# -*- coding: utf-8 -*-
import os, re

rr = re.compile(r'[\u4e00-\u9fff]')
BASE = '/app/working/workspaces/default/check_tmp'
OUT = '/app/working/workspaces/default/publish_ready'
os.makedirs(OUT, exist_ok=True)

stories = [
    ('r3-yue-tang-ping-yue-qiang', '01-越躺平越强，我摆烂成了全服第一'),
    ('r3-shen-ming-zhong-jie', '02-神明中介：我帮神仙接单，神仙帮我躺赢'),
    ('r3-man-ji-da-lao-hui-xin-shou-cun', '03-满级大佬回新手村，我装了三天萌新'),
    ('r4-tun-jin-shou-xi-tong', '04-分手后我觉醒了吞金兽系统'),
]

for repo_dir, story_name in stories:
    path = os.path.join(BASE, repo_dir)
    ch_files = sorted([f for f in os.listdir(path) if f.endswith('.md') and f.startswith('第')])
    
    intro = ''
    for f in os.listdir(path):
        if '简介' in f:
            with open(os.path.join(path, f), 'r', encoding='utf-8') as fp:
                intro = fp.read().strip()
    
    parts = [story_name + '\n']
    if intro:
        parts.append('【简介】\n' + intro + '\n')
    
    for f in ch_files:
        with open(os.path.join(path, f), 'r', encoding='utf-8') as fp:
            c = fp.read().strip()
        # 处理标题
        lines = c.split('\n')
        # 清理文件头部的# 标题行
        clean_lines = []
        for l in lines:
            if l.startswith('# '):
                continue  # 去掉markdown标题
            clean_lines.append(l)
        parts.append('\n'.join(clean_lines))
    
    all_text = '\n\n'.join(parts)
    total_h = len(rr.findall(all_text))
    
    out_fn = story_name + '.txt'
    out_path = os.path.join(OUT, out_fn)
    with open(out_path, 'w', encoding='utf-8') as f:
        f.write(all_text)
    
    print(f'OK: {story_name}')
    print(f'   汉字: {total_h}')
    print(f'   文件: {out_path}')
    print()

print('全部4部发布文件已生成!')