1 from .common
import InfoExtractor
2 from .youtube
import YoutubeIE
3 from ..utils
import unified_timestamp
6 class CtsNewsIE(InfoExtractor
):
8 _VALID_URL
= r
'https?://news\.cts\.com\.tw/[a-z]+/[a-z]+/\d+/(?P<id>\d+)\.html'
10 'url': 'http://news.cts.com.tw/cts/international/201501/201501291578109.html',
11 'md5': 'a9875cb790252b08431186d741beaabe',
13 'id': '201501291578109',
15 'title': '以色列.真主黨交火 3人死亡 - 華視新聞網',
16 '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...',
17 'timestamp': 1422528540,
18 'upload_date': '20150129',
21 # News count not appear on page but still available in database
22 'url': 'http://news.cts.com.tw/cts/international/201309/201309031304098.html',
23 'md5': '3aee7e0df7cdff94e43581f54c22619e',
25 'id': '201309031304098',
27 'title': '韓國31歲童顏男 貌如十多歲小孩 - 華視新聞網',
28 '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...',
29 'thumbnail': r
're:^https?://.*\.jpg$',
30 'timestamp': 1378205880,
31 'upload_date': '20130903',
34 # With Youtube embedded video
35 'url': 'http://news.cts.com.tw/cts/money/201501/201501291578003.html',
36 'md5': 'e4726b2ccd70ba2c319865e28f0a91d1',
40 'title': 'iPhone6熱銷 蘋果財報亮眼',
41 'description': 'md5:f395d4f485487bb0f992ed2c4b07aa7d',
42 'thumbnail': r
're:^https?://.*\.jpg$',
43 'upload_date': '20150128',
44 'uploader_id': 'TBSCTS',
47 'add_ie': ['Youtube'],
50 def _real_extract(self
, url
):
51 news_id
= self
._match
_id
(url
)
52 page
= self
._download
_webpage
(url
, news_id
)
54 news_id
= self
._hidden
_inputs
(page
).get('get_id')
57 mp4_feed
= self
._download
_json
(
58 'http://news.cts.com.tw/action/test_mp4feed.php',
59 news_id
, note
='Fetching feed', query
={'news_id': news_id
})
60 video_url
= mp4_feed
['source_url']
62 self
.to_screen('Not CTSPlayer video, trying Youtube...')
63 youtube_url
= YoutubeIE
._extract
_url
(page
)
65 return self
.url_result(youtube_url
, ie
='Youtube')
67 description
= self
._html
_search
_meta
('description', page
)
68 title
= self
._html
_search
_meta
('title', page
, fatal
=True)
69 thumbnail
= self
._html
_search
_meta
('image', page
)
71 datetime_str
= self
._html
_search
_regex
(
72 r
'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page
, 'date and time', fatal
=False)
75 timestamp
= unified_timestamp(datetime_str
) - 8 * 3600
81 'description': description
,
82 'thumbnail': thumbnail
,
83 'timestamp': timestamp
,