2 from __future__ import unicode_literals
4 from .common import InfoExtractor
5 from ..utils import unified_timestamp
8 class CtsNewsIE(InfoExtractor):
10 _VALID_URL = r'https?://news\.cts\.com\.tw/[a-z]+/[a-z]+/\d+/(?P<id>\d+)\.html'
12 'url': 'http://news.cts.com.tw/cts/international/201501/201501291578109.html',
13 'md5': 'a9875cb790252b08431186d741beaabe',
15 'id': '201501291578109',
17 'title': '以色列.真主黨交火 3人死亡',
18 '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...',
19 'timestamp': 1422528540,
20 'upload_date': '20150129',
23 # News count not appear on page but still available in database
24 'url': 'http://news.cts.com.tw/cts/international/201309/201309031304098.html',
25 'md5': '3aee7e0df7cdff94e43581f54c22619e',
27 'id': '201309031304098',
29 'title': '韓國31歲童顏男 貌如十多歲小孩',
30 '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...',
31 'thumbnail': r're:^https?://.*\.jpg$',
32 'timestamp': 1378205880,
33 'upload_date': '20130903',
36 # With Youtube embedded video
37 'url': 'http://news.cts.com.tw/cts/money/201501/201501291578003.html',
38 'md5': 'e4726b2ccd70ba2c319865e28f0a91d1',
42 'title': 'iPhone6熱銷 蘋果財報亮眼',
43 'description': 'md5:f395d4f485487bb0f992ed2c4b07aa7d',
44 'thumbnail': r're:^https?://.*\.jpg$',
45 'upload_date': '20150128',
46 'uploader_id': 'TBSCTS',
49 'add_ie': ['Youtube'],
52 def _real_extract(self, url):
53 news_id = self._match_id(url)
54 page = self._download_webpage(url, news_id)
56 news_id = self._hidden_inputs(page).get('get_id')
59 mp4_feed = self._download_json(
60 'http://news.cts.com.tw/action/test_mp4feed.php',
61 news_id, note='Fetching feed', query={'news_id': news_id})
62 video_url = mp4_feed['source_url']
64 self.to_screen('Not CTSPlayer video, trying Youtube...')
65 youtube_url = self._search_regex(
66 r'src="(//www\.youtube\.com/embed/[^"]+)"', page, 'youtube url')
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,