1 # -*- coding: utf-8 -*-
2 from __future__ import unicode_literals
4 from .common import InfoExtractor
5 from ..utils import parse_iso8601, ExtractorError
8 class CtsNewsIE(InfoExtractor):
9 # https connection failed (Connection reset)
10 _VALID_URL = r'http://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': 'md5:95e9b295c898b7ff294f09d450178d7d',
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': 'md5:f183feeba3752b683827aab71adad584',
31 'thumbnail': 're:^https?://.*\.jpg$',
32 'timestamp': 1378205880,
33 'upload_date': '20130903',
36 # With Youtube embedded video
37 'url': 'http://news.cts.com.tw/cts/money/201501/201501291578003.html',
38 'md5': '1d842c771dc94c8c3bca5af2cc1db9c5',
39 'add_ie': ['Youtube'],
43 'title': 'iPhone6熱銷 蘋果財報亮眼',
44 'description': 'md5:f395d4f485487bb0f992ed2c4b07aa7d',
45 'thumbnail': 're:^https?://.*\.jpg$',
46 'upload_date': '20150128',
47 'uploader_id': 'TBSCTS',
52 def _real_extract(self, url):
53 news_id = self._match_id(url)
54 page = self._download_webpage(url, news_id)
56 if self._search_regex(r'(CTSPlayer2)', page, 'CTSPlayer2 identifier', default=None):
57 feed_url = self._html_search_regex(
58 r'(http://news\.cts\.com\.tw/action/mp4feed\.php\?news_id=\d+)',
60 video_url = self._download_webpage(
61 feed_url, news_id, note='Fetching feed')
63 self.to_screen('Not CTSPlayer video, trying Youtube...')
64 youtube_url = self._search_regex(
65 r'src="(//www\.youtube\.com/embed/[^"]+)"', page, 'youtube url',
68 raise ExtractorError('The news includes no videos!', expected=True)
76 description = self._html_search_meta('description', page)
77 title = self._html_search_meta('title', page)
78 thumbnail = self._html_search_meta('image', page)
80 datetime_str = self._html_search_regex(
81 r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page, 'date and time')
82 # Transform into ISO 8601 format with timezone info
83 datetime_str = datetime_str.replace('/', '-') + ':00+0800'
84 timestamp = parse_iso8601(datetime_str, delimiter=' ')
90 'description': description,
91 'thumbnail': thumbnail,
92 'timestamp': timestamp,