"""
Custom Chess Tokenizer for the Chess Challenge.

This tokenizer treats each move as a single token using the extended UCI notation
from the Lichess dataset (e.g., WPe2e4, BNg8f6).

The dataset format uses:
- W/B prefix for White/Black
- Piece letter: P=Pawn, N=Knight, B=Bishop, R=Rook, Q=Queen, K=King
- Source and destination squares (e.g., e2e4)
- Special suffixes: (x)=capture, (+)=check, (+*)=checkmate, (o)/(O)=castling
"""

from __future__ import annotations

import json
import os
from pathlib import Path
from typing import Dict, List, Optional

from transformers import PreTrainedTokenizer


class ChessTokenizer_v0(PreTrainedTokenizer):
    """
    A custom tokenizer for chess moves using extended UCI notation.
    
    This tokenizer maps each possible chess move to a unique token ID.
    The vocabulary is built from the training dataset to ensure all moves
    encountered during training have a corresponding token.
    
    Example:
        >>> tokenizer = ChessTokenizer_v0()
        >>> tokenizer.encode("WPe2e4 BPe7e5")
        [1, 42, 87, 2]  # [BOS, e2e4, e7e5, EOS]
    """
    
    model_input_names = ["input_ids", "attention_mask"]
    vocab_files_names = {"vocab_file": "vocab.json"}
    
    # Special tokens
    PAD_TOKEN = "[PAD]"
    BOS_TOKEN = "[BOS]"
    EOS_TOKEN = "[EOS]"
    UNK_TOKEN = "[UNK]"
    
    def __init__(
        self,
        vocab_file: Optional[str] = None,
        vocab: Optional[Dict[str, int]] = None,
        **kwargs,
    ):
        """
        Initialize the chess tokenizer.
        
        Args:
            vocab_file: Path to a JSON file containing the vocabulary mapping.
            vocab: Dictionary mapping tokens to IDs (alternative to vocab_file).
            **kwargs: Additional arguments passed to PreTrainedTokenizer.
        """
        # Initialize special tokens
        self._pad_token = self.PAD_TOKEN
        self._bos_token = self.BOS_TOKEN
        self._eos_token = self.EOS_TOKEN
        self._unk_token = self.UNK_TOKEN

        # Remove any duplicate special-token entries passed through kwargs
        # to avoid "multiple values for keyword" errors when loading from disk.
        kwargs.pop("pad_token", None)
        kwargs.pop("bos_token", None)
        kwargs.pop("eos_token", None)
        kwargs.pop("unk_token", None)
        
        # Load or create vocabulary
        if vocab is not None:
            self._vocab = vocab
        elif vocab_file is not None and os.path.exists(vocab_file):
            with open(vocab_file, "r", encoding="utf-8") as f:
                self._vocab = json.load(f)
        else:
            # Create a minimal vocabulary with just special tokens
            # The full vocabulary should be built from the dataset
            self._vocab = self._create_default_vocab()
        
        # Create reverse mapping
        self._ids_to_tokens = {v: k for k, v in self._vocab.items()}
        
        # Call parent init AFTER setting up vocab
        super().__init__(
            pad_token=self._pad_token,
            bos_token=self._bos_token,
            eos_token=self._eos_token,
            unk_token=self._unk_token,
            **kwargs,
        )
    
    def _create_default_vocab(self) -> Dict[str, int]:
        """
        Create a minimal default vocabulary with just special tokens.
        
        For the full vocabulary, use `build_vocab_from_dataset()`.
        This minimal vocab is just a placeholder - you should build from data.
        """
        special_tokens = [self.PAD_TOKEN, self.BOS_TOKEN, self.EOS_TOKEN, self.UNK_TOKEN]
        vocab = {token: idx for idx, token in enumerate(special_tokens)}
        return vocab
    
    @classmethod
    def build_vocab_from_iterator(
        cls,
        iterator,
        min_frequency: int = 1,
    ) -> "ChessTokenizer_v0":
        """
        Build a tokenizer vocabulary from an iterator of game strings.
        
