from html.parser import HTMLParser

class PythonConferenceParser(HTMLParser):

    def __init__(self, html):
        super(PythonConferenceParser, self).__init__()
        self.conferences = []
        self.feed(html)

    #attrs是含有tuple对象的列表
    def handle_starttag(self, name, attrs):
        attrs = dict(attrs)
        if name == 'h3' and 'class' in attrs and 'event-title' == attrs['class']:
            self.conferences.append({})
            self.conferences[-1]['event_title'] = None
        elif name == 'time' and 'datetime' in attrs:
            self.conferences[-1]['datetime'] = None
        elif name == 'span' and 'class' in attrs and 'event-location' == attrs['class']:
            self.conferences[-1]['event_location'] = None


    def handle_data(self, data):
        key = self._get_next_attr()
        if key:
            self.conferences[-1][key] = data

    def _get_next_attr(self):
        if not self.conferences:
            return None
        for k, v in self.conferences[-1].items():
            if v == None:
                return k
        return None

    def __str__(self):
        return str(self.conferences)

conferences = PythonConferenceParser(html).conferences
for conference in conferences:
    print(conference)