RO-Rtechs's picture
Update app.py
bf432bc verified
raw
history blame
9.8 kB
import gradio as gr
from gradio import utils
import os
import re
import requests
from concurrent.futures import ThreadPoolExecutor
import time
from yt_dlp import YoutubeDL
import subprocess
import shutil
from typing import List, Tuple
def sanitize_title(title):
return re.sub(r'[\\/*?:"<>|]', "", title)
def format_time(seconds):
return time.strftime('%H:%M:%S', time.gmtime(seconds))
def get_video_info(video_url):
with YoutubeDL({'quiet': True, 'no_warnings': True}) as ydl:
try:
info = ydl.extract_info(video_url, download=False)
formats = info.get('formats', [])
# Function to safely get bitrate
def get_bitrate(format_dict, key):
return format_dict.get(key, 0) or 0
# Prefer adaptive formats (separate video and audio)
video_formats = [f for f in formats if f.get('vcodec') != 'none' and f.get('acodec') == 'none']
audio_formats = [f for f in formats if f.get('acodec') != 'none' and f.get('vcodec') == 'none']
if video_formats and audio_formats:
video_format = max(video_formats, key=lambda f: get_bitrate(f, 'vbr'))
audio_format = max(audio_formats, key=lambda f: get_bitrate(f, 'abr'))
return info['title'], video_format['url'], audio_format['url']
else:
# Fallback to best combined format
combined_formats = [f for f in formats if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
if combined_formats:
best_format = max(combined_formats, key=lambda f: get_bitrate(f, 'tbr'))
return info['title'], best_format['url'], None
else:
raise Exception("No suitable video formats found")
except Exception as e:
raise Exception(f"Error extracting video info: {str(e)}")
def download_segment(url, start_time, end_time, output_path):
command = [
'ffmpeg',
'-ss', format_time(start_time),
'-i', url,
'-t', format_time(end_time - start_time),
'-c', 'copy',
'-y',
output_path
]
process = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, universal_newlines=True)
while True:
output = process.stderr.readline()
if output == '' and process.poll() is not None:
break
if output:
yield output.strip()
rc = process.poll()
return rc == 0
def combine_segments(video_segments, audio_segments, output_path):
temp_video = 'temp_video.mp4'
temp_audio = 'temp_audio.m4a'
# Concatenate video segments
with open('video_list.txt', 'w') as f:
for segment in video_segments:
f.write(f"file '{segment}'\n")
subprocess.run(['ffmpeg', '-f', 'concat', '-safe', '0', '-i', 'video_list.txt', '-c', 'copy', temp_video])
# Concatenate audio segments if they exist
if audio_segments:
with open('audio_list.txt', 'w') as f:
for segment in audio_segments:
f.write(f"file '{segment}'\n")
subprocess.run(['ffmpeg', '-f', 'concat', '-safe', '0', '-i', 'audio_list.txt', '-c', 'copy', temp_audio])
# Combine video and audio
subprocess.run(['ffmpeg', '-i', temp_video, '-i', temp_audio, '-c', 'copy', output_path])
else:
shutil.move(temp_video, output_path)
# Clean up temporary files
os.remove('video_list.txt')
if os.path.exists('audio_list.txt'):
os.remove('audio_list.txt')
if os.path.exists(temp_video):
os.remove(temp_video)
if os.path.exists(temp_audio):
os.remove(temp_audio)
def add_segment(start_hours, start_minutes, start_seconds, end_hours, end_minutes, end_seconds, segments):
start_time = f"{start_hours:02d}:{start_minutes:02d}:{start_seconds:02d}"
end_time = f"{end_hours:02d}:{end_minutes:02d}:{end_seconds:02d}"
new_segment = f"{start_time}-{end_time}"
new_row = [new_segment]
return segments + [new_row]
def remove_segment(segments, index):
return segments[:index] + segments[index+1:]
def move_segment(segments, old_index, new_index):
segments_list = segments.values.tolist() # Convert Dataframe to list
if 0 <= old_index < len(segments_list) and 0 <= new_index < len(segments_list):
segment = segments_list.pop(old_index)
segments_list.insert(new_index, segment)
return segments_list
def parse_segments(segments: List[str]) -> List[Tuple[int, int]]:
parsed_segments = []
for segment in segments:
start, end = map(lambda x: sum(int(i) * 60 ** j for j, i in enumerate(reversed(x.split(':')))), segment.split('-'))
if start < end:
parsed_segments.append((start, end))
