main_improved.py

import os
import sys

# Suppress unnecessary logs before importing heavy libs
os.environ["ORT_LOGGING_LEVEL"] = "3" 
os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"

import warnings
warnings.filterwarnings("ignore")

import json
import shutil
import subprocess
import argparse
import time
from scripts import (
    download_video,
    transcribe_video,
    create_viral_segments,
    cut_segments,
    edit_video,
    transcribe_cuts,
    adjust_subtitles,
    burn_subtitles,
    save_json,
    organize_output,
    translate_json,
)
from i18n.i18n import I18nAuto

# Inicializa sistema de tradução
i18n = I18nAuto()
#
# Configurações de Legenda (ASS Style)
# Cores no formato BGR (Blue-Green-Red) para o ASS
COLORS = {
    "red": "0000FF",  # Red
    "yellow": "00FFFF",   # Yellow
    "green": "00FF00",     # Green
    "white": "FFFFFF",    # White
    "black": "000000",     # Black
    "grey": "808080",     # Grey
}

def get_subtitle_config(config_path=None):
    """

    Returns the subtitle configuration dictionary.

    Can be expanded to load from a JSON/YAML file in the future.

    """
    # Default Config
    base_color_transparency = "00"
    outline_transparency = "FF" 
    highlight_color_transparency = "00"
    shadow_color_transparency = "00"
    
    config = {
        "font": "Montserrat-Regular",
        "base_size": 30,
        "base_color": f"&H{base_color_transparency}{COLORS['white']}&",
        "highlight_size": 35,
        "words_per_block": 3,
        "gap_limit": 0.5,
        "mode": 'highlight', # Options: 'no_highlight', 'word_by_word', 'highlight'
        "highlight_color": f"&H{highlight_color_transparency}{COLORS['green']}&",
        "vertical_position": 210, # 1=170(top), ... 4=60(default)
        "alignment": 2, # 2=Center
        "bold": 0,
        "italic": 0,
        "underline": 0,
        "strikeout": 0,
        "border_style": 2, # 1=outline, 3=box
        "outline_thickness": 1.5,
        "outline_color": f"&H{outline_transparency}{COLORS['grey']}&",
        "shadow_size": 2,
        "shadow_color": f"&H{shadow_color_transparency}{COLORS['black']}&",
        "remove_punctuation": True,
    }

    if config_path and os.path.exists(config_path):
        try:
            with open(config_path, 'r', encoding='utf-8') as f:
                loaded_config = json.load(f)
                config.update(loaded_config)
                print(i18n("Loaded subtitle config from {}").format(config_path))
        except Exception as e:
            print(i18n("Error loading subtitle config: {}. Using defaults.").format(e))
    
    return config

def interactive_input_int(prompt_text):
    """Solicita um inteiro ao usuário via terminal."""
    while True:
        try:
            value = int(input(i18n(prompt_text)))
            if value > 0:
                return value
            print(i18n("\nError: Number must be greater than 0."))
        except ValueError:
            print(i18n("\nError: The value you entered is not an integer. Please try again."))

def main():
    # Configuração de Argumentos via Linha de Comando (CLI)
    parser = argparse.ArgumentParser(description="ViralCutter CLI")
    parser.add_argument("--url", help="YouTube Video URL")
    parser.add_argument("--segments", type=int, help="Number of segments to create")
    parser.add_argument("--viral", action="store_true", help="Enable viral mode")
    parser.add_argument("--themes", help="Comma-separated themes (if not viral mode)")
    parser.add_argument("--burn-only", action="store_true", help="Skip processing and only burn subtitles")
    parser.add_argument("--min-duration", type=int, default=15, help="Minimum segment duration (seconds)")
    parser.add_argument("--max-duration", type=int, default=90, help="Maximum segment duration (seconds)")
    parser.add_argument("--model", default="large-v3-turbo", help="Whisper model to use")
    
