[srmediathek] improve extraction
[youtube-dl] / youtube_dl / extractor / ign.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6
7
8 class IGNIE(InfoExtractor):
9     """
10     Extractor for some of the IGN sites, like www.ign.com, es.ign.com de.ign.com.
11     Some videos of it.ign.com are also supported
12     """
13
14     _VALID_URL = r'https?://.+?\.ign\.com/(?P<type>videos|show_videos|articles|(?:[^/]*/feature))(/.+)?/(?P<name_or_id>.+)'
15     IE_NAME = 'ign.com'
16
17     _CONFIG_URL_TEMPLATE = 'http://www.ign.com/videos/configs/id/%s.config'
18     _DESCRIPTION_RE = [
19         r'<span class="page-object-description">(.+?)</span>',
20         r'id="my_show_video">.*?<p>(.*?)</p>',
21         r'<meta name="description" content="(.*?)"',
22     ]
23
24     _TESTS = [
25         {
26             'url': 'http://www.ign.com/videos/2013/06/05/the-last-of-us-review',
27             'md5': 'eac8bdc1890980122c3b66f14bdd02e9',
28             'info_dict': {
29                 'id': '8f862beef863986b2785559b9e1aa599',
30                 'ext': 'mp4',
31                 'title': 'The Last of Us Review',
32                 'description': 'md5:c8946d4260a4d43a00d5ae8ed998870c',
33             }
34         },
35         {
36             'url': 'http://me.ign.com/en/feature/15775/100-little-things-in-gta-5-that-will-blow-your-mind',
37             'info_dict': {
38                 'id': '100-little-things-in-gta-5-that-will-blow-your-mind',
39             },
40             'playlist': [
41                 {
42                     'info_dict': {
43                         'id': '5ebbd138523268b93c9141af17bec937',
44                         'ext': 'mp4',
45                         'title': 'GTA 5 Video Review',
46                         'description': 'Rockstar drops the mic on this generation of games. Watch our review of the masterly Grand Theft Auto V.',
47                     },
48                 },
49                 {
50                     'info_dict': {
51                         'id': '638672ee848ae4ff108df2a296418ee2',
52                         'ext': 'mp4',
53                         'title': '26 Twisted Moments from GTA 5 in Slow Motion',
54                         'description': 'The twisted beauty of GTA 5 in stunning slow motion.',
55                     },
56                 },
57             ],
58             'params': {
59                 'skip_download': True,
60             },
61         },
62         {
63             'url': 'http://www.ign.com/articles/2014/08/15/rewind-theater-wild-trailer-gamescom-2014?watch',
64             'md5': '618fedb9c901fd086f6f093564ef8558',
65             'info_dict': {
66                 'id': '078fdd005f6d3c02f63d795faa1b984f',
67                 'ext': 'mp4',
68                 'title': 'Rewind Theater - Wild Trailer Gamescom 2014',
69                 'description': (
70                     'Giant skeletons, bloody hunts, and captivating'
71                     ' natural beauty take our breath away.'
72                 ),
73             },
74         },
75     ]
76
77     def _find_video_id(self, webpage):
78         res_id = [
79             r'"video_id"\s*:\s*"(.*?)"',
80             r'class="hero-poster[^"]*?"[^>]*id="(.+?)"',
81             r'data-video-id="(.+?)"',
82             r'<object id="vid_(.+?)"',
83             r'<meta name="og:image" content=".*/(.+?)-(.+?)/.+.jpg"',
84         ]
85         return self._search_regex(res_id, webpage, 'video id')
86
87     def _real_extract(self, url):
88         mobj = re.match(self._VALID_URL, url)
89         name_or_id = mobj.group('name_or_id')
90         page_type = mobj.group('type')
91         webpage = self._download_webpage(url, name_or_id)
92         if page_type != 'video':
93             multiple_urls = re.findall(
94                 '<param name="flashvars"[^>]*value="[^"]*?url=(https?://www\.ign\.com/videos/.*?)["&]',
95                 webpage)
96             if multiple_urls:
97                 entries = [self.url_result(u, ie='IGN') for u in multiple_urls]
98                 return {
99                     '_type': 'playlist',
100                     'id': name_or_id,
101                     'entries': entries,
102                 }
103
104         video_id = self._find_video_id(webpage)
105         result = self._get_video_info(video_id)
106         description = self._html_search_regex(self._DESCRIPTION_RE,
107                                               webpage, 'video description', flags=re.DOTALL)
108         result['description'] = description
109         return result
110
111     def _get_video_info(self, video_id):
112         config_url = self._CONFIG_URL_TEMPLATE % video_id
113         config = self._download_json(config_url, video_id)
114         media = config['playlist']['media']
115
116         return {
117             'id': media['metadata']['videoId'],
118             'url': media['url'],
119             'title': media['metadata']['title'],
120             'thumbnail': media['poster'][0]['url'].replace('{size}', 'grande'),
121         }
122
123
124 class OneUPIE(IGNIE):
125     _VALID_URL = r'https?://gamevideos\.1up\.com/(?P<type>video)/id/(?P<name_or_id>.+)\.html'
126     IE_NAME = '1up.com'
127
128     _DESCRIPTION_RE = r'<div id="vid_summary">(.+?)</div>'
129
130     _TESTS = [{
131         'url': 'http://gamevideos.1up.com/video/id/34976.html',
132         'md5': '68a54ce4ebc772e4b71e3123d413163d',
133         'info_dict': {
134             'id': '34976',
135             'ext': 'mp4',
136             'title': 'Sniper Elite V2 - Trailer',
137             'description': 'md5:5d289b722f5a6d940ca3136e9dae89cf',
138         }
139     }]
140
141     def _real_extract(self, url):
142         mobj = re.match(self._VALID_URL, url)
143         result = super(OneUPIE, self)._real_extract(url)
144         result['id'] = mobj.group('name_or_id')
145         return result