219 lines
7.3 KiB
Python
219 lines
7.3 KiB
Python
import logging
|
|
import os
|
|
import subprocess
|
|
from typing import Dict
|
|
from xml.etree import ElementTree
|
|
|
|
import acoustid
|
|
import cv2
|
|
import imagehash
|
|
import librosa
|
|
import numpy as np
|
|
from PIL import Image
|
|
from fastdtw import fastdtw
|
|
from scipy.spatial.distance import cosine
|
|
from skimage.metrics import structural_similarity as ssim
|
|
|
|
logger = logging.getLogger("desktopenv.metrics.vlc")
|
|
|
|
|
|
def is_vlc_playing(actual_status_path: str, rule: Dict[str, str]) -> float:
|
|
"""
|
|
Checks if VLC is currently playing a file.
|
|
"""
|
|
with open(actual_status_path, 'rb') as file:
|
|
actual_status = file.read().decode('utf-8')
|
|
|
|
tree = ElementTree.fromstring(actual_status)
|
|
status = tree.find('state').text
|
|
if status == 'playing':
|
|
if rule['type'] == 'file_name':
|
|
file_info = tree.find('information/category[@name="meta"]/info[@name="filename"]').text
|
|
if file_info:
|
|
return 1 if file_info.endswith(rule['file_name']) else 0
|
|
elif rule['type'] == 'url':
|
|
file_info = tree.find('information/category[@name="meta"]/info[@name="url"]').text
|
|
if file_info:
|
|
return 1 if file_info.endswith(rule['url']) else 0
|
|
else:
|
|
logger.error(f"Unknown type: {rule['type']}")
|
|
return 0
|
|
else:
|
|
return 0
|
|
|
|
|
|
# fixme: part of this function can be moved to getters
|
|
def is_vlc_recordings_folder(actual_config_path: str, rule: Dict[str, str]) -> float:
|
|
"""
|
|
Checks if VLC's recording folder is set to the expected value.
|
|
"""
|
|
with open(actual_config_path, 'rb') as file:
|
|
config_file = file.read().decode('utf-8')
|
|
|
|
expected_recording_file_path = rule['recording_file_path']
|
|
|
|
try:
|
|
for line in config_file:
|
|
# Skip comments and empty lines
|
|
if line.startswith('#') or not line.strip():
|
|
continue
|
|
# Check if the line contains the recording path setting
|
|
if 'recorded_files_path' in line:
|
|
# Extract the value of the recording path and remove surrounding whitespace
|
|
current_path = line.split('=')[-1].strip()
|
|
# Compare with the Desktop path
|
|
if current_path == expected_recording_file_path:
|
|
return 1
|
|
else:
|
|
return 0
|
|
# The configuration key was not found in the file
|
|
return 0
|
|
except FileNotFoundError:
|
|
logger.error("VLC configuration file not found.")
|
|
return 0
|
|
except Exception as e:
|
|
logger.error(f"An error occurred: {e}")
|
|
return 0
|
|
|
|
|
|
def is_vlc_fullscreen(actual_window_size, screen_size):
|
|
if actual_window_size['width'] == screen_size['width'] and actual_window_size['height'] == screen_size['height']:
|
|
return 1
|
|
else:
|
|
return 0
|
|
|
|
|
|
def compare_images(image1_path, image2_path):
|
|
# You would call this function with the paths to the two images you want to compare:
|
|
# score = compare_images('path_to_image1', 'path_to_image2')
|
|
# print("Similarity score:", score)
|
|
|
|
if not image1_path or not image2_path:
|
|
return 0
|
|
|
|
# Open the images and convert to grayscale
|
|
image1 = Image.open(image1_path).convert('L')
|
|
image2 = Image.open(image2_path).convert('L')
|
|
|
|
# Resize images to the smaller one's size for comparison
|
|
image1_size = image1.size
|
|
image2_size = image2.size
|
|
new_size = min(image1_size, image2_size)
|
|
|
|
image1 = image1.resize(new_size, Image.Resampling.LANCZOS)
|
|
image2 = image2.resize(new_size, Image.Resampling.LANCZOS)
|
|
|
|
# Convert images to numpy arrays
|
|
image1_array = np.array(image1)
|
|
image2_array = np.array(image2)
|
|
|
|
# Calculate SSIM between two images
|
|
similarity_index = ssim(image1_array, image2_array)
|
|
|
|
return similarity_index
|
|
|
|
|
|
def compare_audios(audio_path_1, audio_path_2):
|
|
"""
|
|
Compare two audio files and return a similarity score in the range [0, 1].