        Args:
            iterator: An iterator yielding game strings (space-separated moves).
            min_frequency: Minimum frequency for a token to be included.
        
        Returns:
            A ChessTokenizer_v0 with the built vocabulary.
        """
        from collections import Counter
        
        token_counts = Counter()
        
        for game in iterator:
            moves = game.strip().split()
            token_counts.update(moves)
        
        # Filter by frequency
        tokens = [
            token for token, count in token_counts.items()
            if count >= min_frequency
        ]
        
        # Sort for reproducibility
        tokens = sorted(tokens)
        
        # Build vocabulary
        special_tokens = [cls.PAD_TOKEN, cls.BOS_TOKEN, cls.EOS_TOKEN, cls.UNK_TOKEN]
        vocab = {token: idx for idx, token in enumerate(special_tokens + tokens)}
        
        return cls(vocab=vocab)
    
    @classmethod
    def build_vocab_from_dataset(
        cls,
        dataset_name: str = "dlouapre/lichess_2025-01_1M",
        split: str = "train",
        column: str = "text",
        min_frequency: int = 500,
        max_samples: Optional[int] = 100000,
    ) -> "ChessTokenizer_v0":
        """
        Build a tokenizer vocabulary from a Hugging Face dataset.
        
        Args:
            dataset_name: Name of the dataset on Hugging Face Hub.
            split: Dataset split to use.
            column: Column containing the game strings.
            min_frequency: Minimum frequency for a token to be included (default: 500).
            max_samples: Maximum number of samples to process (default: 100k).
        
        Returns:
            A ChessTokenizer_v0 with the built vocabulary.
        """
        from datasets import load_dataset
        
        dataset = load_dataset(dataset_name, split=split)
        
        if max_samples is not None:
            dataset = dataset.select(range(min(max_samples, len(dataset))))
        
        def game_iterator():
            for example in dataset:
                yield example[column]
        
        return cls.build_vocab_from_iterator(game_iterator(), min_frequency=min_frequency)
    
    @property
    def vocab_size(self) -> int:
        """Return the size of the vocabulary."""
        return len(self._vocab)
    
    def get_vocab(self) -> Dict[str, int]:
        """Return the vocabulary as a dictionary."""
        return dict(self._vocab)
    
    def _tokenize(self, text: str) -> List[str]:
        """
        Tokenize a string of moves into a list of tokens.
        
        Args:
            text: A string of space-separated moves.
        
        Returns:
            List of move tokens.
        """
        return text.strip().split()
    
    def _convert_token_to_id(self, token: str) -> int:
        """Convert a token to its ID."""
        return self._vocab.get(token, self._vocab.get(self.UNK_TOKEN, 0))
    
    def _convert_id_to_token(self, index: int) -> str:
        """Convert an ID to its token."""
        return self._ids_to_tokens.get(index, self.UNK_TOKEN)
    
    def convert_tokens_to_string(self, tokens: List[str]) -> str:
        """Convert a list of tokens back to a string."""
        # Filter out special tokens for cleaner output
        special = {self.PAD_TOKEN, self.BOS_TOKEN, self.EOS_TOKEN, self.UNK_TOKEN}
        return " ".join(t for t in tokens if t not in special)
    
    def save_vocabulary(
        self,
        save_directory: str,
        filename_prefix: Optional[str] = None,
    ) -> tuple:
        """
        Save the vocabulary to a JSON file.
        
        Args:
            save_directory: Directory to save the vocabulary.
            filename_prefix: Optional prefix for the filename.
        
