Created
April 24, 2020 09:21
-
-
Save yogendratamang48/17fff61dddde54cce6f7cba30e4b224a to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from datetime import datetime, timedelta | |
from nltk import word_tokenize | |
MOMENTS = { | |
'seconds': ['seconds', 'second', 'sec', 'just'], | |
'minutes': ['mins', 'minutes', 'today'], | |
'hours': ['hours', 'hour'], | |
'days': ['day', 'days', 'veröffentlicht', 'Tagen' ], | |
'today': ['today',], | |
'weeks': ['week', 'weeks'], | |
'months': ['month', 'months'], | |
} | |
class DateHelper: | |
@staticmethod | |
def handle_T_format(raw_text): | |
''' | |
2017-05-26T12:00 | |
''' | |
date_obj = datetime.strptime(raw_text, '%Y-%m-%dT%H:%M') | |
return date_obj | |
@staticmethod | |
def handle_TZ_format(raw_text): | |
''' | |
2017-05-26T12:00 | |
''' | |
if 'z' in raw_text.lower(): | |
raw_text = raw_text.upper().replace('Z', '').strip() | |
date_obj = datetime.strptime(raw_text, '%Y-%m-%dT%H:%M:%S') | |
return date_obj | |
@staticmethod | |
def handle_timestamp(raw_text): | |
date_obj = datetime.fromtimestamp(int(raw_text)) | |
return date_obj | |
@staticmethod | |
def handle_format(raw_date, _format): | |
date_obj = datetime.strptime(raw_date, _format) | |
return date_obj | |
def handle_moments(raw_date): | |
''' | |
''' | |
if raw_date is None: | |
return None | |
tokens = word_tokenize(raw_date.lower()) | |
numbers = [_.strip('+') for _ in tokens if _.strip('+').isnumeric()] | |
if numbers: | |
number = numbers[0] | |
else: | |
number = 1 | |
unit = '' | |
for moment, variants in MOMENTS.items(): | |
for variant in variants: | |
if variant in tokens: | |
unit = moment | |
break | |
if unit: | |
break | |
if unit == 'seconds': | |
_delta = timedelta(seconds=int(number)) | |
elif unit == 'minutes': | |
_delta = timedelta(minutes=int(number)) | |
elif unit == 'hours': | |
_delta = timedelta(hours=int(number)) | |
elif unit == 'days': | |
_delta = timedelta(days=int(number)) | |
elif unit == 'today': | |
_delta = timedelta(days=0) | |
elif unit == 'weeks': | |
_delta = timedelta(days=7*int(number)) | |
elif unit == 'months': | |
_delta = timedelta(days=int(number)) | |
return (datetime.now()-_delta) | |
def test_helper(): | |
raw_data = '2017-05-26T12:00' | |
date_obj = DateHelper.handle_T_format(raw_data) | |
print(date_obj.__class__) | |
def test_date_moments(): | |
texts = ['3 seconds ago', | |
'minutes a go', | |
'just now', | |
'4 hours ago', | |
'1 week ago', | |
'1 hour ago', | |
'5 hours ago', | |
'a week ago', | |
'3 weeks ago', | |
'15+ days ago', | |
'30+ days ago', | |
] | |
for _text in texts: | |
raw_date = DateHelper.handle_moments(_text) | |
print(_text, ":", raw_date) | |
if __name__ == '__main__': | |
test_date_moments() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment