]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/breakcom.py
debian/copyright: Use HTTPS for upstream's source.
[youtubedl] / youtube_dl / extractor / breakcom.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..compat import compat_str
7 from ..utils import (
8 int_or_none,
9 parse_age_limit,
10 )
11
12
13 class BreakIE(InfoExtractor):
14 _VALID_URL = r'https?://(?:www\.)?(?P<site>break|screenjunkies)\.com/video/(?P<display_id>[^/]+?)(?:-(?P<id>\d+))?(?:[/?#&]|$)'
15 _TESTS = [{
16 'url': 'http://www.break.com/video/when-girls-act-like-guys-2468056',
17 'info_dict': {
18 'id': '2468056',
19 'ext': 'mp4',
20 'title': 'When Girls Act Like D-Bags',
21 'age_limit': 13,
22 }
23 }, {
24 'url': 'http://www.screenjunkies.com/video/best-quentin-tarantino-movie-2841915',
25 'md5': '5c2b686bec3d43de42bde9ec047536b0',
26 'info_dict': {
27 'id': '2841915',
28 'display_id': 'best-quentin-tarantino-movie',
29 'ext': 'mp4',
30 'title': 'Best Quentin Tarantino Movie',
31 'thumbnail': r're:^https?://.*\.jpg',
32 'duration': 3671,
33 'age_limit': 13,
34 'tags': list,
35 },
36 }, {
37 'url': 'http://www.screenjunkies.com/video/honest-trailers-the-dark-knight',
38 'info_dict': {
39 'id': '2348808',
40 'display_id': 'honest-trailers-the-dark-knight',
41 'ext': 'mp4',
42 'title': 'Honest Trailers - The Dark Knight',
43 'thumbnail': r're:^https?://.*\.(?:jpg|png)',
44 'age_limit': 10,
45 'tags': list,
46 },
47 }, {
48 # requires subscription but worked around
49 'url': 'http://www.screenjunkies.com/video/knocking-dead-ep-1-the-show-so-far-3003285',
50 'info_dict': {
51 'id': '3003285',
52 'display_id': 'knocking-dead-ep-1-the-show-so-far',
53 'ext': 'mp4',
54 'title': 'State of The Dead Recap: Knocking Dead Pilot',
55 'thumbnail': r're:^https?://.*\.jpg',
56 'duration': 3307,
57 'age_limit': 13,
58 'tags': list,
59 },
60 }, {
61 'url': 'http://www.break.com/video/ugc/baby-flex-2773063',
62 'only_matching': True,
63 }]
64
65 _DEFAULT_BITRATES = (48, 150, 320, 496, 864, 2240, 3264)
66
67 def _real_extract(self, url):
68 site, display_id, video_id = re.match(self._VALID_URL, url).groups()
69
70 if not video_id:
71 webpage = self._download_webpage(url, display_id)
72 video_id = self._search_regex(
73 (r'src=["\']/embed/(\d+)', r'data-video-content-id=["\'](\d+)'),
74 webpage, 'video id')
75
76 webpage = self._download_webpage(
77 'http://www.%s.com/embed/%s' % (site, video_id),
78 display_id, 'Downloading video embed page')
79 embed_vars = self._parse_json(
80 self._search_regex(
81 r'(?s)embedVars\s*=\s*({.+?})\s*</script>', webpage, 'embed vars'),
82 display_id)
83
84 youtube_id = embed_vars.get('youtubeId')
85 if youtube_id:
86 return self.url_result(youtube_id, 'Youtube')
87
88 title = embed_vars['contentName']
89
90 formats = []
91 bitrates = []
92 for f in embed_vars.get('media', []):
93 if not f.get('uri') or f.get('mediaPurpose') != 'play':
94 continue
95 bitrate = int_or_none(f.get('bitRate'))
96 if bitrate:
97 bitrates.append(bitrate)
98 formats.append({
99 'url': f['uri'],
100 'format_id': 'http-%d' % bitrate if bitrate else 'http',
101 'width': int_or_none(f.get('width')),
102 'height': int_or_none(f.get('height')),
103 'tbr': bitrate,
104 'format': 'mp4',
105 })
106
107 if not bitrates:
108 # When subscriptionLevel > 0, i.e. plus subscription is required
109 # media list will be empty. However, hds and hls uris are still
110 # available. We can grab them assuming bitrates to be default.
111 bitrates = self._DEFAULT_BITRATES
112
113 auth_token = embed_vars.get('AuthToken')
114
115 def construct_manifest_url(base_url, ext):
116 pieces = [base_url]
117 pieces.extend([compat_str(b) for b in bitrates])
118 pieces.append('_kbps.mp4.%s?%s' % (ext, auth_token))
119 return ','.join(pieces)
120
121 if bitrates and auth_token:
122 hds_url = embed_vars.get('hdsUri')
123 if hds_url:
124 formats.extend(self._extract_f4m_formats(
125 construct_manifest_url(hds_url, 'f4m'),
126 display_id, f4m_id='hds', fatal=False))
127 hls_url = embed_vars.get('hlsUri')
128 if hls_url:
129 formats.extend(self._extract_m3u8_formats(
130 construct_manifest_url(hls_url, 'm3u8'),
131 display_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
132 self._sort_formats(formats)
133
134 return {
135 'id': video_id,
136 'display_id': display_id,
137 'title': title,
138 'thumbnail': embed_vars.get('thumbUri'),
139 'duration': int_or_none(embed_vars.get('videoLengthInSeconds')) or None,
140 'age_limit': parse_age_limit(embed_vars.get('audienceRating')),
141 'tags': embed_vars.get('tags', '').split(','),
142 'formats': formats,
143 }