parse_twitter_date.py

def parse_date(date):
    """parse out non-standard date format used by Twitter
    
    Usage: parse_date("Sun Jul 13 12:44:07 +0000 2008")
           >>> datetime.datetime(2008, 06, 13, 12, 44, 07)
    """
    date_regex = re.compile(r"""
        ^
        (?P<day_name>[a-zA-Z]+)
        \s+
        (?P<month_name>[a-zA-Z]+)
        \s+
        (?P<day>\d+)
        \s+
        (?P<hours>\d+)
        :
        (?P<minutes>\d+)
        :
        (?P<seconds>\d+)
        \s+
        \+\d+
        \s+
        (?P<year>\d+)
        $
    """, re.VERBOSE)
    
    if date_regex.match(date):
        date_string = date_regex.sub(r'\g<year> \g<month_name> \g<day> \g<hours>:\g<minutes>:\g<seconds>', date)
        return datetime.datetime(*time.strptime(date_string, '%Y %b %d %H:%M:%S')[:6])
    else:
        raise MalformedDate