videocut/cut_head.py

import subprocess
import json
import os
import glob
import sys

def get_video_duration(video_path):
    """Uses ffprobe to extract the exact duration of the baseline video."""
    print(f"Analyzing baseline duration: {video_path}...")
    cmd = [
        'ffprobe', '-v', 'error', '-show_entries', 'format=duration', 
        '-of', 'default=noprint_wrappers=1:nokey=1', video_path
    ]
    result = subprocess.run(cmd, capture_output=True, text=True)
    try:
        duration = float(result.stdout.strip())
        print(f"Baseline duration found: {duration} seconds.")
        return duration
    except ValueError:
        print("Error: Could not determine the duration of the baseline video.")
        return None

def get_next_keyframe(input_video, cut_timestamp):
    """Finds the next keyframe immediately following the cut timestamp."""
    cmd = [
        'ffprobe', '-v', 'quiet', '-select_streams', 'v',
        '-skip_frame', 'nokey', '-show_frames', 
        '-show_entries', 'frame=pkt_pts_time,pkt_dts_time,best_effort_timestamp_time',
        '-of', 'json', '-read_intervals', f'{cut_timestamp}%+180', input_video
    ]
    result = subprocess.run(cmd, capture_output=True, text=True)
    frames = json.loads(result.stdout).get('frames', [])
    
    for frame in frames:
        # Safely try multiple timestamp keys
        time_str = (frame.get('best_effort_timestamp_time') or 
                    frame.get('pkt_pts_time') or 
                    frame.get('pkt_dts_time'))
        
        if time_str is not None:
            keyframe_time = float(time_str)
            if keyframe_time > cut_timestamp:
                return keyframe_time
    return None

def smart_cut(input_video, output_video, cut_timestamp):
    """Executes the Smart Cut process: Re-encode a tiny segment, copy the rest, and stitch."""
    print(f"Processing: {input_video}...")
    next_keyframe = get_next_keyframe(input_video, cut_timestamp)
    
    if not next_keyframe:
        print(f"Skipping {input_video}: Could not find a keyframe after {cut_timestamp}s.")
        return

    part1 = f"temp_part1_{input_video}"
    part2 = f"temp_part2_{input_video}"
    concat_list = f"concat_{input_video}.txt"

    try:
        # 1. Re-encode the tiny segment (from exact cut to next keyframe)
        # Audio is removed (-an) to prevent sync/overlap issues during concatenation
        subprocess.run([
            'ffmpeg', '-y', '-v', 'error', '-i', input_video, 
            '-ss', str(cut_timestamp), '-to', str(next_keyframe), 
            '-c:v', 'libx264', '-crf', '18', '-an', part1
        ], check=True)
        
        # 2. Copy the rest of the video (from next keyframe to the end)
        # Audio is removed (-an) here too
        subprocess.run([
            'ffmpeg', '-y', '-v', 'error', '-ss', str(next_keyframe), '-i', input_video, 
            '-c:v', 'copy', '-an', part2
        ], check=True)
        
        # 3. Concatenate video parts and cleanly mux with the original extracted audio
        with open(concat_list, 'w', encoding='utf-8') as f:
            f.write(f"file '{part1}'\nfile '{part2}'\n")
            
        subprocess.run([
            'ffmpeg', '-y', '-v', 'error', 
            '-f', 'concat', '-safe', '0', '-i', concat_list, 
            '-ss', str(cut_timestamp), '-i', input_video,
            '-map', '0:v', '-map', '1:a?', 
            '-c:v', 'copy', '-c:a', 'copy', output_video
        ], check=True)
        print(f"Success! Saved to {output_video}")

    except subprocess.CalledProcessError as e:
        print(f"Error processing {input_video}. FFmpeg failed.")
    finally:
        # 4. Clean up temporary files safely
        for temp_file in [part1, part2, concat_list]:
            if os.path.exists(temp_file):
                os.remove(temp_file)

def main():
    # 1. Check if the user provided an argument
    if len(sys.argv) < 2:
        print("Usage: python cut.py <path_to_file_or_folder>")
        return
        
    # 2. Get the absolute path of the provided argument
    # 2. Get the absolute path and strip any accidental quotes
    raw_input = sys.argv[1].strip(' "''')
    input_arg = os.path.abspath(raw_input)
    # input_arg = os.path.abspath(sys.argv[1])
    print(input_arg)
    
    # 3. Determine if it's a file or a folder
    if os.path.isfile(input_arg):
        if not input_arg.lower().endswith('.mp4'):
            print("Error: The specified file is not an .mp4 video.")
            return
        target_dir = os.path.dirname(input_arg)
        target_files = [os.path.basename(input_arg)] # Process only this single file
        
    elif os.path.isdir(input_arg):
        target_dir = input_arg
        target_files = None # We will find all files later
        
    else:
        print(f"Error: Path '{input_arg}' does not exist.")
        return

    # 4. Change the working directory to the target folder
    # This prevents FFmpeg from crashing due to complex Windows path names during concatenation
    os.chdir(target_dir)

    # 5. If it was a folder, grab all .mp4 files
    if target_files is None:
        target_files = glob.glob("*.mp4")

    baseline_file = "i.mp4"

    # Check if baseline exists in the target directory
    if not os.path.exists(baseline_file):
        print(f"Error: '{baseline_file}' not found in the target folder: {target_dir}")
        return

    # Remove the baseline file from the processing list
    if baseline_file in target_files:
        target_files.remove(baseline_file)

    if not target_files:
        print("No other .mp4 files found to process.")
        return

    # Get exact cut timestamp from the baseline
    cut_timestamp = get_video_duration(baseline_file)
    if not cut_timestamp:
        return

    # Create output directory
    output_dir = "processed_videos"
    os.makedirs(output_dir, exist_ok=True)

    print(f"Found {len(target_files)} video(s) to process in '{target_dir}'.\n" + "-"*30)

    # Process each video
    for video in target_files:
        output_path = os.path.join(output_dir, video)
        smart_cut(video, output_path, cut_timestamp)

    print("-" * 30 + "\nBatch processing complete!")

if __name__ == "__main__":
    main()