# -*- coding: utf-8 -*-
from __future__ import unicode_literals
import re
from datetime import datetime
from dateutil.relativedelta import relativedelta
from dateutil.parser import parse
_UNITS = r'year|month|week|day|hour|minute|second'
PATTERN = re.compile(r'(\d+)\s*(%s)\b' % _UNITS, re.I | re.S | re.U)
[docs]class FreshnessDateDataParser(object):
""" Parses date string like "1 year, 2 months ago" and "3 hours, 50 minutes ago" """
def __init__(self, now=None):
self.now = now or datetime.utcnow()
def _are_all_words_units(self, date_string):
skip = [_UNITS,
r'ago|\d+',
r':|[ap]m']
date_string = re.sub(r'\s+', ' ', date_string.strip())
words = filter(lambda x: x if x else False, re.split('\W', date_string))
words = filter(lambda x: not re.match(r'%s' % '|'.join(skip), x), words)
return not list(words)
def _parse_time(self, date_string):
"""Attemps to parse time part of date strings like '1 day ago, 2 PM' """
date_string = PATTERN.sub('', date_string)
date_string = re.sub(r'\bago\b', '', date_string)
try:
return parse(date_string).time()
except:
pass
[docs] def parse(self, date_string):
date, period = self._parse(date_string)
if date:
time = self._parse_time(date_string)
if time:
date = date.replace(hour=time.hour, minute=time.minute,
second=time.second, microsecond=time.microsecond)
return date, period
def _parse(self, date_string):
if not self._are_all_words_units(date_string):
return None, None
kwargs = self.get_kwargs(date_string)
if not kwargs:
return None, None
period = 'day'
if 'days' not in kwargs:
for k in ['weeks', 'months', 'years']:
if k in kwargs:
period = k[:-1]
break
td = relativedelta(**kwargs)
date = self.now - td
return date, period
[docs] def get_kwargs(self, date_string):
m = PATTERN.findall(date_string)
if not m:
return {}
kwargs = {}
for num, unit in m:
kwargs[unit + 's'] = int(num)
return kwargs
[docs] def get_date_data(self, date_string):
date, period = self.parse(date_string)
return dict(date_obj=date, period=period)
freshness_date_parser = FreshnessDateDataParser()