|
|
audio_path_1, audio_path_2: paths to the audio files to compare
|
|
"""
|
|
# Example Usage:
|
|
# similarity = compare_audios_simple('path_to_audio1.mp3', 'path_to_audio2.mp3')
|
|
# print(f'Similarity Score: {similarity}')
|
|
|
|
# Convert to common format if necessary and load audio
|
|
if not audio_path_1 or not audio_path_2:
|
|
return 0
|
|
|
|
# Load the audio files and extract MFCC features
|
|
y1, sr1 = librosa.load(audio_path_1)
|
|
mfcc1 = librosa.feature.mfcc(y=y1, sr=sr1)
|
|
|
|
y2, sr2 = librosa.load(audio_path_2)
|
|
mfcc2 = librosa.feature.mfcc(y=y2, sr=sr2)
|
|
|
|
# Normalize the MFCC features
|
|
mfcc1 = librosa.util.normalize(mfcc1, axis=1)
|
|
mfcc2 = librosa.util.normalize(mfcc2, axis=1)
|
|
|
|
# Define a lambda function to compute cosine distance
|
|
dist_func = lambda x, y: cosine(x, y)
|
|
|
|
# Use the DTW algorithm to find the best alignment path
|
|
distance, path = fastdtw(mfcc1.T, mfcc2.T, dist=dist_func)
|
|
|
|
# Calculate the similarity score, here we use 1/(1+distance) to convert distance to a similarity score
|
|
similarity = 1 / (1 + distance)
|
|
|
|
return similarity
|
|
|
|
|
|
def compare_audios_by_dl_model(audio_path_1, audio_path_2):
|
|
pass
|
|
|
|
|
|
def compare_videos(video_path1, video_path2, max_frames_to_check=100, threshold=5):
|
|
# Open both video files
|
|
cap1 = cv2.VideoCapture(video_path1)
|
|
cap2 = cv2.VideoCapture(video_path2)
|
|
|
|
frames_checked = 0
|
|
mismatch_count = 0
|
|
|
|
while frames_checked < max_frames_to_check:
|
|
# Read frames from both videos
|
|
ret1, frame1 = cap1.read()
|
|
ret2, frame2 = cap2.read()
|
|
|
|
# If a video ends, then check if both ended to confirm they are of the same length
|
|
if not ret1 or not ret2:
|
|
return ret1 == ret2
|
|
|
|
# Convert frames to PIL Images
|
|
frame1 = Image.fromarray(cv2.cvtColor(frame1, cv2.COLOR_BGR2RGB))
|
|
frame2 = Image.fromarray(cv2.cvtColor(frame2, cv2.COLOR_BGR2RGB))
|
|
|
|
# Compute the perceptual hash for each frame
|
|
hash1 = imagehash.phash(frame1)
|
|
hash2 = imagehash.phash(frame2)
|
|
|
|
# Increment the frames checked
|
|
frames_checked += 1
|
|
|
|
# Compute the difference in the hashes
|
|
if hash1 - hash2 > threshold:
|
|
mismatch_count += 1
|
|
# If there's a significant difference, the frames are not the same
|
|
if mismatch_count > threshold:
|
|
return False
|
|
|
|
# If we reach here, the content appears to be the same
|
|
return True
|
|
|
|
|
|
def are_audio_files_similar(mp3_file_path, mp4_file_path):
|
|
# Extract audio fingerprint from MP3 file
|
|
mp3_fingerprint, mp3_duration = acoustid.fingerprint_file(mp3_file_path)
|
|
|
|
# Extract the audio stream from the MP4 file
|
|
mp4_audio_path = os.path.splitext(mp4_file_path)[0] + '_extracted.mp3'
|
|
try:
|
|
subprocess.run(["ffmpeg", "-i", mp4_file_path, "-vn", "-ar", "44100", "-ac", "2", "-ab", "192k", "-f", "mp3",
|
|
mp4_audio_path], check=True)
|
|
except subprocess.CalledProcessError as e:
|
|
print(f"An error occurred during audio extraction from MP4: {e}")
|
|
return False
|
|
|
|
# Extract audio fingerprint from the extracted audio
|
|
mp4_fingerprint, mp4_duration = acoustid.fingerprint_file(mp4_audio_path)
|
|
|
|
# Clean up temporary extracted audio file
|
|
os.remove(mp4_audio_path)
|
|
|
|
# Compare fingerprints (rudimentary comparison)
|
|
if mp3_duration >= mp4_duration and mp3_fingerprint == mp4_fingerprint:
|
|
return True
|
|
|
|
return False
|