    parser.add_argument("--ai-backend", choices=["manual", "gemini", "g4f", "local"], help="AI backend for viral analysis")
    parser.add_argument("--api-key", help="Gemini API Key (required if ai-backend is gemini)")
    
    parser.add_argument("--chunk-size", help="Override Chunk Size")
    parser.add_argument("--ai-model-name", help="Override AI Model Name")

    parser.add_argument("--project-path", help="Path to existing project folder (overrides URL/Latest)")
    parser.add_argument("--workflow", choices=["1", "2", "3"], default="1", help="Workflow choice: 1=Full, 2=Cut Only, 3=Subtitles Only")
    parser.add_argument("--face-model", choices=["insightface", "mediapipe"], default="insightface", help="Face detection model")
    parser.add_argument("--face-mode", choices=["auto", "1", "2"], default="auto", help="Face tracking mode: auto, 1, 2")
    parser.add_argument("--subtitle-config", help="Path to subtitle configuration JSON file")
    parser.add_argument("--no-face-mode", choices=["padding", "zoom"], default="padding", help="Method to handle segments with no face detected: 'padding' (9:16 frame with black bars) or 'zoom' (Center Crop Zoom)")
    parser.add_argument("--face-detect-interval", type=str, default="0.17,1.0", help="Face detection interval in seconds. Single value or 'interval_1face,interval_2face'")
    parser.add_argument("--face-filter-threshold", type=float, default=0.35, help="Relative area threshold to ignore background faces (default: 0.35)")
    parser.add_argument("--face-two-threshold", type=float, default=0.60, help="Relative area threshold to trigger 2-face mode (default: 0.60)")
    parser.add_argument("--face-confidence-threshold", type=float, default=0.30, help="Face detection confidence threshold (0.0 - 1.0) (default: 0.30)")
    parser.add_argument("--face-dead-zone", type=str, default="40", help="Camera movement dead zone in pixels (default: 40)") # str to support future "auto"
    parser.add_argument("--focus-active-speaker", action="store_true", help="Enable experimental active speaker focus (InsightFace only)")
    parser.add_argument("--active-speaker-mar", type=float, default=0.03, help="Mouth Aspect Ratio threshold for active speaker (0.0 - 1.0) (default: 0.03)")
    parser.add_argument("--active-speaker-score-diff", type=float, default=1.5, help="Score difference to focus on active speaker (default: 1.5)")
    parser.add_argument("--include-motion", action="store_true", help="Include motion (body/head movement) in activity score")
    parser.add_argument("--active-speaker-motion-threshold", type=float, default=3.0, help="Motion deadzone in pixels (default: 3.0)")
    parser.add_argument("--active-speaker-motion-sensitivity", type=float, default=0.05, help="Motion sensitivity multiplier (default: 0.05)")
    parser.add_argument("--active-speaker-decay", type=float, default=2.0, help="Activity score decay rate (default: 2.0)")
    parser.add_argument("--skip-prompts", action="store_true", help="Skip interactive prompts and use defaults/existing files")
    parser.add_argument("--video-quality", choices=["best", "1080p", "720p", "480p"], default="best", help="Video download quality")
    parser.add_argument("--skip-youtube-subs", action="store_true", help="Skip downloading YouTube subtitles")
    parser.add_argument("--translate-target", help="Target language code for subtitle translation (e.g. 'pt', 'en').")

    args = parser.parse_args()
    
    # Workflow Logic
    workflow_choice = args.workflow
    
    # If Subtitles Only, checking project path
    if workflow_choice == "3" and not args.project_path and not args.url and not args.skip_prompts:
        # Prompt for project path or use latest if not provided?
        pass # Will handle in main flow

    # Modo Apenas Queimar Legenda (Legacy support, mapped to Workflow 3 internally if burn-only is set)
    # Verifica o argumento CLI ou uma variável local hardcoded (para compatibilidade)
    burn_only_mode = args.burn_only

    if burn_only_mode:
        print(i18n("Burn only mode activated. Switching to Workflow 3..."))
        workflow_choice = "3"