        Returns:
            Tuple containing the path to the saved vocabulary file.
        """
        if not os.path.isdir(save_directory):
            os.makedirs(save_directory, exist_ok=True)
        
        vocab_file = os.path.join(
            save_directory,
            (filename_prefix + "-" if filename_prefix else "") + "vocab.json",
        )
        
        with open(vocab_file, "w", encoding="utf-8") as f:
            json.dump(self._vocab, f, ensure_ascii=False, indent=2)
        
        return (vocab_file,)


def count_vocab_from_dataset(
    dataset_name: str = "dlouapre/lichess_2025-01_1M",
    split: str = "train",
    column: str = "text",
    max_samples: Optional[int] = 10000,
) -> Dict[str, int]:
    """
    Count token frequencies in a dataset (useful for vocabulary analysis).
    
    Args:
        dataset_name: Name of the dataset on Hugging Face Hub.
        split: Dataset split to use.
        column: Column containing the game strings.
        max_samples: Maximum number of samples to process.
    
    Returns:
        Dictionary mapping tokens to their frequencies.
    """
    from collections import Counter
    from datasets import load_dataset
    
    dataset = load_dataset(dataset_name, split=split)
    
    if max_samples is not None:
        dataset = dataset.select(range(min(max_samples, len(dataset))))
    
    token_counts = Counter()
    
    for example in dataset:
        moves = example[column].strip().split()
        token_counts.update(moves)
    
    return dict(token_counts)


# ============================================================================
# V1 IMPROVEMENTS: Sub-word tokenizer that decomposes moves into components
# ============================================================================

import re

# Regex to parse extended UCI move format: WPe2e4(x)(+) etc.
MOVE_PATTERN = re.compile(
    r"^(?P<side>[WB])"
    r"(?P<piece>[PNBRQK])"
    r"(?P<src>[a-h][1-8])"
    r"(?P<dst>[a-h][1-8])"
    r"(?P<suffix>.*)$"
)


class ChessTokenizer(PreTrainedTokenizer):
    """
    Sub-word chess tokenizer that decomposes moves into components.
    
    Instead of treating each move as a single token (requiring ~1500 tokens),
    this tokenizer breaks moves into:
    - Side: [W], [B]
    - Piece: [P], [N], [B], [R], [Q], [K]
    - Source square: [a1] through [h8]
    - Destination square: [a1] through [h8]
    - Optional suffixes: [x] (capture), [+] (check), [#] (checkmate), 
                        [O-O], [O-O-O], [=Q], [=R], [=B], [=N]
    
    Total vocabulary: ~90 tokens (vs ~1500 for whole-move tokenizer)
    
    Trade-off: Each move becomes 4-6 tokens instead of 1, but:
    - Saves ~100-200K embedding parameters
    - Model learns piece/square patterns independently
    - Zero OOV - can represent any legal move
    
    Example:
        "WPe2e4" -> ["[W]", "[P]", "[e2]", "[e4]"]
        "BNg8f6(x)(+)" -> ["[B]", "[N]", "[g8]", "[f6]", "[x]", "[+]"]
    """
    
    model_input_names = ["input_ids", "attention_mask"]
    vocab_files_names = {"vocab_file": "vocab.json"}
    
    # Special tokens
    PAD_TOKEN = "[PAD]"
    BOS_TOKEN = "[BOS]"
    EOS_TOKEN = "[EOS]"
    UNK_TOKEN = "[UNK]"
    
    def __init__(
        self,
        vocab_file: Optional[str] = None,
        vocab: Optional[Dict[str, int]] = None,
        **kwargs,
    ):
        self._pad_token = self.PAD_TOKEN
        self._bos_token = self.BOS_TOKEN
        self._eos_token = self.EOS_TOKEN
        self._unk_token = self.UNK_TOKEN

        kwargs.pop("pad_token", None)
        kwargs.pop("bos_token", None)
        kwargs.pop("eos_token", None)
        kwargs.pop("unk_token", None)
        
        if vocab is not None:
            self._vocab = vocab
        elif vocab_file is not None and os.path.exists(vocab_file):
            with open(vocab_file, "r", encoding="utf-8") as f:
                self._vocab = json.load(f)
        else:
            self._vocab = self._create_default_vocab()
        
        self._ids_to_tokens = {v: k for k, v in self._vocab.items()}
        
        super().__init__(
            pad_token=self._pad_token,
            bos_token=self._bos_token,
            eos_token=self._eos_token,
            unk_token=self._unk_token,
            **kwargs,
        )
    
    def _create_default_vocab(self) -> Dict[str, int]:
        """
        Create the fixed sub-word vocabulary.
        
