import os
import torch
import numpy as np
import torchaudio
import yaml
from . import asteroid_test


def get_conf():
    conf_filterbank = {
        'n_filters': 64,
        'kernel_size': 16,
        'stride': 8
    }

    conf_masknet = {
        'in_chan': 64,
        'n_src': 2,
        'out_chan': 64,
        'ff_hid': 256,
        'ff_activation': "relu",
        'norm_type': "gLN",
        'chunk_size': 100,
        'hop_size': 50,
        'n_repeats': 2,
        'mask_act': 'sigmoid',
        'bidirectional': True,
        'dropout': 0
    }
    return conf_filterbank, conf_masknet


def load_dpt_model():
    print('Load Separation Model...')
    now_path = os.path.split(os.path.realpath(__file__))[0]
    conf_filterbank, conf_masknet = get_conf()
    model_path = os.path.join(now_path, "trained_model/train_dptnet_aishell_partOverlap_B2_300epoch_quan-int8.p")
    model = getattr(asteroid_test, "DPTNet")(**conf_filterbank, **conf_masknet)
    model = torch.quantization.quantize_dynamic(model, {torch.nn.LSTM, torch.nn.Linear}, dtype=torch.qint8)
    state_dict = torch.load(model_path, map_location="cpu")
    model.load_state_dict(state_dict)
    model.eval()
    return model

def dpt_sep_process(wav_path, model=None, outfilename=None):
    if model is None:
        model = load_model()

    x, sr = torchaudio.load(wav_path)
    x = x.cpu()

    with torch.no_grad():
        est_sources = model(x)  # shape: (1, 2, T)

    # 確保 est_sources 是 (1, 2, T)，再拆分
    est_sources = est_sources.squeeze(0)  # shape: (2, T)

    sep_1, sep_2 = est_sources  # 拆成兩個 (T, ) 的 tensor

    # 正規化
    max_abs = x[0].abs().max().item()
    sep_1 = sep_1 * max_abs / sep_1.abs().max().item()
    sep_2 = sep_2 * max_abs / sep_2.abs().max().item()

    # 增加 channel 維度，變為 (1, T)
    sep_1 = sep_1.unsqueeze(0)
    sep_2 = sep_2.unsqueeze(0)

    if outfilename is not None:
        torchaudio.save(outfilename.replace('.wav', '_sep1.wav'), sep_1, sr)
        torchaudio.save(outfilename.replace('.wav', '_sep2.wav'), sep_2, sr)
        torchaudio.save(outfilename.replace('.wav', '_mix.wav'), x, sr)
    else:
        torchaudio.save(wav_path.replace('.wav', '_sep1.wav'), sep_1, sr)
        torchaudio.save(wav_path.replace('.wav', '_sep2.wav'), sep_2, sr)
        
# def dpt_sep_process(wav_path, model=None, outfilename=None):
#     if model == None:
#         model = load_model()
#     x, sr = torchaudio.load(wav_path)
#     x = x.cpu()
#     with torch.no_grad():
#         est_sources = model(x)

#     est_sources_np = est_sources.squeeze(0)

#     sep_1, sep_2 = est_sources_np
#     sep_1 = sep_1 * x[0].abs().max().item() / sep_1.abs().max().item()
#     sep_2 = sep_2 * x[0].abs().max().item() / sep_2.abs().max().item()

#     if outfilename != None:
#         torchaudio.save(outfilename.replace('.wav', '_sep1.wav'), sep_1, sr)
#         torchaudio.save(outfilename.replace('.wav', '_sep2.wav'), sep_2, sr)
#         torchaudio.save(outfilename.replace('.wav', '_mix.wav'), x, sr)
#     else:
#         torchaudio.save(wav_path.replace('.wav', '_sep1.wav'), sep_1, sr)
#         torchaudio.save(wav_path.replace('.wav', '_sep2.wav'), sep_2, sr)

if __name__ == '__main__':
    print("This module should be used via Flask or Gradio.")