#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import json
import re

# 读取原始文件
with open('src/data/expertSupportData.js', 'r', encoding='utf-8') as f:
    content = f.read()

# 提取JSON部分（去掉JavaScript变量声明和结尾）
json_start = content.index('{')
json_end = content.rfind('}') + 1
json_content = content[json_start:json_end]

# 解析JSON
data = json.loads(json_content)

# 递归处理所有字符串值
def clean_text(obj):
    if isinstance(obj, str):
        # 处理转义的双引号 \" -> "
        obj = obj.replace('\\"', '"')
        # 处理转义的换行符 \\n -> \n (保留真正的换行)
        obj = obj.replace('\\n', '\n')
        return obj
    elif isinstance(obj, dict):
        return {k: clean_text(v) for k, v in obj.items()}
    elif isinstance(obj, list):
        return [clean_text(item) for item in obj]
    else:
        return obj

# 清理数据
cleaned_data = clean_text(data)

# 生成新的JavaScript文件内容
output = """// 从化工问答内容.json转换的专家支持中心数据
const expertSupportData = """ + json.dumps(cleaned_data, ensure_ascii=False, indent=2) + """;

export default expertSupportData;
"""

# 写入新文件
with open('src/data/expertSupportData_cleaned.js', 'w', encoding='utf-8') as f:
    f.write(output)

print("数据清理完成！已生成 expertSupportData_cleaned.js")
print("\n示例内容修复:")
print("原始: 废水处理厂为什么要建\\\"调节池\\\"？")
print("修复后: 废水处理厂为什么要建\"调节池\"？")
print("\n原始: PLC：响应速度快...\\nDCS：更擅长连续过程控制...")
print("修复后: PLC：响应速度快...\nDCS：更擅长连续过程控制...")