        This vocabulary is complete - no need to build from data.
        """
        vocab_list = []
        
        # 1. Special tokens (4)
        vocab_list.extend([self.PAD_TOKEN, self.BOS_TOKEN, self.EOS_TOKEN, self.UNK_TOKEN])
        
        # 2. Side tokens (2)
        vocab_list.extend(["[W]", "[B]"])
        
        # 3. Piece tokens (6)
        vocab_list.extend(["[P]", "[N]", "[Bi]", "[R]", "[Q]", "[K]"])
        
        # 4. Square tokens (64)
        for rank in "12345678":
            for file in "abcdefgh":
                vocab_list.append(f"[{file}{rank}]")
        
        # 5. Suffix tokens
        vocab_list.extend([
            "[x]",      # capture
            "[+]",      # check
            "[#]",      # checkmate
            "[O-O]",    # kingside castling
            "[O-O-O]",  # queenside castling
            "[=Q]",     # promotion to queen
            "[=R]",     # promotion to rook
            "[=B]",     # promotion to bishop
            "[=N]",     # promotion to knight
        ])
        
        return {token: idx for idx, token in enumerate(vocab_list)}
    
    @property
    def vocab_size(self) -> int:
        return len(self._vocab)
    
    def get_vocab(self) -> Dict[str, int]:
        return dict(self._vocab)
    
    def _tokenize(self, text: str) -> List[str]:
        """
        Tokenize a string of moves into sub-word tokens.
        
        Args:
            text: A string of space-separated moves (e.g., "WPe2e4 BPe7e5")
        
        Returns:
            List of sub-word tokens
        """
        tokens = []
        moves = text.strip().split()
        
        for move in moves:
            tokens.extend(self._tokenize_move(move))
        
        return tokens
    
    def _tokenize_move(self, move: str) -> List[str]:
        """Parse a single move into component tokens."""
        # Handle castling first
        if "O-O-O" in move or "o-o-o" in move:
            side = "[W]" if move.startswith("W") else "[B]"
            return [side, "[O-O-O]"]
        
        if "O-O" in move or "o-o" in move:
            side = "[W]" if move.startswith("W") else "[B]"
            return [side, "[O-O]"]
        
        # Parse regular move
        match = MOVE_PATTERN.match(move)
        if not match:
            return [self.UNK_TOKEN]
        
        tokens = []
        
        # Side
        side = match.group("side")
        tokens.append(f"[{side}]")
        
        # Piece (use [Bi] for bishop to avoid confusion with [B] for black)
        piece = match.group("piece")
        if piece == "B":
            tokens.append("[Bi]")
        else:
            tokens.append(f"[{piece}]")
        
        # Source and destination squares
        tokens.append(f"[{match.group('src')}]")
        tokens.append(f"[{match.group('dst')}]")
        
        # Parse suffix for capture, check, checkmate, promotion
        suffix = match.group("suffix") or ""
        
        if "x" in suffix:
            tokens.append("[x]")
        
        # Checkmate before check (since checkmate contains +)
        if "*" in suffix or "#" in suffix:
            tokens.append("[#]")
        elif "+" in suffix:
            tokens.append("[+]")
        
        # Promotion
        if "=" in suffix:
            idx = suffix.find("=")
            if idx + 1 < len(suffix):
                promo_piece = suffix[idx + 1].upper()
                if promo_piece in "QRBN":
                    tokens.append(f"[={promo_piece}]")
        
        return tokens
    
    def _convert_token_to_id(self, token: str) -> int:
        return self._vocab.get(token, self._vocab.get(self.UNK_TOKEN, 0))
    
    def _convert_id_to_token(self, index: int) -> str:
        return self._ids_to_tokens.get(index, self.UNK_TOKEN)
    
    def convert_tokens_to_string(self, tokens: List[str]) -> str:
        """
        Convert tokens back to a readable string.
        
