[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / twitcasting.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..utils import urlencode_postdata
6
7 import re
8
9
10 class TwitCastingIE(InfoExtractor):
11     _VALID_URL = r'https?://(?:[^/]+\.)?twitcasting\.tv/(?P<uploader_id>[^/]+)/movie/(?P<id>\d+)'
12     _TESTS = [{
13         'url': 'https://twitcasting.tv/ivetesangalo/movie/2357609',
14         'md5': '745243cad58c4681dc752490f7540d7f',
15         'info_dict': {
16             'id': '2357609',
17             'ext': 'mp4',
18             'title': 'Live #2357609',
19             'uploader_id': 'ivetesangalo',
20             'description': "Moi! I'm live on TwitCasting from my iPhone.",
21             'thumbnail': r're:^https?://.*\.jpg$',
22         },
23         'params': {
24             'skip_download': True,
25         },
26     }, {
27         'url': 'https://twitcasting.tv/mttbernardini/movie/3689740',
28         'info_dict': {
29             'id': '3689740',
30             'ext': 'mp4',
31             'title': 'Live playing something #3689740',
32             'uploader_id': 'mttbernardini',
33             'description': "I'm live on TwitCasting from my iPad. password: abc (Santa Marinella/Lazio, Italia)",
34             'thumbnail': r're:^https?://.*\.jpg$',
35         },
36         'params': {
37             'skip_download': True,
38             'videopassword': 'abc',
39         },
40     }]
41
42     def _real_extract(self, url):
43         mobj = re.match(self._VALID_URL, url)
44         video_id = mobj.group('id')
45         uploader_id = mobj.group('uploader_id')
46
47         video_password = self._downloader.params.get('videopassword')
48         request_data = None
49         if video_password:
50             request_data = urlencode_postdata({
51                 'password': video_password,
52             })
53         webpage = self._download_webpage(url, video_id, data=request_data)
54
55         title = self._html_search_regex(
56             r'(?s)<[^>]+id=["\']movietitle[^>]+>(.+?)</',
57             webpage, 'title', default=None) or self._html_search_meta(
58             'twitter:title', webpage, fatal=True)
59
60         m3u8_url = self._search_regex(
61             (r'data-movie-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
62              r'(["\'])(?P<url>http.+?\.m3u8.*?)\1'),
63             webpage, 'm3u8 url', group='url')
64
65         formats = self._extract_m3u8_formats(
66             m3u8_url, video_id, ext='mp4', entry_protocol='m3u8_native',
67             m3u8_id='hls')
68
69         thumbnail = self._og_search_thumbnail(webpage)
70         description = self._og_search_description(
71             webpage, default=None) or self._html_search_meta(
72             'twitter:description', webpage)
73
74         return {
75             'id': video_id,
76             'title': title,
77             'description': description,
78             'thumbnail': thumbnail,
79             'uploader_id': uploader_id,
80             'formats': formats,
81         }