模糊匹配省市区地址

用户输入地址不可能一定规范，如按习惯省略掉：“省”、“市”、“区”等关键字，此时安装正则匹配很容易查找不到正确的地址。
以下代码按照用户输入的先后顺序，相同的词组进行匹配，可靠性与适配性大大提高，记录于此以供参考：

def get_area_code(biz_address_code):
    """获取省市编码"""
    print(biz_address_code, type(biz_address_code))
    data, error_info = None, None
    # 判断传入数据是否为空
    if (not biz_address_code) or (not isinstance(biz_address_code, str)):
        error_info = '根据 开户银行省市:“{}” ,获取省市编码失败，请按规范填写！'.format(biz_address_code)
        return None, error_info

    num = -1  # 定义开始行数
    # 打开对照表
    comparison_table = os.path.join(BASE_PATH, 'static', '省市区编号对照表.xlsx')
    sheet = xlrd.open_workbook(comparison_table)

    table = sheet.sheets()[0]
    cols = table.col_values(1)

    res_address = None  # 终值
    res_weight = 0  # 权值
    for regions in cols:
        num += 1
        region_list = regions.split(',')  # eg: ['中国', '', '天津市', '河东区']
        address_str = ''
        weight = 0
        data = table.cell(num, 0).value
        biz_address = biz_address_code
        for index, region in enumerate(region_list):  # 使用index作为权值
            if not region:
                continue
            # 找到大于两个字符的共同部分
            ret = ''
            for zip_li in zip(region, biz_address):
                if len(set(zip_li)) == 1:
                    ret += zip_li[0]
                else:
                    break
            if len(ret) >= 2:
                weight += index
                address_str += region
                biz_address = biz_address.replace(ret, '')  # 将匹配到的部分去除掉，理论上该匹配没有问题

        if res_weight < weight:
            res_weight = weight
            res_address = data
    if not res_address:
        error_info = '根据 开户银行省市:“{}”，获取省市编码失败，请按规范填写！'.format(biz_address_code)
        return res_address, error_info
    else:
        res_address = str(int(res_address))

        return res_address, error_info**

省市区编号对照表.xlsx 是记录好所有的地区编号对应信息的excel表格，如下入所示：