]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/revision3.py
Fix extraction from youtube.
[youtubedl] / youtube_dl / extractor / revision3.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import compat_str
8 from ..utils import (
9 int_or_none,
10 parse_iso8601,
11 unescapeHTML,
12 qualities,
13 )
14
15
16 class Revision3EmbedIE(InfoExtractor):
17 IE_NAME = 'revision3:embed'
18 _VALID_URL = r'(?:revision3:(?:(?P<playlist_type>[^:]+):)?|https?://(?:(?:(?:www|embed)\.)?(?:revision3|animalist)|(?:(?:api|embed)\.)?seekernetwork)\.com/player/embed\?videoId=)(?P<playlist_id>\d+)'
19 _TEST = {
20 'url': 'http://api.seekernetwork.com/player/embed?videoId=67558',
21 'md5': '83bcd157cab89ad7318dd7b8c9cf1306',
22 'info_dict': {
23 'id': '67558',
24 'ext': 'mp4',
25 'title': 'The Pros & Cons Of Zoos',
26 'description': 'Zoos are often depicted as a terrible place for animals to live, but is there any truth to this?',
27 'uploader_id': 'dnews',
28 'uploader': 'DNews',
29 }
30 }
31 _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
32
33 def _real_extract(self, url):
34 mobj = re.match(self._VALID_URL, url)
35 playlist_id = mobj.group('playlist_id')
36 playlist_type = mobj.group('playlist_type') or 'video_id'
37 video_data = self._download_json(
38 'http://revision3.com/api/getPlaylist.json', playlist_id, query={
39 'api_key': self._API_KEY,
40 'codecs': 'h264,vp8,theora',
41 playlist_type: playlist_id,
42 })['items'][0]
43
44 formats = []
45 for vcodec, media in video_data['media'].items():
46 for quality_id, quality in media.items():
47 if quality_id == 'hls':
48 formats.extend(self._extract_m3u8_formats(
49 quality['url'], playlist_id, 'mp4',
50 'm3u8_native', m3u8_id='hls', fatal=False))
51 else:
52 formats.append({
53 'url': quality['url'],
54 'format_id': '%s-%s' % (vcodec, quality_id),
55 'tbr': int_or_none(quality.get('bitrate')),
56 'vcodec': vcodec,
57 })
58 self._sort_formats(formats)
59
60 return {
61 'id': playlist_id,
62 'title': unescapeHTML(video_data['title']),
63 'description': unescapeHTML(video_data.get('summary')),
64 'uploader': video_data.get('show', {}).get('name'),
65 'uploader_id': video_data.get('show', {}).get('slug'),
66 'duration': int_or_none(video_data.get('duration')),
67 'formats': formats,
68 }
69
70
71 class Revision3IE(InfoExtractor):
72 IE_NAME = 'revision'
73 _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)'
74 _TESTS = [{
75 'url': 'http://www.revision3.com/technobuffalo/5-google-predictions-for-2016',
76 'md5': 'd94a72d85d0a829766de4deb8daaf7df',
77 'info_dict': {
78 'id': '71089',
79 'display_id': 'technobuffalo/5-google-predictions-for-2016',
80 'ext': 'webm',
81 'title': '5 Google Predictions for 2016',
82 'description': 'Google had a great 2015, but it\'s already time to look ahead. Here are our five predictions for 2016.',
83 'upload_date': '20151228',
84 'timestamp': 1451325600,
85 'duration': 187,
86 'uploader': 'TechnoBuffalo',
87 'uploader_id': 'technobuffalo',
88 }
89 }, {
90 # Show
91 'url': 'http://revision3.com/variant',
92 'only_matching': True,
93 }, {
94 # Tag
95 'url': 'http://revision3.com/vr',
96 'only_matching': True,
97 }]
98 _PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s'
99
100 def _real_extract(self, url):
101 domain, display_id = re.match(self._VALID_URL, url).groups()
102 site = domain.split('.')[0]
103 page_info = self._download_json(
104 self._PAGE_DATA_TEMPLATE % (domain, display_id, domain), display_id)
105
106 page_data = page_info['data']
107 page_type = page_data['type']
108 if page_type in ('episode', 'embed'):
109 show_data = page_data['show']['data']
110 page_id = compat_str(page_data['id'])
111 video_id = compat_str(page_data['video']['data']['id'])
112
113 preference = qualities(['mini', 'small', 'medium', 'large'])
114 thumbnails = [{
115 'url': image_url,
116 'id': image_id,
117 'preference': preference(image_id)
118 } for image_id, image_url in page_data.get('images', {}).items()]
119
120 info = {
121 'id': page_id,
122 'display_id': display_id,
123 'title': unescapeHTML(page_data['name']),
124 'description': unescapeHTML(page_data.get('summary')),
125 'timestamp': parse_iso8601(page_data.get('publishTime'), ' '),
126 'author': page_data.get('author'),
127 'uploader': show_data.get('name'),
128 'uploader_id': show_data.get('slug'),
129 'thumbnails': thumbnails,
130 'extractor_key': site,
131 }
132
133 if page_type == 'embed':
134 info.update({
135 '_type': 'url_transparent',
136 'url': page_data['video']['data']['embed'],
137 })
138 return info
139
140 info.update({
141 '_type': 'url_transparent',
142 'url': 'revision3:%s' % video_id,
143 })
144 return info
145 else:
146 list_data = page_info[page_type]['data']
147 episodes_data = page_info['episodes']['data']
148 num_episodes = page_info['meta']['totalEpisodes']
149 processed_episodes = 0
150 entries = []
151 page_num = 1
152 while True:
153 entries.extend([{
154 '_type': 'url',
155 'url': 'http://%s%s' % (domain, episode['path']),
156 'id': compat_str(episode['id']),
157 'ie_key': 'Revision3',
158 'extractor_key': site,
159 } for episode in episodes_data])
160 processed_episodes += len(episodes_data)
161 if processed_episodes == num_episodes:
162 break
163 page_num += 1
164 episodes_data = self._download_json(self._PAGE_DATA_TEMPLATE % (
165 domain, display_id + '/' + compat_str(page_num), domain),
166 display_id)['episodes']['data']
167
168 return self.playlist_result(
169 entries, compat_str(list_data['id']),
170 list_data.get('name'), list_data.get('summary'))