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