return parsed_segments
def process_video(video_url, segments, combine, progress=gr.Progress()):
if not video_url.strip():
return 0, "Error: Please provide a valid YouTube URL", None
# Extract segments from the Dataframe
segment_list = [segment[0] for segment in segments if segment[0].strip()]
parsed_segments = parse_segments(segment_list)
if not parsed_segments:
return 0, "Error: No valid segments provided", None
output_dir = 'output'
os.makedirs(output_dir, exist_ok=True)
try:
video_title, video_url, audio_url = get_video_info(video_url)
except Exception as e:
return 0, f"Error: {str(e)}", None
video_segments = []
audio_segments = []
total_segments = len(parsed_segments)
for i, (start_time, end_time) in enumerate(parsed_segments):
video_output = os.path.join(output_dir, f"{sanitize_title(video_title)}_video_segment_{i+1}.mp4")
for output in download_segment(video_url, start_time, end_time, video_output):
progress((i / total_segments) + (1 / total_segments) * 0.5)
yield i * 100 // total_segments, f"Downloading video segment {i+1}/{total_segments}: {output}", None
video_segments.append(video_output)
if audio_url:
audio_output = os.path.join(output_dir, f"{sanitize_title(video_title)}_audio_segment_{i+1}.m4a")
for output in download_segment(audio_url, start_time, end_time, audio_output):
progress((i / total_segments) + (1 / total_segments) * 0.75)
yield i * 100 // total_segments + 50, f"Downloading audio segment {i+1}/{total_segments}: {output}", None
audio_segments.append(audio_output)
if combine:
output_path = os.path.join(output_dir, f"{sanitize_title(video_title)}_combined.mp4")
combine_segments(video_segments, audio_segments, output_path)
yield 100, f"Segments combined and saved as {output_path}", output_path
else:
# If not combining, return the first video segment (you might want to modify this behavior)
output_path = video_segments[0] if video_segments else None
yield 100, "All segments downloaded successfully", output_path
# Clean up individual segments if combined
if combine:
for segment in video_segments + audio_segments:
os.remove(segment)
# Disable Gradio analytics
utils.colab_check = lambda: True
with gr.Blocks(title="Advanced YouTube Segment Downloader", theme=gr.themes.Soft()) as iface:
gr.Markdown("## Advanced YouTube Segment Downloader")
gr.Markdown("Download segments of YouTube videos using adaptive streaming and ffmpeg, with optional combining.")
with gr.Row():
video_url = gr.Textbox(label="YouTube URL", placeholder="Enter YouTube URL here")
with gr.Row():
with gr.Column(scale=1):
with gr.Row():
start_hours = gr.Number(label="Start Hours", minimum=0, maximum=23, step=1, value=0)
start_minutes = gr.Number(label="Start Minutes", minimum=0, maximum=59, step=1, value=0)
start_seconds = gr.Number(label="Start Seconds", minimum=0, maximum=59, step=1, value=0)
with gr.Row():
end_hours = gr.Number(label="End Hours", minimum=0, maximum=23, step=1, value=0)
end_minutes = gr.Number(label="End Minutes", minimum=0, maximum=59, step=1, value=0)
end_seconds = gr.Number(label="End Seconds", minimum=0, maximum=59, step=1, value=0)
add_btn = gr.Button("Add Segment")
with gr.Column(scale=2):
segments = gr.Dataframe(
headers=["Segment"],
row_count=5,
col_count=1,
interactive=True,
label="Segments"
)
combine = gr.Checkbox(label="Combine Segments")
submit_btn = gr.Button("Download Segments", variant="primary")
progress = gr.Slider(label="Progress", minimum=0, maximum=100, step=1)
status = gr.Textbox(label="Status", lines=10)
output_file = gr.File(label="Download Video")
add_btn.click(
add_segment,
inputs=[start_hours, start_minutes, start_seconds, end_hours, end_minutes, end_seconds, segments],
outputs=[segments]
)
submit_btn.click(
process_video,
inputs=[video_url, segments, combine],
outputs=[progress, status, output_file]
)
segments.change(
move_segment,
inputs=[segments, gr.Slider(0, 100, step=1, label="Old Index"), gr.Slider(0, 100, step=1, label="New Index")],
outputs=[segments]
)
remove_btn = gr.Button("Remove Selected Segment")
remove_btn.click(
remove_segment,
inputs=[segments, gr.Slider(0, 100, step=1, label="Index to Remove")],
outputs=[segments]
)
iface.launch()