import torch
import torch.nn as nn
import numpy as np
from typing import Optional, Tuple
from .patch_embedding import PatchEmbedding
from .transformer_block import TransformerBlock

class Encoder(nn.Module):
    """
    MAE Encoder based on Vision Transformer.
    """
    def __init__(
        self, 
        img_size: int = 224, 
        patch_size: int = 16, 
        in_chans: int = 3, 
        embed_dim: int = 768, 
        depth: int = 12, 
        num_heads: int = 12, 
        mlp_ratio: float = 4.0, 
        norm_layer: nn.Module = nn.LayerNorm
    ):
        super().__init__()
        
        self.patch_embed = PatchEmbedding(img_size, patch_size, in_chans, embed_dim)
        num_patches = self.patch_embed.num_patches
        
        self.cls_token = nn.Parameter(torch.zeros(1, 1, embed_dim))
        self.pos_embed = nn.Parameter(torch.zeros(1, num_patches + 1, embed_dim), requires_grad=False)
        
        self.blocks = nn.ModuleList([
            TransformerBlock(embed_dim, num_heads, mlp_ratio, qkv_bias=True, norm_layer=norm_layer)
            for _ in range(depth)
        ])
        self.norm = norm_layer(embed_dim)
        
        self.initialize_weights()

    def initialize_weights(self):
        # Initialize pos_embed with sin-cos embedding
        pos_embed = self.get_2d_sincos_pos_embed(self.pos_embed.shape[-1], int(self.patch_embed.num_patches**.5), cls_token=True)
        self.pos_embed.data.copy_(torch.from_numpy(pos_embed).float().unsqueeze(0))
        
        # Initialize patch_embed like nn.Linear
        w = self.patch_embed.proj.weight.data
        torch.nn.init.xavier_uniform_(w.view([w.shape[0], -1]))
        
        # Initialize cls_token
        torch.nn.init.normal_(self.cls_token, std=.02)
        
        # Initialize transformer blocks
        self.apply(self._init_weights)

    def _init_weights(self, m):
        if isinstance(m, nn.Linear):
            torch.nn.init.xavier_uniform_(m.weight)
            if isinstance(m, nn.Linear) and m.bias is not None:
                nn.init.constant_(m.bias, 0)
        elif isinstance(m, nn.LayerNorm):
            nn.init.constant_(m.bias, 0)
            nn.init.constant_(m.weight, 1.0)

    def get_2d_sincos_pos_embed(self, embed_dim, grid_size, cls_token=False):
        """
        grid_size: int of the grid height and width
        return:
        pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim] (w/ or w/o cls_token)
        """
        grid_h = np.arange(grid_size, dtype=np.float32)
        grid_w = np.arange(grid_size, dtype=np.float32)
        grid = np.meshgrid(grid_w, grid_h)  # here w goes first
        grid = np.stack(grid, axis=0)
        grid = grid.reshape([2, 1, grid_size, grid_size])
        
        pos_embed = self.get_2d_sincos_pos_embed_from_grid(embed_dim, grid)
        if cls_token:
            pos_embed = np.concatenate([np.zeros([1, embed_dim]), pos_embed], axis=0)
        return pos_embed

    def get_2d_sincos_pos_embed_from_grid(self, embed_dim, grid):
        assert embed_dim % 2 == 0
        # use half of dimensions to encode grid_h
        emb_h = self.get_1d_sincos_pos_embed_from_grid(embed_dim // 2, grid[0])  # (H*W, D/2)
        emb_w = self.get_1d_sincos_pos_embed_from_grid(embed_dim // 2, grid[1])  # (H*W, D/2)
        emb = np.concatenate([emb_h, emb_w], axis=1) # (H*W, D)
        return emb

    def get_1d_sincos_pos_embed_from_grid(self, embed_dim, pos):
        assert embed_dim % 2 == 0
        omega = np.arange(embed_dim // 2, dtype=np.float32)
        omega /= embed_dim / 2.
        omega = 1. / 10000**omega  # (D/2,)

        pos = pos.reshape(-1)  # (M,)
        out = np.einsum('m,d->md', pos, omega)  # (M, D/2), outer product

        emb_sin = np.sin(out) # (M, D/2)
        emb_cos = np.cos(out) # (M, D/2)

        emb = np.concatenate([emb_sin, emb_cos], axis=1)  # (M, D)
        return emb

    def forward(self, x: torch.Tensor, mask_ratio: float = 0.0) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
        # Embed patches
        x = self.patch_embed(x)
        
        # Add pos embed w/o cls token
        x = x + self.pos_embed[:, 1:, :]
        
        # Masking
        if mask_ratio > 0:
            from .masking import random_masking
            x, mask, ids_restore = random_masking(x, mask_ratio)
        else:
            mask, ids_restore = None, None
            
        # Append cls token
        cls_token = self.cls_token + self.pos_embed[:, :1, :]
        cls_tokens = cls_token.expand(x.shape[0], -1, -1)
        x = torch.cat((cls_tokens, x), dim=1)
        
        # Apply transformer blocks
        for block in self.blocks:
            x = block(x)
        x = self.norm(x)
        
        return x, mask, ids_restore