        This reconstructs moves from their component tokens.
        """
        special = {self.PAD_TOKEN, self.BOS_TOKEN, self.EOS_TOKEN, self.UNK_TOKEN}
        
        # Filter special tokens
        filtered = [t for t in tokens if t not in special]
        
        # Simple approach: just join with spaces
        # A more sophisticated approach would reconstruct full moves
        return " ".join(filtered)
    
    def decode_to_moves(self, token_ids: List[int]) -> List[str]:
        """
        Decode token IDs back to chess moves.
        
        Returns a list of reconstructed moves like ["WPe2e4", "BPe7e5"].
        """
        tokens = [self._convert_id_to_token(tid) for tid in token_ids]
        special = {self.PAD_TOKEN, self.BOS_TOKEN, self.EOS_TOKEN, self.UNK_TOKEN}
        
        moves = []
        current_move = []
        
        for token in tokens:
            if token in special:
                continue
            
            # Start new move on side token
            if token in ("[W]", "[B]"):
                if current_move:
                    moves.append(self._reconstruct_move(current_move))
                current_move = [token]
            else:
                current_move.append(token)
        
        # Don't forget last move
        if current_move:
            moves.append(self._reconstruct_move(current_move))
        
        return moves
    
    def _reconstruct_move(self, tokens: List[str]) -> str:
        """Reconstruct a move string from component tokens."""
        if not tokens:
            return ""
        
        # Handle castling
        if "[O-O-O]" in tokens:
            side = "W" if "[W]" in tokens else "B"
            return f"{side}KO-O-O"
        if "[O-O]" in tokens:
            side = "W" if "[W]" in tokens else "B"
            return f"{side}KO-O"
        
        move = ""
        
        for token in tokens:
            # Strip brackets
            inner = token[1:-1] if token.startswith("[") and token.endswith("]") else token
            
            if inner in ("W", "B"):
                move += inner
            elif inner == "Bi":
                move += "B"  # Bishop
            elif inner in "PNRQK":
                move += inner
            elif len(inner) == 2 and inner[0] in "abcdefgh" and inner[1] in "12345678":
                move += inner
            elif inner == "x":
                move += "(x)"
            elif inner == "+":
                move += "(+)"
            elif inner == "#":
                move += "(+*)"
            elif inner.startswith("="):
                move += f"({inner})"
        
        return move
    
    def save_vocabulary(
        self,
        save_directory: str,
        filename_prefix: Optional[str] = None,
    ) -> tuple:
        if not os.path.isdir(save_directory):
            os.makedirs(save_directory, exist_ok=True)
        
        vocab_file = os.path.join(
            save_directory,
            (filename_prefix + "-" if filename_prefix else "") + "vocab.json",
        )
        
        with open(vocab_file, "w", encoding="utf-8") as f:
            json.dump(self._vocab, f, ensure_ascii=False, indent=2)
        
        return (vocab_file,)
    
    def get_vocab_stats(self) -> Dict[str, int]:
        """Get statistics about vocabulary composition."""
        return {
            "special": 4,
            "sides": 2,
            "pieces": 6,
            "squares": 64,
            "suffixes": 9,
            "total": self.vocab_size,
        }
    
    # For compatibility - no need to build vocab from data anymore
    @classmethod
    def build_vocab_from_dataset(cls, **kwargs) -> "ChessTokenizer":
        """Return a tokenizer with the fixed vocabulary (no data needed)."""
        return cls()
    
    @classmethod
    def build_vocab_from_iterator(cls, iterator, **kwargs) -> "ChessTokenizer":
        """Return a tokenizer with the fixed vocabulary (no data needed)."""
        return cls()