utils.py

import re
import math

def group_sentences(text, target_sec=10):
    raw_sentences = re.split(r'(?<=[.?!])\s+', text.strip())
    scenes = []
    curr_text, curr_time = "", 0
    for s in raw_sentences:
        if not s: continue
        s_time = max(5, math.ceil(len(s)/35)*5)
        if curr_time + s_time > target_sec and curr_text:
            scenes.append({"text": curr_text, "time": curr_time})
            curr_text, curr_time = s, s_time
        else:
            curr_text = (curr_text + " " + s).strip()
            curr_time += s_time
    if curr_text: scenes.append({"text": curr_text, "time": curr_time})
    return scenes

def get_smart_filename(index, text):
    clean_text = re.sub(r'[\\/*?:"<>|]', "", text).strip()
    name_part = clean_text[:12] if len(clean_text) <= 12 else f"{clean_text[:6]}~{clean_text[-4:]}"
    return f"{index+1:02d}_{name_part}.png"

def make_scene_snippet(text, head_len=10, tail_len=10, max_len=60):
    clean_text = " ".join((text or "").strip().split())
    clean_text = re.sub(r'[\\/*?:"<>|]', "", clean_text)
    clean_text = re.sub(r"[^0-9A-Za-z가-힣\s.~]", " ", clean_text)
    clean_text = re.sub(r"[.]{2,}", ".", clean_text)
    clean_text = re.sub(r"\s+", " ", clean_text).strip()
    if not clean_text:
        return "scene"
    if len(clean_text) <= head_len + tail_len + 1:
        snippet = clean_text
    else:
        snippet = f"{clean_text[:head_len]}~{clean_text[-tail_len:]}"
    snippet = re.sub(r"\s+", " ", snippet).strip()
    return snippet[:max_len].rstrip(" .")