2 from __future__ import unicode_literals
4 from .common import InfoExtractor
5 from ..utils import unified_timestamp
6 from .youtube import YoutubeIE
9 class CtsNewsIE(InfoExtractor):
11 _VALID_URL = r'https?://news\.cts\.com\.tw/[a-z]+/[a-z]+/\d+/(?P<id>\d+)\.html'
13 'url': 'http://news.cts.com.tw/cts/international/201501/201501291578109.html',
14 'md5': 'a9875cb790252b08431186d741beaabe',
16 'id': '201501291578109',
18 'title': '以色列.真主黨交火 3人死亡 - 華視新聞網',
19 'description': '以色列和黎巴嫩真主黨,爆發五年最嚴重衝突,雙方砲轟交火,兩名以軍死亡,還有一名西班牙籍的聯合國維和人員也不幸罹難。大陸陝西、河南、安徽、江蘇和湖北五個省份出現大暴雪,嚴重影響陸空交通,不過九華山卻出現...',
20 'timestamp': 1422528540,
21 'upload_date': '20150129',
24 # News count not appear on page but still available in database
25 'url': 'http://news.cts.com.tw/cts/international/201309/201309031304098.html',
26 'md5': '3aee7e0df7cdff94e43581f54c22619e',
28 'id': '201309031304098',
30 'title': '韓國31歲童顏男 貌如十多歲小孩 - 華視新聞網',
31 'description': '越有年紀的人,越希望看起來年輕一點,而南韓卻有一位31歲的男子,看起來像是11、12歲的小孩,身...',
32 'thumbnail': r're:^https?://.*\.jpg$',
33 'timestamp': 1378205880,
34 'upload_date': '20130903',
37 # With Youtube embedded video
38 'url': 'http://news.cts.com.tw/cts/money/201501/201501291578003.html',
39 'md5': 'e4726b2ccd70ba2c319865e28f0a91d1',
43 'title': 'iPhone6熱銷 蘋果財報亮眼',
44 'description': 'md5:f395d4f485487bb0f992ed2c4b07aa7d',
45 'thumbnail': r're:^https?://.*\.jpg$',
46 'upload_date': '20150128',
47 'uploader_id': 'TBSCTS',
50 'add_ie': ['Youtube'],
53 def _real_extract(self, url):
54 news_id = self._match_id(url)
55 page = self._download_webpage(url, news_id)
57 news_id = self._hidden_inputs(page).get('get_id')
60 mp4_feed = self._download_json(
61 'http://news.cts.com.tw/action/test_mp4feed.php',
62 news_id, note='Fetching feed', query={'news_id': news_id})
63 video_url = mp4_feed['source_url']
65 self.to_screen('Not CTSPlayer video, trying Youtube...')
66 youtube_url = YoutubeIE._extract_url(page)
68 return self.url_result(youtube_url, ie='Youtube')
70 description = self._html_search_meta('description', page)
71 title = self._html_search_meta('title', page, fatal=True)
72 thumbnail = self._html_search_meta('image', page)
74 datetime_str = self._html_search_regex(
75 r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page, 'date and time', fatal=False)
78 timestamp = unified_timestamp(datetime_str) - 8 * 3600
84 'description': description,
85 'thumbnail': thumbnail,
86 'timestamp': timestamp,