Python中文日期时间提取：jieba和dateutil实战

import jieba.posseg as psg
import re
from datetime import datetime, timedelta
from dateutil.parser import parse
import os

# 定义中文数字和单位的映射字典
UTIL_CN_NUM = {
    '零': 0, '一': 1, '二': 2, '两': 2, '三': 3, '四': 4, '五': 5, '六': 6, '七': 7, '八': 8, '九': 9,
    '0': 0, '1': 1, '2': 2, '3': 3, '4': 4, '5': 5, '6': 6, '7': 7, '8': 8, '9': 9,
}
UTIL_CN_UTIL = {'十': 10, '百': 100, '千': 1000, '万': 10000}

# 检查时间字符串是否合法
def check_time_valid(word):
    m = re.match(r'\d+$', word)
    if m:
        if len(word) <= 6:
            return None
    word_1 = re.sub(r'[号|日]\d+$', '日', word)
    if word_1 != word:
        return check_time_valid(word_1)
    else:
        return word_1

# 将中文年份转换为数字
def year2dig(year):
    res = ''
    for item in year:
        if item in UTIL_CN_NUM.keys():
            res = res + str(UTIL_CN_NUM[item])
        else:
            res = res + item
    m = re.match(r'\d+', res)
    if m:
        if len(m.group(0)) == 2:
            return int(datetime.today().year / 100) * 100 + int(m.group(0))
        else:
            return int(m.group(0))
    else:
        return None

# 将中文数字转换为阿拉伯数字
def cn2dig(src):
    if src == '':
        return None
    m = re.match(r'\d+', src)
    if m:
        return int(m.group(0))
    rsl = 0
    unit = 1
    for item in src[::-1]:
        if item in UTIL_CN_UTIL.keys():
            unit = UTIL_CN_UTIL[item]
        elif item in UTIL_CN_NUM.keys():
            num = UTIL_CN_NUM[item]
            rsl += num * unit
        else:
            return None
    if rsl < unit:
        rsl += unit
    return rsl

# 将时间字符串解析为标准格式
def parse_datetime(msg):
    if msg is None or len(msg) == 0:
        return None
    try:
        dt = parse(msg)
        return dt.strftime('%Y-%m-%d %H:%M:%S')
    except Exception as e:
        m = re.match(r'([0-9零一二两三四五六七八九十]+ 年)? ([0-9一二两三四五六七八九十]+ 月)? '
                     r'([0-9一二两三四五六七八九十]+ [号日])? ([上中下午晚早]+)?'
                     r'([0-9零一二两三四五六七八九十百]+[点:.时])?([0-9零一二三四五六七八九十百]+ 分?)?'
                     r'([0-9零一二三四五六七八九十百]+ 秒)?', msg)
        if m and m.group(0) is not None:
            res = {
                'year': m.group(1),
                'month': m.group(2),
                'day': m.group(3),
                'hour': m.group(5) if m.group(5) is not None else '00',
                'minute': m.group(6) if m.group(6) is not None else '00',
                'second': m.group(7) if m.group(7) is not None else '00',
            }
            params = {}
            for name in res:
                if res[name] is not None and len(res[name]) != 0:
                    if name == 'year':
                        tmp = year2dig(res[name][:-1])
                    else:
                        tmp = cn2dig(res[name][:-1])
                    if tmp is not None:
                        params[name] = int(tmp)
            target_date = datetime.today().replace(**params)
            is_pm = m.group(4)
            if is_pm is not None:
                if is_pm == u'下午' or is_pm == u'晚上' or is_pm == u'中午':
                    hour = target_date.time().hour
                    if hour < 12:
                        target_date = target_date.replace(hour=hour + 12)
            return target_date.strftime('%Y-%m-%d %H:%M:%S')
        else:
            return None

# 从文本中提取时间信息
def time_extract(text):
    time_res = []
    word = ''
    key_date = {'今天': 0, '明天': 1, '后天': 2}
    for k, v in psg.cut(text):
        if k in key_date:
            if word != '':
                time_res.append(word)
            word = (datetime.today() + timedelta(days=key_date.get(k, 0))) \
                .strftime('%Y {0} %m {1} %d {2} ').format('年', '月', '日')
        elif word != '':
            if v in ['m', 't']:
                word = word + k
            else:
                time_res.append(word)
                word = ''
        elif v in ['m', 't']:
            word = k
    if word != '':
        time_res.append(word)
    result = list(filter(lambda x: x is not None, [check_time_valid(w) for w in time_res]))
    final_res = [parse_datetime(w) for w in result]
    return [x for x in final_res if x is not None]


# 读取语料文本
with open('d:/Users/Administrator/Desktop/data/corpus.txt', encoding='utf-8') as f:
    text = f.read()
    f.close()

# 提取时间信息并打印
result = time_extract(text)
for r in result:
    print(r)