    # Obtenção de Inputs (CLI ou Interativo)
    url = args.url
    project_path_arg = args.project_path
    input_video = None

    # Se project_path for fornecido, ignoramos URL
    if project_path_arg:
        if os.path.exists(project_path_arg):
             print(i18n("Using provided project path: {}").format(project_path_arg))
             # Tentar achar o input.mp4 pra manter compatibilidade de variaveis, embora Workflow 3 não precise de download
             possible_input = os.path.join(project_path_arg, "input.mp4")
             if os.path.exists(possible_input):
                 input_video = possible_input
             else:
                 # Se não tiver input.mp4, tudo bem para workflow 3, mas definimos um dummy para não quebrar logica
                 input_video = os.path.join(project_path_arg, "dummy_input.mp4")
             
             # Se for workflow 3, não precisamos de URL
        else:
             print(i18n("Error: Provided project path does not exist."))
             sys.exit(1)

    # Se não temos URL via CLI nem Project Path, pedimos agora
    if not url and not project_path_arg:
        if args.skip_prompts:
             print(i18n("No URL provided and skipping prompts. Trying to load latest project..."))
             # Fallthrough to project loading logic
        else:
            user_input = input(i18n("Enter the YouTube video URL (or press Enter to use latest project): ")).strip()
            if user_input:
                url = user_input
    
    if not url and not input_video:
        # Usuário apertou Enter (Vazio) -> Tentar pegar último projeto
        base_virals = "VIRALS"
        if os.path.exists(base_virals):
            subdirs = [os.path.join(base_virals, d) for d in os.listdir(base_virals) if os.path.isdir(os.path.join(base_virals, d))]
            if subdirs:
                latest_project = max(subdirs, key=os.path.getmtime)
                detected_video = os.path.join(latest_project, "input.mp4")
                if os.path.exists(detected_video):
                    input_video = detected_video
                    print(i18n("Using latest project: {}").format(latest_project))
                else:
                    print(i18n("Latest project found but 'input.mp4' is missing."))
                    sys.exit(1)
            else:
                print(i18n("No existing projects found in VIRALS folder."))
                sys.exit(1)
        else:
             print(i18n("VIRALS folder not found. Cannot load latest project."))
             sys.exit(1)

    # -------------------------------------------------------------------------
    # Checagem Antecipada de Segmentos Virais (Para pular configurações se já existirem)
    # -------------------------------------------------------------------------
    viral_segments = None
    project_folder_anticipated = None

    if input_video:
        # Se já temos o vídeo, podemos deduzir a pasta
        project_folder_anticipated = os.path.dirname(input_video)
        viral_segments_file = os.path.join(project_folder_anticipated, "viral_segments.txt")
        
        if os.path.exists(viral_segments_file):
             print(i18n("\nExisting viral segments found: {}").format(viral_segments_file))
             if args.skip_prompts:
                 use_existing_json = 'yes'
             else:
                 use_existing_json = input(i18n("Use existing viral segments? (yes/no) [default: yes]: ")).strip().lower()
             
             if use_existing_json in ['', 'y', 'yes']:
                try:
                    with open(viral_segments_file, 'r', encoding='utf-8') as f:
                        viral_segments = json.load(f)
                    print(i18n("Loaded existing viral segments. Skipping configuration prompts."))
                    if viral_segments and "segments" in viral_segments:
                        print(f"DEBUG: Loaded {len(viral_segments['segments'])} segments from file.")
                    else:
                        print("DEBUG: Loaded JSON but 'segments' key is missing or empty.")
                except Exception as e:
                    print(i18n("Error loading JSON: {}.").format(e))

