import os
import re

def is_index_line(line):
    # 檢查是否為索引行，通常索引行以數字開頭
    return bool(re.match(r'^\d+$', line))

def is_time_line(line):
    # 檢查是否為時間線，包含"-->"的行
    return bool(re.search(r'-->', line))

def clean_srt_file(file_path):
    with open(file_path, 'r', encoding='utf-8') as file:
        lines = file.readlines()

    new_content = []
    subtitle_block = []
    in_subtitle = False

    for line in lines:
        # 清除空白行
        if not line.strip():
            if subtitle_block:  # 如果當前有字幕塊，則添加一個換行符
                new_content.append(line)
            in_subtitle = False  # 重置字幕塊狀態
            continue

        if is_index_line(line) or is_time_line(line):
            # 如果當前行為索引或時間線，則保存之前的字幕塊（如果存在）
            if subtitle_block:
                new_content.extend(subtitle_block)
                new_content.append('')  # 保留字幕塊之間的換行符
                subtitle_block = []
            # 添加索引行和時間線
            new_content.append(line)
            in_subtitle = True
        elif in_subtitle:
            # 僅保留字幕中的中文內容
            chinese_only_line = re.sub(r'[^\u4e00-\u9fa5\n]', '', line)
            if chinese_only_line.strip():  # 如果行內有中文字符
                subtitle_block.append(chinese_only_line)

    # 添加最后一個字幕塊
    if subtitle_block:
        new_content.extend(subtitle_block)
        new_content.append('')  # 在文件末尾添加一個換行符

    # 保存為新的srt文件
    new_file_path = file_path.replace('.srt', '_cleaned.srt')
    with open(new_file_path, 'w', encoding='utf-8') as new_file:
        new_file.writelines(new_content)

    print(f'Processed and saved cleaned file to: {new_file_path}')

def process_directory(directory_path):
    for filename in os.listdir(directory_path):
        if filename.endswith('.srt'):
            file_path = os.path.join(directory_path, filename)
            clean_srt_file(file_path)

# 指定目錄路徑
directory_path = r'C:\caijian\29-51'
process_directory(directory_path)

import os
import re

def clean_and_save_srt(file_path):
    with open(file_path, 'r', encoding='utf-8') as file:
        lines = file.readlines()

    new_content = []
    in_subtitle_block = False
    subtitle_block = []

    for line in lines:
        line = line.strip()

        if is_index_line(line) or is_time_line(line):
            if subtitle_block:
                new_content.extend(subtitle_block)
                new_content.append('')  # 保留字幕塊之間的換行符
                subtitle_block = []
            new_content.append(line)
            in_subtitle_block = True
        elif in_subtitle_block:
            if line:  # 非空行
                cleaned_line = re.sub(r'[^\S\r\n]+', '', line)  # 去除多余的空格
                cleaned_line = re.sub(r'[^\u4e00-\u9fa5，。？！、《》（）【】："，。？！]', '', cleaned_line)  # 只保留中文和部分標點
                if cleaned_line:
                    subtitle_block.append(cleaned_line)
            else:  # 空行
                if subtitle_block:  # 如果字幕塊不為空，則結束當前字幕塊
                    new_content.extend(subtitle_block)
                    new_content.append('')  # 保留字幕塊之間的換行符
                    subtitle_block = []
                    in_subtitle_block = False

    if subtitle_block:  # 添加最后一個字幕塊
        new_content.extend(subtitle_block)

    new_file_path = file_path.replace('.srt', '_cleaned.srt')
    with open(new_file_path, 'w', encoding='utf-8') as new_file:
        new_file.write('\n'.join(new_content))

    print(f'Processed and saved cleaned file to: {new_file_path}')

def is_index_line(line):
    return bool(re.match(r'^\d+$', line))

def is_time_line(line):
    return bool(re.search(r'-->', line))

def process_directory(directory_path):
    for filename in os.listdir(directory_path):
        if filename.endswith('.srt'):
            file_path = os.path.join(directory_path, filename)
            clean_and_save_srt(file_path)

# 指定目錄路徑
directory_path = r'C:\caijian\29-51'
process_directory(directory_path)