"""
TXT 文本解析器

简单的文本文件读取，支持 UTF-8 和 GBK 编码自动检测。
"""

import logging
import os

logger = logging.getLogger(__name__)

# 文件大小限制
MAX_TXT_SIZE = 20 * 1024 * 1024  # 20MB


def extract_text_from_txt(filepath):
    """从TXT提取文本"""
    # 检查文件大小
    try:
        file_size = os.path.getsize(filepath)
        if file_size > MAX_TXT_SIZE:
            logger.error(f"TXT文件过大: {file_size / 1024 / 1024:.1f}MB")
            return ""
    except OSError:
        return ""
    
    try:
        with open(filepath, 'r', encoding='utf-8') as f:
            return f.read()
    except UnicodeDecodeError:
        # 尝试其他编码
        try:
            with open(filepath, 'r', encoding='gbk') as f:
                return f.read()
        except Exception as e:
            logger.error(f"TXT解析错误 {filepath}: {e}")
            return ""
    except Exception as e:
        print(f"      TXT解析错误 {filepath}: {e}")
        return ""