    # Variaveis de config de IA (só necessárias se não tivermos os segmentos)
    num_segments = None
    viral_mode = False
    themes = ""
    ai_backend = "manual" # default
    api_key = None
    
    if not viral_segments:
        num_segments = args.segments
        if not num_segments:
            if args.skip_prompts:
                print(i18n("No segments count provided and skip-prompts is ON. Using default 3."))
                num_segments = 3
            else:
                num_segments = interactive_input_int("Enter the number of viral segments to create: ")

        viral_mode = args.viral
        if not args.viral and not args.themes:
            if args.skip_prompts:
                print(i18n("Viral mode not set, defaulting to True."))
                viral_mode = True
            else:
                response = input(i18n("Do you want viral mode? (yes/no): ")).lower()
                viral_mode = response in ['yes', 'y']
        
        themes = args.themes if args.themes else ""
        if not viral_mode and not themes:
            if not args.skip_prompts:
                 themes = input(i18n("Enter themes (comma-separated, leave blank if viral mode is True): "))

        # Duration Config
        print(i18n("\nCurrent duration settings: {}s - {}s").format(args.min_duration, args.max_duration))
        if not args.skip_prompts:
            change_dur = input(i18n("Change duration? (y/n) [default: n]: ")).strip().lower()
            if change_dur in ['y', 'yes']:
                 try:
                     min_d = input(i18n("Minimum duration [{}]: ").format(args.min_duration)).strip()
                     if min_d: args.min_duration = int(min_d)
                     
                     max_d = input(i18n("Maximum duration [{}]: ").format(args.max_duration)).strip()
                     if max_d: args.max_duration = int(max_d)
                 except ValueError:
                     print(i18n("Invalid number. Using previous values."))

        # Load API Config
        config_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'api_config.json')
        api_config = {}
        if os.path.exists(config_path):
            try:
                with open(config_path, 'r', encoding='utf-8') as f:
                    api_config = json.load(f)
            except:
                pass

        # Seleção do Backend de IA
        ai_backend = args.ai_backend
        
        # Try to load backend from config if not in args
        if not ai_backend and api_config.get("selected_api"):
            ai_backend = api_config.get("selected_api")
            print(i18n("Using AI Backend from config: {}").format(ai_backend))

        if not ai_backend:
            if args.skip_prompts:
                print(i18n("No AI backend selected, defaulting to Manual."))
                ai_backend = "manual"
            else:
                print("\n" + i18n("Select AI Backend for Viral Analysis:"))
                print(i18n("1. Gemini API (Best / Recommended)"))
                print(i18n("2. G4F (Free / Experimental)"))
                print(i18n("3. Local (GGUF via llama.cpp)"))
                print(i18n("4. Manual (Copy/Paste Prompt)"))
                choice = input(i18n("Choose (1-4): ")).strip()
                
                if choice == "1":
                    ai_backend = "gemini"
                elif choice == "2":
                    ai_backend = "g4f"
                elif choice == "3":
                    ai_backend = "local"
                    # Interactive model selection for local
                    # List models
                    models_dir = os.path.join(os.path.dirname(os.path.abspath(__file__)), "models")
                    if not os.path.exists(models_dir): os.makedirs(models_dir)
                    models = [f for f in os.listdir(models_dir) if f.endswith(".gguf")]
                    
                    if not models:
                        print(i18n("\nNo .gguf models found in 'models' directory."))
                        print(i18n("Please place a module file in: {}").format(models_dir))
                        print(i18n("Falling back to Manual..."))
                        ai_backend = "manual"
                    else:
                        print(i18n("\nAvailable Models:"))
                        for idx, m in enumerate(models):
                            print(f"{idx+1}. {m}")
                        
