"""
Forked Tongue — model definition.
"""
from __future__ import annotations

from dataclasses import dataclass

import torch
import torch.nn as nn
import torch.nn.functional as F


@dataclass
class GPTConfig:
    vocab_size: int
    block_size: int = 128
    n_layer: int = 4
    n_head: int = 4
    n_embd: int = 128
    dropout: float = 0.0


class CausalSelfAttention(nn.Module):
    def __init__(self, cfg: GPTConfig):
        super().__init__()
        assert cfg.n_embd % cfg.n_head == 0
        self.n_head = cfg.n_head
        self.n_embd = cfg.n_embd
        self.c_attn = nn.Linear(cfg.n_embd, 3 * cfg.n_embd)
        self.c_proj = nn.Linear(cfg.n_embd, cfg.n_embd)
        self.attn_dropout = nn.Dropout(cfg.dropout)
        self.resid_dropout = nn.Dropout(cfg.dropout)
        self.register_buffer(
            "mask",
            torch.tril(torch.ones(cfg.block_size, cfg.block_size)).view(
                1, 1, cfg.block_size, cfg.block_size
            ),
        )

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        B, T, C = x.shape
        q, k, v = self.c_attn(x).split(self.n_embd, dim=2)
        head = C // self.n_head
        q = q.view(B, T, self.n_head, head).transpose(1, 2)
        k = k.view(B, T, self.n_head, head).transpose(1, 2)
        v = v.view(B, T, self.n_head, head).transpose(1, 2)
        att = (q @ k.transpose(-2, -1)) * (1.0 / (head ** 0.5))
        att = att.masked_fill(self.mask[:, :, :T, :T] == 0, float("-inf"))
        att = F.softmax(att, dim=-1)
        att = self.attn_dropout(att)
        y = att @ v
        y = y.transpose(1, 2).contiguous().view(B, T, C)
        return self.resid_dropout(self.c_proj(y))


class MLP(nn.Module):
    def __init__(self, cfg: GPTConfig):
        super().__init__()
        self.c_fc = nn.Linear(cfg.n_embd, 4 * cfg.n_embd)
        self.c_proj = nn.Linear(4 * cfg.n_embd, cfg.n_embd)
        self.dropout = nn.Dropout(cfg.dropout)

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        return self.dropout(self.c_proj(F.gelu(self.c_fc(x))))


class Block(nn.Module):
    def __init__(self, cfg: GPTConfig):
        super().__init__()
        self.ln_1 = nn.LayerNorm(cfg.n_embd)
        self.attn = CausalSelfAttention(cfg)
        self.ln_2 = nn.LayerNorm(cfg.n_embd)
        self.mlp = MLP(cfg)

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        x = x + self.attn(self.ln_1(x))
        x = x + self.mlp(self.ln_2(x))
        return x


class TinyGPT(nn.Module):
    """A minimal GPT. ``config`` is stored so released weights are self-describing."""

    def __init__(self, cfg: GPTConfig):
        super().__init__()
        self.cfg = cfg
        self.wte = nn.Embedding(cfg.vocab_size, cfg.n_embd)
        self.wpe = nn.Embedding(cfg.block_size, cfg.n_embd)
        self.drop = nn.Dropout(cfg.dropout)
        self.blocks = nn.ModuleList([Block(cfg) for _ in range(cfg.n_layer)])
        self.ln_f = nn.LayerNorm(cfg.n_embd)
        self.lm_head = nn.Linear(cfg.n_embd, cfg.vocab_size, bias=False)
        self.wte.weight = self.lm_head.weight  # weight tying

    def forward(self, idx: torch.Tensor) -> torch.Tensor:
        B, T = idx.shape
        assert T <= self.cfg.block_size, "sequence longer than block size"
        pos = torch.arange(0, T, dtype=torch.long, device=idx.device).unsqueeze(0)
        x = self.drop(self.wte(idx) + self.wpe(pos))
        for block in self.blocks:
            x = block(x)
        x = self.ln_f(x)
        return self.lm_head(x)

    @torch.no_grad()
    def generate(
        self,
        idx: torch.Tensor,
        max_new_tokens: int,
        eos_id: int | None = None,
    ) -> torch.Tensor:
        """Greedy (argmax) decoding. Deterministic. Stops early on ``eos_id``."""
        self.eval()
        for _ in range(max_new_tokens):
            idx_cond = idx[:, -self.cfg.block_size :]
            logits = self(idx_cond)
            next_id = torch.argmax(logits[:, -1, :], dim=-1, keepdim=True)
            idx = torch.cat([idx, next_id], dim=1)
            if eos_id is not None and int(next_id) == eos_id:
                break
        return idx