]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/voxmedia.py
Merge pull request #1 from e7appew/python3
[youtubedl] / youtube_dl / extractor / voxmedia.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..compat import compat_urllib_parse_unquote
6
7
8 class VoxMediaIE(InfoExtractor):
9 _VALID_URL = r'https?://(?:www\.)?(?:theverge|vox|sbnation|eater|polygon|curbed|racked)\.com/(?:[^/]+/)*(?P<id>[^/?]+)'
10 _TESTS = [{
11 'url': 'http://www.theverge.com/2014/6/27/5849272/material-world-how-google-discovered-what-software-is-made-of',
12 'md5': '73856edf3e89a711e70d5cf7cb280b37',
13 'info_dict': {
14 'id': '11eXZobjrG8DCSTgrNjVinU-YmmdYjhe',
15 'ext': 'mp4',
16 'title': 'Google\'s new material design direction',
17 'description': 'md5:2f44f74c4d14a1f800ea73e1c6832ad2',
18 },
19 'add_ie': ['Ooyala'],
20 }, {
21 # data-ooyala-id
22 'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet',
23 'md5': 'd744484ff127884cd2ba09e3fa604e4b',
24 'info_dict': {
25 'id': 'RkZXU4cTphOCPDMZg5oEounJyoFI0g-B',
26 'ext': 'mp4',
27 'title': 'The Nexus 6: hands-on with Google\'s phablet',
28 'description': 'md5:87a51fe95ff8cea8b5bdb9ac7ae6a6af',
29 },
30 'add_ie': ['Ooyala'],
31 }, {
32 # volume embed
33 'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill',
34 'md5': '375c483c5080ab8cd85c9c84cfc2d1e4',
35 'info_dict': {
36 'id': 'wydzk3dDpmRz7PQoXRsTIX6XTkPjYL0b',
37 'ext': 'mp4',
38 'title': 'The new frontier of LGBTQ civil rights, explained',
39 'description': 'md5:0dc58e94a465cbe91d02950f770eb93f',
40 },
41 'add_ie': ['Ooyala'],
42 }, {
43 # youtube embed
44 'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance',
45 'md5': '83b3080489fb103941e549352d3e0977',
46 'info_dict': {
47 'id': 'FcNHTJU1ufM',
48 'ext': 'mp4',
49 'title': 'How "the robot" became the greatest novelty dance of all time',
50 'description': 'md5:b081c0d588b8b2085870cda55e6da176',
51 'upload_date': '20160324',
52 'uploader_id': 'voxdotcom',
53 'uploader': 'Vox',
54 },
55 'add_ie': ['Youtube'],
56 }, {
57 # SBN.VideoLinkset.entryGroup multiple ooyala embeds
58 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
59 'info_dict': {
60 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
61 'title': '25 lies you will tell yourself on National Signing Day',
62 'description': 'It\'s the most self-delusional time of the year, and everyone\'s gonna tell the same lies together!',
63 },
64 'playlist': [{
65 'md5': '721fededf2ab74ae4176c8c8cbfe092e',
66 'info_dict': {
67 'id': 'p3cThlMjE61VDi_SD9JlIteSNPWVDBB9',
68 'ext': 'mp4',
69 'title': 'Buddy Hield vs Steph Curry (and the world)',
70 'description': 'Let’s dissect only the most important Final Four storylines.',
71 },
72 }, {
73 'md5': 'bf0c5cc115636af028be1bab79217ea9',
74 'info_dict': {
75 'id': 'BmbmVjMjE6esPHxdALGubTrouQ0jYLHj',
76 'ext': 'mp4',
77 'title': 'Chasing Cinderella 2016: Syracuse basketball',
78 'description': 'md5:e02d56b026d51aa32c010676765a690d',
79 },
80 }],
81 }]
82
83 def _real_extract(self, url):
84 display_id = self._match_id(url)
85 webpage = compat_urllib_parse_unquote(self._download_webpage(url, display_id))
86
87 def create_entry(provider_video_id, provider_video_type, title=None, description=None):
88 return {
89 '_type': 'url_transparent',
90 'url': provider_video_id if provider_video_type == 'youtube' else '%s:%s' % (provider_video_type, provider_video_id),
91 'title': title or self._og_search_title(webpage),
92 'description': description or self._og_search_description(webpage),
93 }
94
95 entries = []
96 entries_data = self._search_regex([
97 r'Chorus\.VideoContext\.addVideo\((\[{.+}\])\);',
98 r'var\s+entry\s*=\s*({.+});',
99 r'SBN\.VideoLinkset\.entryGroup\(\s*(\[.+\])',
100 ], webpage, 'video data', default=None)
101 if entries_data:
102 entries_data = self._parse_json(entries_data, display_id)
103 if isinstance(entries_data, dict):
104 entries_data = [entries_data]
105 for video_data in entries_data:
106 provider_video_id = video_data.get('provider_video_id')
107 provider_video_type = video_data.get('provider_video_type')
108 if provider_video_id and provider_video_type:
109 entries.append(create_entry(
110 provider_video_id, provider_video_type,
111 video_data.get('title'), video_data.get('description')))
112
113 provider_video_id = self._search_regex(
114 r'data-ooyala-id="([^"]+)"', webpage, 'ooyala id', default=None)
115 if provider_video_id:
116 entries.append(create_entry(provider_video_id, 'ooyala'))
117
118 volume_uuid = self._search_regex(
119 r'data-volume-uuid="([^"]+)"', webpage, 'volume uuid', default=None)
120 if volume_uuid:
121 volume_webpage = self._download_webpage(
122 'http://volume.vox-cdn.com/embed/%s' % volume_uuid, volume_uuid)
123 video_data = self._parse_json(self._search_regex(
124 r'Volume\.createVideo\(({.+})\s*,\s*{.*}\s*,\s*\[.*\]\s*,\s*{.*}\);', volume_webpage, 'video data'), volume_uuid)
125 for provider_video_type in ('ooyala', 'youtube'):
126 provider_video_id = video_data.get('%s_id' % provider_video_type)
127 if provider_video_id:
128 description = video_data.get('description_long') or video_data.get('description_short')
129 entries.append(create_entry(
130 provider_video_id, provider_video_type, video_data.get('title_short'), description))
131 break
132
133 if len(entries) == 1:
134 return entries[0]
135 else:
136 return self.playlist_result(entries, display_id, self._og_search_title(webpage), self._og_search_description(webpage))