                        try:
                            m_idx = int(input(i18n("Select Model (Number): "))) - 1
                            if 0 <= m_idx < len(models):
                                args.ai_model_name = models[m_idx] # Set global arg
                            else:
                                print(i18n("Invalid selection. Using first model."))
                                args.ai_model_name = models[0]
                        except:
                             print(i18n("Invalid input. Using first model."))
                             args.ai_model_name = models[0]
                             
                else:
                    ai_backend = "manual"

        api_key = args.api_key
        # Check config for API Key if using Gemini
        if ai_backend == "gemini" and not api_key:
            cfg_key = api_config.get("gemini", {}).get("api_key", "")
            if cfg_key and cfg_key != "SUA_KEY_AQUI":
                api_key = cfg_key
        
        if ai_backend == "gemini" and not api_key:
             if args.skip_prompts:
                 print(i18n("Gemini API key missing, but skip-prompts is ON. Might fail."))
             else:
                 print(i18n("Gemini API Key not found in api_config.json or arguments."))
                 api_key = input(i18n("Enter your Gemini API Key: ")).strip()

    # Workflow & Face Config Inputs
    workflow_choice = args.workflow
    face_model = args.face_model
    face_mode = args.face_mode

    # If args weren't provided and we are not skipping prompts, ask user
    # Note: argparse defaults are set, so they "are provided" effectively.
    # To truly detect "not provided", request default=None in argparse. 
    # But for "Simplified Mode", defaults are good.
    # Advanced users use params.
    # We will assume CLI defaults are what we want if skip_prompts is on.
    
    # Logic for detection intervals (Moved out of interactive block to support CLI/WebUI)
    detection_intervals = None
    if args.face_detect_interval:
        try:
            parts = args.face_detect_interval.split(',')
            if len(parts) == 1:
                val = float(parts[0])
                detection_intervals = {'1': val, '2': val}
            elif len(parts) >= 2:
                val1 = float(parts[0])
                val2 = float(parts[1])
                detection_intervals = {'1': val1, '2': val2}
        except ValueError:
            pass

    if not args.burn_only and not args.skip_prompts:
        # Interactive Face Config
        print(i18n("\n--- Face Detection Settings ---"))
        print(i18n("Current Face Model: {} | Mode: {}").format(face_model, face_mode))
        
        if detection_intervals:
             print(i18n("Custom detection intervals: {}").format(detection_intervals))
        else:
             print(i18n("Using dynamic intervals: 1s for 2-face, ~0.16s for 1-face."))


    # Pipeline Execution
    try:
        # 1. Download & Project Setup
        print(f"DEBUG: Checking input_video state. input_video={input_video}")
        
        if not input_video:
            if not url:
                print(i18n("Error: No URL provided and no existing video selected."))
                sys.exit(1)
                
            print(i18n("Starting download..."))
            download_subs = not args.skip_youtube_subs
            download_result = download_video.download(url, download_subs=download_subs, quality=args.video_quality)
            
            if isinstance(download_result, tuple):
                input_video, project_folder = download_result
            else:
                input_video = download_result
                project_folder = os.path.dirname(input_video)
                
            print(f"DEBUG: Download finished. input_video={input_video}, project_folder={project_folder}")
            
        else:
            # Reuso de video existente
            print("DEBUG: Using existing video logic.")
            project_folder = os.path.dirname(input_video)
            
        print(f"Project Folder: {project_folder}")
        
        # 2. Transcribe
        if workflow_choice == "3":
            print(i18n("Workflow 3: Skipping Transcribe."))
            # We assume transcription exists (SRT/JSON) or we won't need it for 'adjust_subtitles' if it uses 'subs/*.json' which are created by 'cut_segments'
            # Actually 'adjust_subtitles' reads from 'project_folder/subs'.
            # viral_segments = True # Removed to avoid overwritting dict loaded earlier
        else:
            print(i18n("Transcribing with model {}...").format(args.model))
            # Se skip config, args.model é default
            srt_file, tsv_file = transcribe_video.transcribe(input_video, args.model, project_folder=project_folder)
 
