如何获取句子中指定日期列表的起止索引？

给定句子：'Foo bar was open on 12.03.2022 and closed on 3.05.22.'
日期列表：list = ['12.03.2022', '4.04.2022', '3.05.22']（注：列表元素需为字符串类型，否则无法直接匹配）

需求：获取列表中能在句子里找到的日期的起止索引，以元组形式返回，期望结果为[(20,29), (45, 51)]。目前已通过正则表达式匹配到日期，但无法获取对应的索引，现有代码如下：

import re

DAY = r'(?:(?:0)[1-9]|[12]\d|3[01])'  # day can be from 1 to 31 with a leading zero 
MONTH = r'(?:(?:0)[1-9]|1[0-2])' # month can be 1 to 12 with a leading zero
YEAR1 = r'(?:(?:20|)\d{2}|(?:19|){9}[0-9])'  # Restricted the year to begin in 20th or 21st century 
                            # Also the first two digits may be skipped if data is represented as dd.mm.yy  
YEAR2 = r'(?:20\d{2}|199[0-9])'
                     
BEGIN_LINE1 = r'(?<!\w)'
DELIM1 = r'(?:[\,\/\-\._])' 
DELIM2 = r'(?:[\,\/\-\._])?'

# combined, several options
NUM_DATE =  f"""(?P<date>
    (?:
        # DAY MONTH YEAR
        (?:{BEGIN_LINE1}{DAY}{DELIM1}{MONTH}{DELIM1}{YEAR1})
        |
        (?:{BEGIN_LINE1}{DAY}{DELIM1}{MONTH})
        |
        (?:{BEGIN_LINE1}{MONTH}{DELIM1}{YEAR1})
        |
        (?:{BEGIN_LINE1}{DAY}{DELIM2}{MONTH}{DELIM2}{YEAR2})
        |
        (?:{BEGIN_LINE1}{MONTH}{DELIM2}{YEAR2})
    )
)"""


myDate = re.compile(f'{NUM_DATE}', re.IGNORECASE | re.VERBOSE | re.UNICODE)


def find_date(subject):
    """_summary_

    Args:
        subject (_type_): _description_

    Returns:
        _type_: _description_
    """
    
    if subject is None:
        return subject
    
    dates = list(set(myDate.findall(subject)))
    
    
    return dates

要获取匹配日期的起止索引，需用re.finditer()替代findall()，该方法返回匹配对象迭代器，每个匹配对象包含start()（起始索引）和end()（结束索引）方法。

修改后的代码 #

import re

DAY = r'(?:(?:0)[1-9]|[12]\d|3[01])'  # 日期：1-31，可带前导零
MONTH = r'(?:(?:0)[1-9]|1[0-2])' # 月份：1-12，可带前导零
YEAR1 = r'(?:(?:20|)\d{2}|(?:19|){9}[0-9])'  # 年份限制在20或21世纪，支持简写为yy格式
YEAR2 = r'(?:20\d{2}|199[0-9])'

BEGIN_LINE1 = r'(?<!\w)'
DELIM1 = r'(?:[\,\/\-\._])' 
DELIM2 = r'(?:[\,\/\-\._])?'

# 组合正则表达式，匹配多种日期格式
NUM_DATE =  f"""(?P<date>
    (?:
        # 日.月.年 格式
        (?:{BEGIN_LINE1}{DAY}{DELIM1}{MONTH}{DELIM1}{YEAR1})
        |
        (?:{BEGIN_LINE1}{DAY}{DELIM1}{MONTH})
        |
        (?:{BEGIN_LINE1}{MONTH}{DELIM1}{YEAR1})
        |
        (?:{BEGIN_LINE1}{DAY}{DELIM2}{MONTH}{DELIM2}{YEAR2})
        |
        (?:{BEGIN_LINE1}{MONTH}{DELIM2}{YEAR2})
    )
)"""

myDate = re.compile(f'{NUM_DATE}', re.IGNORECASE | re.VERBOSE | re.UNICODE)

def find_date_indices(subject, target_dates):
    if subject is None or not target_dates:
        return []
    
    # 收集所有匹配的日期及其索引
    date_matches = []
    for match in myDate.finditer(subject):
        date_str = match.group('date')
        start_idx = match.start()
        end_idx = match.end()
        date_matches.append((date_str, start_idx, end_idx))
    
    # 筛选出目标列表中存在的日期的索引
    result = []
    target_set = set(target_dates)
    for date_str, start, end in date_matches:
        if date_str in target_set:
            result.append((start, end))
    
    return result

# 测试示例
sentence = 'Foo bar was open on 12.03.2022 and closed on 3.05.22.'
target_list = ['12.03.2022', '4.04.2022', '3.05.22']
print(find_date_indices(sentence, target_list))  # 输出: [(20, 29), (45, 51)]

关键说明 #

1. re.finditer()的使用：该方法遍历句子中所有匹配的日期，返回的每个Match对象可直接调用start()和end()获取索引。
2. 目标日期匹配：将目标列表转为集合，可快速判断匹配到的日期是否在目标范围内，提升查找效率。
3. 格式一致性：确保目标列表中的日期为字符串类型，避免因类型不匹配导致的匹配失败。

内容的提问来源于stack exchange，提问作者Yana