ass_generator.py

import os
from styles import STYLES

def seconds_to_ass_time(seconds):
    """Converts seconds (float) to 0:00:00.00 format"""
    mins, secs = divmod(seconds, 60)
    hours, mins = divmod(mins, 60)
    cs = int((secs - int(secs)) * 100)
    return f"{int(hours)}:{int(mins):02d}:{int(secs):02d}.{cs:02d}"

def generate_ass(transcript, style_name="hormozi", output_file="subs.ass"):
    """
    Generates an .ass file from a transcript list.
    transcript: [{'text': 'Hello', 'start': 0.0, 'end': 0.5}, ...]
    """
    config = STYLES.get(style_name, STYLES["hormozi"])
    
    header = f"""[Script Info]
ScriptType: v4.00+
PlayResX: 1280
PlayResY: 720

[V4+ Styles]
Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
Style: Default,{config['Fontname']},{config['Fontsize']},{config['PrimaryColour']},{config['PrimaryColour']},{config['OutlineColour']},{config['BackColour']},{config['Bold']},{config['Italic']},0,0,100,100,0,0,{config['BorderStyle']},{config['Outline']},{config['Shadow']},{config['Alignment']},{config['MarginL']},{config['MarginR']},{config['MarginV']},1

[Events]
Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
"""
    
    # We need to group words into "Phrases" (Lines) to avoid flickering
    # For simplicity, we will put roughly 3-4 words per line, or use the provided breaks if advanced.
    # Here we assume the transcript IS the phrasing (chunks of words).
    # If transcript is just words, we might want to group them.
    
    # Let's assume input is a list of WORDS for "Karaoke" style effect.
    # We will group them into chunks of ~1.5 - 2 seconds.
    
    lines = []
    current_chunk = []
    chunk_start = 0.0
    chunk_duration = 0.0
    
    events_str = ""
    
    # Simple Logic: 1 line = 1 transcript item (assuming transcript handles grouping)
    # OR Logic: Input is single words, we must group.
    
    # Let's code for: Input is BLOCKS of words (Phrases), loop through them.
    # If the user sends single words, this will visualize strictly one word at a time.
    
    for item in transcript:
        start = seconds_to_ass_time(item['start'])
        end = seconds_to_ass_time(item['end'])
        
        # Apply Active Style
        # If the text has multiple words, we treat the whole block as 'Active'
        # To do word-by-word karaoke, we'd need nested logic.
        
        text = item['text']
        style_code = config['active_anim']  # Apply the "active" look to this chunk
        
        # Simple line:
        line_str = f"Dialogue: 0,{start},{end},Default,,0,0,0,,{style_code}{text}"
        events_str += line_str + "\n"

    with open(output_file, "w", encoding="utf-8") as f:
        f.write(header + events_str)
    
    return output_file