import math
import torch
import torch.nn as nn
from torch.nn import functional as F

class HierarchicalPositionEncoding(nn.Module):
    """
    Hierarchical Position Encoding that captures position information at multiple scales:
    - Fine-grained local position (token level)
    - Medium-scale position (segment level)
    - Coarse-grained position (document level)
    """
    def __init__(self, d_model, max_len=1024, base=10000):
        super().__init__()
        self.d_model = d_model
        self.max_len = max_len
        self.base = base
        
        # Split embedding dimensions for different scales
        self.local_dim = d_model // 2
        self.segment_dim = d_model // 4
        self.doc_dim = d_model - self.local_dim - self.segment_dim
        
        # Create position encodings for different scales
        self.register_buffer('local_pe', self._create_pe(max_len, self.local_dim))
        self.register_buffer('segment_pe', self._create_pe(max_len//8, self.segment_dim))
        self.register_buffer('doc_pe', self._create_pe(max_len//32, self.doc_dim))
        
    def _create_pe(self, max_len, d_model):
        pe = torch.zeros(max_len, d_model)
        position = torch.arange(0, max_len, dtype=torch.float).unsqueeze(1)
        div_term = torch.exp(torch.arange(0, d_model, 2).float() * (-math.log(self.base) / d_model))
        pe[:, 0::2] = torch.sin(position * div_term)
        pe[:, 1::2] = torch.cos(position * div_term)
        return pe.unsqueeze(0)
    
    def forward(self, x):
        B, T, C = x.shape
        
        # Get positional encodings at different scales
        local_pos = self.local_pe[:, :T, :]
        segment_pos = self.segment_pe[:, :(T//8), :].repeat_interleave(8, dim=1)[:, :T, :]
        doc_pos = self.doc_pe[:, :(T//32), :].repeat_interleave(32, dim=1)[:, :T, :]
        
        # Combine all scales
        pos_encoding = torch.cat([local_pos, segment_pos, doc_pos], dim=-1)
        return pos_encoding

class MultiScaleAttention(nn.Module):
    """
    Multi-scale attention mechanism that processes information at different temporal scales
    """
    def __init__(self, config):
        super().__init__()
        assert config.n_embd % config.n_head == 0
        
        # key, query, value projections for all heads, but in a batch
        self.c_attn = nn.Linear(config.n_embd, 3 * config.n_embd, bias=config.bias)
        # output projection
        self.c_proj = nn.Linear(config.n_embd, config.n_embd, bias=config.bias)
        # regularization
        self.attn_dropout = nn.Dropout(config.dropout)
        self.resid_dropout = nn.Dropout(config.dropout)
        self.n_head = config.n_head
        self.n_embd = config.n_embd
        self.dropout = config.dropout
        
    def forward(self, x):
        B, T, C = x.shape # batch size, sequence length, embedding dimensionality

        # calculate query, key, values for all heads in batch and move head forward to be the batch dim
        q, k, v  = self.c_attn(x).split(self.n_embd, dim=2)
        k = k.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # (B, nh, T, hs)
        q = q.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # (B, nh, T, hs)
        v = v.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # (B, nh, T, hs)

        # causal self-attention; Self-attend: (B, nh, T, hs) x (B, nh, hs, T) -> (B, nh, T, T)
        att = (q @ k.transpose(-2, -1)) * (1.0 / math.sqrt(k.size(-1)))
        att = F.softmax(att, dim=-1)
        att = self.attn_dropout(att)
        y = att @ v # (B, nh, T, T) x (B, nh, T, hs) -> (B, nh, T, hs)
        y = y.transpose(1, 2).contiguous().view(B, T, C) # re-assemble all head outputs side by side

