import numpy as np
import soundfile as sf
import json
import subprocess
import os

def mel_frequencies(n_mels=128, fmin=0.0, fmax=8000.0):
    """Create a list of frequencies uniformly spaced on the Mel scale."""
    fmin_mel = 2595.0 * np.log10(1.0 + fmin / 700.0)
    fmax_mel = 2595.0 * np.log10(1.0 + fmax / 700.0)
    mels = np.linspace(fmin_mel, fmax_mel, num=n_mels)
    freqs = 700.0 * (10.0**(mels / 2595.0) - 1.0)
    return freqs

def update_waybar():
    # Update the Waybar module with the new data
    subprocess.run(['pkill', '-SIGRTMIN+1', 'waybar'])

home_dir = os.path.expanduser('~')
audio_file_path = os.path.join(home_dir, 'audio.wav')

data, samplerate = sf.read(audio_file_path)
fft = np.fft.rfft(data, axis=0)
fft_mean = np.mean(np.abs(fft), axis=1)

num_bins = 5
mel_freqs = mel_frequencies(n_mels=num_bins+2, fmin=0, fmax=samplerate/2)  # Ensure fmax does not exceed Nyquist frequency
bins = np.round((len(fft_mean) - 1) * mel_freqs / (samplerate/2)).astype(int)
fft_mean_binned = np.add.reduceat(fft_mean, bins[:-1]) / np.diff(bins)
fft_mean_binned = np.nan_to_num(fft_mean_binned)

# Normalize the binned data to the range 1-8 (inclusive)
normalized_bins = np.interp(fft_mean_binned, (fft_mean_binned.min(), fft_mean_binned.max()), (1, 8))

# Convert the normalized data to block characters
block_chars = [' ', '▁', '▂', '▃', '▄', '▅', '▆', '▇', '█']
visualizer_string = ''.join(block_chars[int(round(bin_value))] for bin_value in normalized_bins)

# Convert the visualizer string to JSON
json_data = json.dumps({"data": visualizer_string})

# Print the JSON data
print(json_data)