]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/ctvnews.py
[ie/matchtv] Fix extractor (#10190)
[yt-dlp.git] / yt_dlp / extractor / ctvnews.py
1 import re
2
3 from .common import InfoExtractor
4 from ..utils import orderedSet
5
6
7 class CTVNewsIE(InfoExtractor):
8 _VALID_URL = r'https?://(?:.+?\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)'
9 _TESTS = [{
10 'url': 'http://www.ctvnews.ca/video?clipId=901995',
11 'md5': '9b8624ba66351a23e0b6e1391971f9af',
12 'info_dict': {
13 'id': '901995',
14 'ext': 'flv',
15 'title': 'Extended: \'That person cannot be me\' Johnson says',
16 'description': 'md5:958dd3b4f5bbbf0ed4d045c790d89285',
17 'timestamp': 1467286284,
18 'upload_date': '20160630',
19 },
20 }, {
21 'url': 'http://www.ctvnews.ca/video?playlistId=1.2966224',
22 'info_dict':
23 {
24 'id': '1.2966224',
25 },
26 'playlist_mincount': 19,
27 }, {
28 'url': 'http://www.ctvnews.ca/video?binId=1.2876780',
29 'info_dict':
30 {
31 'id': '1.2876780',
32 },
33 'playlist_mincount': 100,
34 }, {
35 'url': 'http://www.ctvnews.ca/1.810401',
36 'only_matching': True,
37 }, {
38 'url': 'http://www.ctvnews.ca/canadiens-send-p-k-subban-to-nashville-in-blockbuster-trade-1.2967231',
39 'only_matching': True,
40 }, {
41 'url': 'http://vancouverisland.ctvnews.ca/video?clipId=761241',
42 'only_matching': True,
43 }]
44
45 def _real_extract(self, url):
46 page_id = self._match_id(url)
47
48 def ninecninemedia_url_result(clip_id):
49 return {
50 '_type': 'url_transparent',
51 'id': clip_id,
52 'url': f'9c9media:ctvnews_web:{clip_id}',
53 'ie_key': 'NineCNineMedia',
54 }
55
56 if page_id.isdigit():
57 return ninecninemedia_url_result(page_id)
58 else:
59 webpage = self._download_webpage(f'http://www.ctvnews.ca/{page_id}', page_id, query={
60 'ot': 'example.AjaxPageLayout.ot',
61 'maxItemsPerPage': 1000000,
62 })
63 entries = [ninecninemedia_url_result(clip_id) for clip_id in orderedSet(
64 re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
65 if not entries:
66 webpage = self._download_webpage(url, page_id)
67 if 'getAuthStates("' in webpage:
68 entries = [ninecninemedia_url_result(clip_id) for clip_id in
69 self._search_regex(r'getAuthStates\("([\d+,]+)"', webpage, 'clip ids').split(',')]
70 return self.playlist_result(entries, page_id)