        # output projection
        y = self.resid_dropout(self.c_proj(y))
        return y

class Block(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.ln_1 = nn.LayerNorm(config.n_embd)
        self.attn = MultiScaleAttention(config)
        self.ln_2 = nn.LayerNorm(config.n_embd)
        self.mlp = nn.ModuleDict(dict(
            c_fc    = nn.Linear(config.n_embd, 4 * config.n_embd, bias=config.bias),
            c_proj  = nn.Linear(4 * config.n_embd, config.n_embd, bias=config.bias),
            act     = nn.GELU(),
            dropout = nn.Dropout(config.dropout),
        ))
        m = self.mlp
        self.mlpf = lambda x: m.dropout(m.c_proj(m.act(m.c_fc(x))))

    def forward(self, x):
        x = x + self.attn(self.ln_1(x))
        x = x + self.mlpf(self.ln_2(x))
        return x

class GPTModified(nn.Module):
    def __init__(self, config):
        super().__init__()
        assert config.vocab_size is not None
        assert config.block_size is not None
        self.config = config

        self.transformer = nn.ModuleDict(dict(
            wte = nn.Embedding(config.vocab_size, config.n_embd),
            hpe = HierarchicalPositionEncoding(config.n_embd, config.block_size),
            drop = nn.Dropout(config.dropout),
            h = nn.ModuleList([Block(config) for _ in range(config.n_layer)]),
            ln_f = nn.LayerNorm(config.n_embd),
        ))
        self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)
        
        # Initialize weights
        self.apply(self._init_weights)
        # Apply special scaled init to the residual projections, per GPT-2 paper
        for pn, p in self.named_parameters():
            if pn.endswith('c_proj.weight'):
                torch.nn.init.normal_(p, mean=0.0, std=0.02/math.sqrt(2 * config.n_layer))

        # Report number of parameters
        print("number of parameters: %.2fM" % (self.get_num_params()/1e6,))

    def get_num_params(self, non_embedding=True):
        n_params = sum(p.numel() for p in self.parameters())
        if non_embedding:
            n_params -= self.transformer.wte.weight.numel()
        return n_params

    def _init_weights(self, module):
        if isinstance(module, nn.Linear):
            torch.nn.init.normal_(module.weight, mean=0.0, std=0.02)
            if module.bias is not None:
                torch.nn.init.zeros_(module.bias)
        elif isinstance(module, nn.Embedding):
            torch.nn.init.normal_(module.weight, mean=0.0, std=0.02)

    def forward(self, idx, targets=None):
        device = idx.device
        b, t = idx.size()
        assert t <= self.config.block_size, f"Cannot forward sequence of length {t}, block size is only {self.config.block_size}"
        pos = torch.arange(0, t, dtype=torch.long, device=device).unsqueeze(0) # shape (1, t)

        # Forward pass
        tok_emb = self.transformer.wte(idx) # token embeddings of shape (b, t, n_embd)
        pos_emb = self.transformer.hpe(tok_emb) # position embeddings of shape (b, t, n_embd)
        x = self.transformer.drop(tok_emb + pos_emb)
        for block in self.transformer.h:
            x = block(x)
        x = self.transformer.ln_f(x)
        logits = self.lm_head(x)

        # If we are given some desired targets also calculate the loss
        loss = None
        if targets is not None:
            loss = F.cross_entropy(logits.view(-1, logits.size(-1)), targets.view(-1), ignore_index=-1)

        return logits, loss

    @torch.no_grad()
    def generate(self, idx, max_new_tokens, temperature=1.0, top_k=None):
        for _ in range(max_new_tokens):
            # If the sequence context is growing too long we must crop it at block_size
            idx_cond = idx if idx.size(1) <= self.config.block_size else idx[:, -self.config.block_size:]
            # Forward the model to get the logits for the index in the sequence
            logits, _ = self(idx_cond)
            # Pluck the logits at the final step and scale by desired temperature
            logits = logits[:, -1, :] / temperature
            # Optionally crop the logits to only the top k options
            if top_k is not None:
                v, _ = torch.topk(logits, min(top_k, logits.size(-1)))
                logits[logits < v[:, [-1]]] = -float('Inf')
            # Apply softmax to convert logits to (normalized) probabilities
            probs = F.softmax(logits, dim=-1)
            # Sample from the distribution
            idx_next = torch.multinomial(probs, num_samples=1)
            # Append sampled index to the running sequence and continue
            idx = torch.cat((idx, idx_next), dim=1)

        return idx