        # 3. Create Viral Segments
        if workflow_choice != "3":
            # Se não carregamos 'viral_segments' lá em cima (ou se era download novo), checamos agora ou criamos
            if not viral_segments:
                # Checagem tardia para downloads novos que por acaso ja tenham json (Ex: URL repetida)
                viral_segments_file_late = os.path.join(project_folder, "viral_segments.txt")
                if os.path.exists(viral_segments_file_late):
                    print(i18n("Found existing viral segments file at {}").format(viral_segments_file_late))
                    if args.skip_prompts:
                        print(i18n("Skipping prompts enabled. Loading existing segments."))
                        try:
                            with open(viral_segments_file_late, 'r', encoding='utf-8') as f:
                                viral_segments = json.load(f)
                        except Exception as e:
                            print(i18n("Error loading existing JSON: {}. Proceeding to create new segments.").format(e))
                    else:
                        print(i18n("Loading existing viral segments found at {}").format(viral_segments_file_late))
                        try:
                            with open(viral_segments_file_late, 'r', encoding='utf-8') as f:
                                viral_segments = json.load(f)
                        except Exception as e:
                            print(i18n("Error loading existing JSON: {}.").format(e))
                    
                if not viral_segments:
                    print(i18n("Creating viral segments using {}...").format(ai_backend.upper()))
                    viral_segments = create_viral_segments.create(
                        num_segments, 
                        viral_mode, 
                        themes, 
                        args.min_duration, 
                        args.max_duration,
                        ai_mode=ai_backend,
                        api_key=api_key,
                        project_folder=project_folder,
                        chunk_size_arg=args.chunk_size,
                        model_name_arg=args.ai_model_name
                    )
                
                if not viral_segments or not viral_segments.get("segments"):
                    print(i18n("Error: No viral segments were generated."))
                    print(i18n("Possible reasons: API error, Model not found, or empty response."))
                    print(i18n("Stopping execution."))
                    sys.exit(1)
                
                save_json.save_viral_segments(viral_segments, project_folder=project_folder) 

        # 3.5. Fix Raw Segments (missing timestamps)
        if workflow_choice != "3" and viral_segments and "segments" in viral_segments:
            segs = viral_segments.get("segments", [])
            if segs and len(segs) > 0:
                 # Check first segment for duration 0 but having start_time_ref or just check duration
                 first = segs[0]
                 # If duration is effectively 0 and we have a ref tag (or even if we dont, we cant cut 0s video)
                 # We assume if duration is 0, it is raw.
                 if first.get("duration", 0) == 0:
                      print(i18n("Detected raw AI segments without timestamps (Duration 0). Running alignment..."))
                      try:
                          # Load transcript
                          transcript = create_viral_segments.load_transcript(project_folder)
                          # Process (Align)
                          # Use None for output_count to keep all found segments
                          viral_segments = create_viral_segments.process_segments(
                              segs, 
                              transcript, 
                              args.min_duration, 
                              args.max_duration, 
                              output_count=None 
                          )
                          save_json.save_viral_segments(viral_segments, project_folder=project_folder)
                          print(i18n("Segments aligned and saved."))
                      except Exception as e:
                          print(i18n("Failed to align raw segments: {}").format(e))
                          # If alignment fails, it might crash later, but we tried. 

        # 4. Cut Segments
        # Se workflow for 3, pulamos corte
        if workflow_choice == "3":
            print(i18n("Workflow 3 (Subtitles Only): Skipping Cut and Edit."))
            # Deduzir cuts folder apenas para log
            cuts_folder = os.path.join(project_folder, "cuts")
        else:
            cuts_folder = os.path.join(project_folder, "cuts")
            skip_cutting = False
            
            if os.path.exists(cuts_folder) and os.listdir(cuts_folder):
                print(i18n("\nExisting cuts found in: {}").format(cuts_folder))
                if args.skip_prompts:
                    cut_again_resp = 'no'
                else:
                    cut_again_resp = input(i18n("Cuts already exist. Cut again? (yes/no) [default: no]: ")).strip().lower()
                
                # Default is no (skip) if they just press enter or say no
                if cut_again_resp not in ['y', 'yes']:
                    skip_cutting = True
            
            if skip_cutting:
                print(i18n("Skipping Video Rendering (using existing cuts), but updating Subtitle JSONs..."))
            else:
                print(i18n("Cutting segments..."))

