zoukankan      html  css  js  c++  java
  • python自动提取文本中的时间(包含中文日期)

    有时在处理不规则数据时需要提取文本包含的时间日期。
    dateutil.parser模块可以统一日期字符串格式。
    datefinder模块可以在字符串中提取日期。
    datefinder模块实现也是用正则,功能很全 但是对中文不友好。
    但是这两个模块都不能支持中文及一些特殊的情况;所以我用正则写了段代码可进行中文日期及一些特殊的时间识别
    例如:
    '2012年12月12日','3小时前','在2012/12/13哈哈','时间2012-12-11 12:22:30','日期2012-13-11','测试2013.12.24','今天12:13'

    import re
    import chardet
    from datetime import datetime,timedelta
    
    
    # 匹配正则表达式
    matchs = {
        1:(r'd{4}%sd{1,2}%sd{1,2}%s d{1,2}%sd{1,2}%sd{1,2}%s','%%Y%s%%m%s%%d%s %%H%s%%M%s%%S%s'),
        2:(r'd{4}%sd{1,2}%sd{1,2}%s d{1,2}%sd{1,2}%s','%%Y%s%%m%s%%d%s %%H%s%%M%s'),
        3:(r'd{4}%sd{1,2}%sd{1,2}%s','%%Y%s%%m%s%%d%s'),
        4:(r'd{2}%sd{1,2}%sd{1,2}%s','%%y%s%%m%s%%d%s'),
       
        # 没有年份
        5:(r'd{1,2}%sd{1,2}%s d{1,2}%sd{1,2}%sd{1,2}%s','%%m%s%%d%s %%H%s%%M%s%%S%s'),
        6:(r'd{1,2}%sd{1,2}%s d{1,2}%sd{1,2}%s','%%m%s%%d%s %%H%s%%M%s'),
        7:(r'd{1,2}%sd{1,2}%s','%%m%s%%d%s'),
        
    
        # 没有年月日
        8:(r'd{1,2}%sd{1,2}%sd{1,2}%s','%%H%s%%M%s%%S%s'),
        9:(r'd{1,2}%sd{1,2}%s','%%H%s%%M%s'),
    }
    
    # 正则中的%s分割
    splits = [
        {1:[('年','月','日','点','分','秒'),('-','-','',':',':',''),('/','/','',':',':',''),('.','.','',':',':','')]},
        {2:[('年','月','日','点','分'),('-','-','',':',''),('/','/','',':',''),('.','.','',':','')]},
        {3:[('年','月','日'),('-','-',''),('/','/',''),('.','.','')]},
        {4:[('年','月','日'),('-','-',''),('/','/',''),('.','.','')]},
    
        {5:[('月','日','点','分','秒'),('-','',':',':',''),('/','',':',':',''),('.','',':',':','')]},
        {6:[('月','日','点','分'),('-','',':',''),('/','',':',''),('.','',':','')]},
        {7:[('月','日'),('-',''),('/',''),('.','')]},
    
        {8:[('点','分','秒'),(':',':','')]},
        {9:[('点','分'),(':','')]},
    ]
    
    def func(parten,tp):
        re.search(parten,parten)
        
    
    parten_other = 'd+天前|d+分钟前|d+小时前|d+秒前'
    
    class TimeFinder(object):
    
        def __init__(self,base_date=None):
            self.base_date = base_date
            self.match_item = []
            
            self.init_args()
            self.init_match_item()
    
        def init_args(self):
            # 格式化基础时间
            if not self.base_date:
                self.base_date = datetime.now()
            if self.base_date and not isinstance(self.base_date,datetime):
                try:
                    self.base_date = datetime.strptime(self.base_date,'%Y-%m-%d %H:%M:%S')
                except Exception as e:
                    raise 'type of base_date must be str of%Y-%m-%d %H:%M:%S or datetime'
    
        def init_match_item(self):
            # 构建穷举正则匹配公式 及提取的字符串转datetime格式映射
            for item in splits:
                for num,value in item.items():
                    match = matchs[num]
                    for sp in value:
                        tmp = []
                        for m in match:
                            tmp.append(m%sp)
                        self.match_item.append(tuple(tmp))
    
        def get_time_other(self,text):
            m = re.search('d+',text)
            if not m:
                return None
            num = int(m.group())
            if '天' in text:
                return self.base_date - timedelta(days=num)
            elif '小时' in text:
                return self.base_date - timedelta(hours=num)
            elif '分钟' in text:
                return self.base_date - timedelta(minutes=num)
            elif '秒' in text:
                return self.base_date - timedelta(seconds=num)
    
            return None
    
        def find_time(self,text):
             # 格式化text为str类型
            if isinstance(text,bytes):
                encoding =chardet.detect(text)['encoding']
                text = text.decode(encoding)
    
            res = []
            parten = '|'.join([x[0] for x in self.match_item])
    
            parten = parten+ '|' +parten_other
            match_list = re.findall(parten,text)
            if not match_list:
                return None
            for match in match_list:
                for item in self.match_item:
                    try:
                        date = datetime.strptime(match,item[1].replace('\',''))
                        if date.year==1900:
                            date = date.replace(year=self.base_date.year)
                            if date.month==1:
                                date = date.replace(month=self.base_date.month)
                                if date.day==1:
                                    date = date.replace(day=self.base_date.day)
                        res.append(datetime.strftime(date,'%Y-%m-%d %H:%M:%S'))
                        break
                    except Exception as e:
                        date = self.get_time_other(match)
                        if date:
                            res.append(datetime.strftime(date,'%Y-%m-%d %H:%M:%S'))
                            break
            if not res:
                return None
            return res
    
    def test():
        timefinder =TimeFinder(base_date='2020-04-23 00:00:00')
        for text in ['2012年12月12日','3小时前','在2012/12/13哈哈','时间2012-12-11 12:22:30','日期2012-13-11','测试2013.12.24','今天12:13']:
            res = timefinder.find_time(text)
            print('text----',text)
            print('res---',res)
    
    if __name__ == '__main__':
        test()
    

    测试运行结果如下

    text---- 2012年12月12日
    res--- ['2012-12-12 00:00:00']
    text---- 3小时前
    res--- ['2020-04-22 21:00:00']
    text---- 在2012/12/13哈哈
    res--- ['2012-12-13 00:00:00']
    text---- 时间2012-12-11 12:22:30
    res--- ['2012-12-11 12:22:30']
    text---- 日期2012-13-11
    res--- None
    text---- 测试2013.12.24
    res--- ['2013-12-24 00:00:00']
    text---- 今天12:13
    res--- ['2020-04-23 12:13:00']
    
  • 相关阅读:
    51nod 1179 最大的最大公约数
    HDU 6090 Rikka with Graph —— 2017 Multi-University Training 5
    HDU 6073 Matching In Multiplication —— 2017 Multi-University Training 4
    HDU 6069 Counting Divisors —— 2017 Multi-University Training 4
    HDU 6058 Kanade's sum —— 2017 Multi-University Training 3
    HDU 6055 Regular polygon —— 2017 Multi-University Training 2
    HDU 6053 TrickGCD —— 2017 Multi-University Training 2
    ORA-01722:无效数字
    ORA-01427 单行子查询返回多个行
    ORA 00972 错误处理
  • 原文地址:https://www.cnblogs.com/i-love-python/p/12763063.html
Copyright © 2011-2022 走看看