hugo Strange")  # Strange's interview tapes
    for tok in re.split(r"[_\s]+", n):
        t = _STRIP_DIGITS.sub("", tok)
        c = _TOK2CHAR.get(t) or _TOK2CHAR.get(tok)
        if c and c not in out:
            out.append(c)
    return out