Ejemplo n.º 1
0
    def test_caching(self):
        from crawl import CachingMentionCounter, to_epoch
        from tempfile import mkdtemp
        import shutil

        day = date(2011, 9, 3)
        self.api_fixture([_mention(to_epoch(day) + 60 * 20, 'english')])

        cache_root = mkdtemp()
        try:
            cmc = CachingMentionCounter(keyword='python',
                                        resolution=timedelta(days=1),
                                        cache_root=cache_root)
            stats1 = cmc.count(day)
            stats2 = cmc.count(day)

        finally:
            shutil.rmtree(cache_root)

        stats_ok = {datetime.combine(day, time()): {
            ('sentiment', None): 1,
            ('generator', 'twitter'): 1,
            ('language', 'english'): 1,
            None: 1
        }}
        self.assertEqual(stats1, stats_ok)
        self.assertEqual(stats2, stats_ok)
        self.assertEqual(self._mock_call.call_count, 1)
Ejemplo n.º 2
0
    def test_split_into_intervals(self):
        from crawl import MentionCounter, to_epoch

        day = date(2011, 9, 3)
        eday = to_epoch(day)

        self.api_fixture([
            _mention(eday + 60 * 20, 'english'),
            _mention(eday + 60 * 30, 'english'),
            _mention(eday + 60 * 40, 'german'),
            _mention(eday + 60 * 90, 'german'),
            _mention(eday + 60 * 100, 'german'),
            _mention(eday + 60 * 110, 'french'),
        ])

        stats = MentionCounter('python', timedelta(hours=1)).count(day)

        bucket0 = stats[datetime.combine(day, time())]
        self.assertEqual(bucket0['language', 'english'], 2)
        self.assertEqual(bucket0['language', 'german'], 1)

        bucket1 = stats[datetime.combine(day, time(1))]
        self.assertEqual(bucket1['language', 'german'], 2)
        self.assertEqual(bucket1['language', 'french'], 1)