# Copyright © 2025 Apple Inc.

from dataclasses import dataclass
from typing import Any, Dict, Optional, Union

import mlx.core as mx
import mlx.nn as nn

from .activations import swiglu
from .base import BaseModelArgs, create_attention_mask, scaled_dot_product_attention
from .rope_utils import initialize_rope


@dataclass
class ModelArgs(BaseModelArgs):
    model_type: str
    hidden_size: int
    num_hidden_layers: int
    intermediate_size: int
    num_attention_heads: int
    rms_norm_eps: float
    vocab_size: int
    head_dim: int
    num_key_value_heads: int
    max_position_embeddings: Optional[int] = None
    attention_bias: bool = False
    rope_theta: float = 10000
    tie_word_embeddings: bool = True


class GLMAttention(nn.Module):
    def __init__(self, args: ModelArgs):
        super().__init__()
        self.hidden_size = args.hidden_size
        self.num_attention_heads = args.num_attention_heads
        self.num_key_value_heads = args.num_key_value_heads
        self.head_dim = args.head_dim or args.hidden_size // self.num_attention_heads
        self.scale = self.head_dim**-0.5

        self.q_proj = nn.Linear(
            self.hidden_size,
            self.num_attention_heads * self.head_dim,
            bias=args.attention_bias,
        )
        self.k_proj = nn.Linear(
            self.hidden_size,
            self.num_key_value_heads * self.head_dim,
            bias=args.attention_bias,
        )
        self.v_proj = nn.Linear(
            self.hidden_size,
            self.num_key_value_heads * self.head_dim,
            bias=args.attention_bias,
        )
        self.o_proj = nn.Linear(
            self.num_attention_heads * self.head_dim, self.hidden_size, bias=False
        )

        self.rope = nn.RoPE(dims=self.head_dim, traditional=True, base=args.rope_theta)

    def __call__(
        self,
        x: mx.array,
        mask: Optional[mx.array] = None,
        cache: Optional[Any] = None,
    ) -> mx.array:
        B, L, D = x.shape

        queries, keys, values = self.q_proj(x), self.k_proj(x), self.v_proj(x)

        queries = queries.reshape(B, L, self.num_attention_heads, -1).transpose(
            0, 2, 1, 3
        )
        keys = keys.reshape(B, L, self.num_key_value_heads, -1).transpose(0, 2, 1, 3)
        values = values.reshape(B, L, self.num_key_value_heads, -1).transpose(
            0, 2, 1, 3
        )

        if cache is not None:
            queries = self.rope(queries, offset=cache.offset)
            keys = self.rope(keys, offset=cache.offset)
            keys, values = cache.update_and_fetch(keys, values)
        else:
            queries = self.rope(queries)
            keys = self.rope(keys)

        output = scaled_dot_product_attention(
            queries, keys, values, cache=cache, scale=self.scale, mask=mask
        )

        output = output.transpose(0, 2, 1, 3).reshape(B, L, -1)
        return self.o_proj(output)


class GLMMLP(nn.Module):
    def __init__(self, args: ModelArgs):
        super().__init__()
        self.gate_up_proj = nn.Linear(
            args.hidden_size, 2 * args.intermediate_size, bias=False
        )
        self.down_proj = nn.Linear(args.intermediate_size, args.hidden_size, bias=False)

    def __call__(self, x) -> mx.array:
        x = self.gate_up_proj(x)
        gate, x = mx.split(x, 2, axis=-1)
        return self.down_proj(swiglu(gate, x))


class GLMBlock(nn.Module):
    def __init__(self, args: ModelArgs):
        super().__init__()
        self.self_attn = GLMAttention(args)
        self.mlp = GLMMLP(args)
        self.input_layernorm = nn.RMSNorm(args.hidden_size, eps=args.rms_norm_eps)
        self.post_attention_layernorm = nn.RMSNorm(
            args.hidden_size, eps=args.rms_norm_eps
        )

    def __call__(
        self,
        x: mx.array,
        mask: Optional[mx.array] = None,
        cache: Optional[Any] = None,
    ) -> mx.array:
        r = self.self_attn(self.input_layernorm(x), mask, cache)
        h = x + r
        r = self.mlp(self.post_attention_layernorm(h))
        out = h + r
        return out


class GLMModel(nn.Module):
    def __init__(self, args: ModelArgs):
        super().__init__()
        self.embed_tokens = nn.Embedding(args.vocab_size, args.hidden_size)
        self.layers = [GLMBlock(args=args) for _ in range(args.num_hidden_layers)]
        self.norm = nn.RMSNorm(args.hidden_size, eps=args.rms_norm_eps)

    def __call__(
        self,
        inputs: mx.array,
        cache: Optional[Any] = None,
    ) -> mx.array:
        h = self.embed_tokens(inputs)

        if cache is None:
            cache = [None] * len(self.layers)

        mask = create_attention_mask(h, cache[0])

        for layer, c in zip(self.layers, cache):
            h = layer(h, mask, c)

        return self.norm(h)


class Model(nn.Module):
    def __init__(self, args: ModelArgs):
        super().__init__()
        self.args = args
        self.model_type = args.model_type
        self.model = GLMModel(args)
        if not args.tie_word_embeddings:
            self.lm_head = nn.Linear(args.hidden_size, args.vocab_size, bias=False)

    def __call__(
        self,
        inputs: mx.array,
        cache: Optional[Any] = None,
    ) -> mx.array:
        out = self.model(inputs, cache)
        if self.args.tie_word_embeddings:
            out = self.model.embed_tokens.as_linear(out)
        else:
            out = self.lm_head(out)
        return out

    def sanitize(self, weights):
        weights = {
            k: v for k, v in weights.items() if "self_attn.rotary_emb.inv_freq" not in k
        }
        if self.args.tie_word_embeddings:
            weights.pop("lm_head.weight", None)
        return weights

    @property
    def layers(self):
        return self.model.layers