            cut_segments.cut(viral_segments, project_folder=project_folder, skip_video=skip_cutting)
        
        # 5. Workflow Check
        if workflow_choice == "2":
            print(i18n("Cut Only selected. Skipping Face Crop and Subtitles."))
            print(i18n(f"Process completed! Check your results in: {project_folder}"))
            sys.exit(0)

        # 5. Edit Video (Face Crop)
        if workflow_choice != "3":
            print(i18n("Editing video with {} (Mode: {})...").format(face_model, face_mode))
            
            # Parse dead zone safely
            try:
                dead_zone_val = float(args.face_dead_zone)
            except:
                dead_zone_val = 40.0
                
            edit_video.edit(
                project_folder=project_folder, 
                face_model=face_model, 
                face_mode=face_mode, 
                detection_period=detection_intervals,
                filter_threshold=args.face_filter_threshold,
                two_face_threshold=args.face_two_threshold,
                confidence_threshold=args.face_confidence_threshold,
                dead_zone=dead_zone_val,
                focus_active_speaker=args.focus_active_speaker,
                active_speaker_mar=args.active_speaker_mar,
                active_speaker_score_diff=args.active_speaker_score_diff,
                include_motion=args.include_motion,
                active_speaker_motion_deadzone=args.active_speaker_motion_threshold,
                active_speaker_motion_sensitivity=args.active_speaker_motion_sensitivity,
                active_speaker_decay=args.active_speaker_decay,
                segments_data=viral_segments.get("segments", []) if viral_segments else None,
                no_face_mode=args.no_face_mode
            )


        else:
            print(i18n("Workflow 3: Skipping Face Crop."))
            # Rename existing files if viral_segments available (since edit_video didn't run)
            if viral_segments and "segments" in viral_segments:
                 segments_data = viral_segments.get("segments", [])
                 final_folder = os.path.join(project_folder, "final")
                 subs_folder = os.path.join(project_folder, "subs")
                 
                 print(i18n("Renaming existing files with titles..."))
                 for idx, segment in enumerate(segments_data):
                     title = segment.get("title", f"Segment_{idx}")
                     safe_title = "".join([c for c in title if c.isalnum() or c in " _-"]).strip()
                     safe_title = safe_title.replace(" ", "_")[:60]
                     
                     new_base_name = f"{idx:03d}_{safe_title}"
                     
                     # 1. MP4
                     old_mp4_name = f"final-output{idx:03d}_processed.mp4"
                     old_mp4_path = os.path.join(final_folder, old_mp4_name)
                     new_mp4_path = os.path.join(final_folder, f"{new_base_name}.mp4")
                     if os.path.exists(old_mp4_path) and not os.path.exists(new_mp4_path):
                         os.rename(old_mp4_path, new_mp4_path)
                         print(f"Renamed (Workflow 3): {old_mp4_name} -> {new_base_name}.mp4")

                     # 2. JSON Sub
                     old_json_name = f"final-output{idx:03d}_processed.json"
                     old_json_path = os.path.join(subs_folder, old_json_name)
                     new_json_path = os.path.join(subs_folder, f"{new_base_name}_processed.json")
                     if os.path.exists(old_json_path) and not os.path.exists(new_json_path):
                         os.rename(old_json_path, new_json_path)
                         print(f"Renamed (Workflow 3): {old_json_name} -> {new_base_name}_processed.json")
                         
                     # 3. Timeline
                     old_tl_name = f"temp_video_no_audio_{idx}_timeline.json"
                     old_tl_path = os.path.join(final_folder, old_tl_name)
                     new_tl_path = os.path.join(final_folder, f"{new_base_name}_timeline.json")
                     if os.path.exists(old_tl_path) and not os.path.exists(new_tl_path):
                         os.rename(old_tl_path, new_tl_path)
                         print(f"Renamed (Workflow 3): {old_tl_name} -> {new_base_name}_timeline.json")

