]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ivi.py
7a400323dc4df3807057a77b25f7401ce5e2a3b8
[youtubedl] / youtube_dl / extractor / ivi.py
1 # encoding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5 import json
6
7 from .common import InfoExtractor
8 from ..compat import (
9 compat_urllib_request,
10 )
11 from ..utils import (
12 ExtractorError,
13 )
14
15
16 class IviIE(InfoExtractor):
17 IE_DESC = 'ivi.ru'
18 IE_NAME = 'ivi'
19 _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<videoid>\d+)'
20
21 _TESTS = [
22 # Single movie
23 {
24 'url': 'http://www.ivi.ru/watch/53141',
25 'md5': '6ff5be2254e796ed346251d117196cf4',
26 'info_dict': {
27 'id': '53141',
28 'ext': 'mp4',
29 'title': 'Иван Васильевич меняет профессию',
30 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
31 'duration': 5498,
32 'thumbnail': 'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg',
33 },
34 'skip': 'Only works from Russia',
35 },
36 # Serial's serie
37 {
38 'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
39 'md5': '221f56b35e3ed815fde2df71032f4b3e',
40 'info_dict': {
41 'id': '9549',
42 'ext': 'mp4',
43 'title': 'Двое из ларца - Серия 1',
44 'duration': 2655,
45 'thumbnail': 'http://thumbs.ivi.ru/f15.vcp.digitalaccess.ru/contents/8/4/0068dc0677041f3336b7c2baad8fc0.jpg',
46 },
47 'skip': 'Only works from Russia',
48 }
49 ]
50
51 # Sorted by quality
52 _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
53
54 # Sorted by size
55 _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480']
56
57 def _extract_description(self, html):
58 m = re.search(r'<meta name="description" content="(?P<description>[^"]+)"/>', html)
59 return m.group('description') if m is not None else None
60
61 def _extract_comment_count(self, html):
62 m = re.search('(?s)<a href="#" id="view-comments" class="action-button dim gradient">\s*Комментарии:\s*(?P<commentcount>\d+)\s*</a>', html)
63 return int(m.group('commentcount')) if m is not None else 0
64
65 def _real_extract(self, url):
66 mobj = re.match(self._VALID_URL, url)
67 video_id = mobj.group('videoid')
68
69 api_url = 'http://api.digitalaccess.ru/api/json/'
70
71 data = {'method': 'da.content.get',
72 'params': [video_id, {'site': 's183',
73 'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
74 'contentid': video_id
75 }
76 ]
77 }
78
79 request = compat_urllib_request.Request(api_url, json.dumps(data))
80
81 video_json_page = self._download_webpage(request, video_id, 'Downloading video JSON')
82 video_json = json.loads(video_json_page)
83
84 if 'error' in video_json:
85 error = video_json['error']
86 if error['origin'] == 'NoRedisValidData':
87 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
88 raise ExtractorError('Unable to download video %s: %s' % (video_id, error['message']), expected=True)
89
90 result = video_json['result']
91
92 formats = [{
93 'url': x['url'],
94 'format_id': x['content_format'],
95 'preference': self._known_formats.index(x['content_format']),
96 } for x in result['files'] if x['content_format'] in self._known_formats]
97
98 self._sort_formats(formats)
99
100 if not formats:
101 raise ExtractorError('No media links available for %s' % video_id)
102
103 duration = result['duration']
104 compilation = result['compilation']
105 title = result['title']
106
107 title = '%s - %s' % (compilation, title) if compilation is not None else title
108
109 previews = result['preview']
110 previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format']))
111 thumbnail = previews[-1]['url'] if len(previews) > 0 else None
112
113 video_page = self._download_webpage(url, video_id, 'Downloading video page')
114 description = self._extract_description(video_page)
115 comment_count = self._extract_comment_count(video_page)
116
117 return {
118 'id': video_id,
119 'title': title,
120 'thumbnail': thumbnail,
121 'description': description,
122 'duration': duration,
123 'comment_count': comment_count,
124 'formats': formats,
125 }
126
127
128 class IviCompilationIE(InfoExtractor):
129 IE_DESC = 'ivi.ru compilations'
130 IE_NAME = 'ivi:compilation'
131 _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
132 _TESTS = [{
133 'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
134 'info_dict': {
135 'id': 'dvoe_iz_lartsa',
136 'title': 'Двое из ларца (2006 - 2008)',
137 },
138 'playlist_mincount': 24,
139 }, {
140 'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
141 'info_dict': {
142 'id': 'dvoe_iz_lartsa/season1',
143 'title': 'Двое из ларца (2006 - 2008) 1 сезон',
144 },
145 'playlist_mincount': 12,
146 }]
147
148 def _extract_entries(self, html, compilation_id):
149 return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
150 for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % compilation_id, html)]
151
152 def _real_extract(self, url):
153 mobj = re.match(self._VALID_URL, url)
154 compilation_id = mobj.group('compilationid')
155 season_id = mobj.group('seasonid')
156
157 if season_id is not None: # Season link
158 season_page = self._download_webpage(url, compilation_id, 'Downloading season %s web page' % season_id)
159 playlist_id = '%s/season%s' % (compilation_id, season_id)
160 playlist_title = self._html_search_meta('title', season_page, 'title')
161 entries = self._extract_entries(season_page, compilation_id)
162 else: # Compilation link
163 compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
164 playlist_id = compilation_id
165 playlist_title = self._html_search_meta('title', compilation_page, 'title')
166 seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
167 if len(seasons) == 0: # No seasons in this compilation
168 entries = self._extract_entries(compilation_page, compilation_id)
169 else:
170 entries = []
171 for season_id in seasons:
172 season_page = self._download_webpage(
173 'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
174 compilation_id, 'Downloading season %s web page' % season_id)
175 entries.extend(self._extract_entries(season_page, compilation_id))
176
177 return self.playlist_result(entries, playlist_id, playlist_title)