"""
pip install faiss-cpu
"""
import re
from Util.SplitDocxUtil import SplitDocxUtil

# 源文件
input_file = '../Txt/小学数学（史校长）.docx'

def split_into_blocks(text):
    """使用正则表达式匹配问题和话题的标题及内容"""
    pattern = r'(问题\d+|话题\d+)([\s\S]+?)(?=问题\d+|话题\d+|$)'
    blocks = re.findall(pattern, text, re.DOTALL)
    return [(i + 1, title + content) for i, (title, content) in enumerate(blocks)]



"""处理文档主函数"""
text = SplitDocxUtil.read_docx(input_file)
# 切开块
blocks = split_into_blocks(text)

# 将块编码并添加到向量数据库
for block in blocks:
    if (len(block[1]) > 10):
        print(block[1])
        print("\r\n")