        # 6. Subtitles
        burn_subtitles_option = True 
        if burn_subtitles_option:
            print(i18n("Processing subtitles..."))
            # transcribe_cuts removido: JSON de legenda já é gerado no corte
            # transcribe_cuts.transcribe(project_folder=project_folder)
            
            # --- Translation Integration ---
            if args.translate_target and args.translate_target.lower() != "none":
                 print(i18n("Translating subtitles to: {}").format(args.translate_target))
                 import asyncio
                 try:
                    asyncio.run(translate_json.translate_project_subs(project_folder, args.translate_target))
                 except Exception as e:
                    print(i18n("Translation failed: {}").format(e))
            # -------------------------------

            sub_config = get_subtitle_config(args.subtitle_config)
            

            
            # Passa o dicionário desempacotado como argumentos, mais o project_folder
            try:
                adjust_subtitles.adjust(project_folder=project_folder, **sub_config)
                burn_subtitles.burn(project_folder=project_folder)
            except FileNotFoundError as fnf_error:
                print(i18n("\n[ERROR] Subtitle processing failed: {}").format(str(fnf_error)))
                print(i18n("Tip: If you are using Workflow 3 (Subtitles Only), ensure the 'subs' folder exists and contains valid JSON files."))
                sys.exit(1)
            except Exception as e:
                print(i18n("\n[ERROR] Unexpected error during subtitle processing: {}").format(str(e)))
                raise e
        else:
            print(i18n("Subtitle burning skipped."))

        # Organização Final (Opcional, pois agora já está tudo em project_folder)
        # organize_output.organize(project_folder=project_folder)
        
        # --- Save Processing Configuration ---
        try:
            # Determine AI Model used
            used_ai_model = args.ai_model_name
            if not used_ai_model and ai_backend != "manual":
                if ai_backend == "gemini":
                    used_ai_model = api_config.get("gemini", {}).get("model", "default")
                elif ai_backend == "g4f":
                    used_ai_model = api_config.get("g4f", {}).get("model", "default")
            
            # Ensure sub_config exists
            current_sub_config = sub_config if 'sub_config' in locals() else get_subtitle_config(args.subtitle_config)
            
            final_config = {
                "timestamp": time.strftime("%Y-%m-%d %H:%M:%S"),
                "workflow": workflow_choice,
                "ai_config": {
                    "backend": ai_backend,
                    "model_name": used_ai_model,
                    "viral_mode": viral_mode,
                    "themes": themes,
                    "num_segments": num_segments,
                    "chunk_size": args.chunk_size
                },
                "face_config": {
                    "model": face_model,
                    "mode": face_mode,
                    "detect_interval": args.face_detect_interval,
                    "filter_threshold": args.face_filter_threshold,
                    "two_face_threshold": args.face_two_threshold,
                    "confidence_threshold": args.face_confidence_threshold,
                    "dead_zone": args.face_dead_zone,
                    "focus_active_speaker": args.focus_active_speaker,
                    "active_speaker_mar": args.active_speaker_mar,
                    "active_speaker_score_diff": args.active_speaker_score_diff,
                    "include_motion": args.include_motion
                },
                "video_config": {
                    "min_duration": args.min_duration,
                    "max_duration": args.max_duration,
                    "whisper_model": args.model
                },
                "subtitle_config": current_sub_config
            }

            config_save_path = os.path.join(project_folder, "process_config.json")
            with open(config_save_path, "w", encoding="utf-8") as f:
                json.dump(final_config, f, indent=4, ensure_ascii=False)
            print(i18n("Configuration saved to: {}").format(config_save_path))
            
        except Exception as e:
            print(i18n("Error saving configuration JSON: {}").format(e))
        # -------------------------------------

        print(i18n("Process completed! Check your results in: {}").format(project_folder))

    except Exception as e:
        print(i18n("\nAn error occurred: {}").format(str(e)))
        import traceback
        traceback.print_exc()
        sys.exit(1)

if __name__ == "__main__":
    main()