[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / gdcvault.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from .kaltura import KalturaIE
7 from ..utils import (
8     HEADRequest,
9     sanitized_Request,
10     smuggle_url,
11     urlencode_postdata,
12 )
13
14
15 class GDCVaultIE(InfoExtractor):
16     _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)(?:/(?P<name>[\w-]+))?'
17     _NETRC_MACHINE = 'gdcvault'
18     _TESTS = [
19         {
20             'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
21             'md5': '7ce8388f544c88b7ac11c7ab1b593704',
22             'info_dict': {
23                 'id': '201311826596_AWNY',
24                 'display_id': 'Doki-Doki-Universe-Sweet-Simple',
25                 'ext': 'mp4',
26                 'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
27             }
28         },
29         {
30             'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
31             'info_dict': {
32                 'id': '201203272_1330951438328RSXR',
33                 'display_id': 'Embracing-the-Dark-Art-of',
34                 'ext': 'flv',
35                 'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
36             },
37             'params': {
38                 'skip_download': True,  # Requires rtmpdump
39             }
40         },
41         {
42             'url': 'http://www.gdcvault.com/play/1015301/Thexder-Meets-Windows-95-or',
43             'md5': 'a5eb77996ef82118afbbe8e48731b98e',
44             'info_dict': {
45                 'id': '1015301',
46                 'display_id': 'Thexder-Meets-Windows-95-or',
47                 'ext': 'flv',
48                 'title': 'Thexder Meets Windows 95, or Writing Great Games in the Windows 95 Environment',
49             },
50             'skip': 'Requires login',
51         },
52         {
53             'url': 'http://gdcvault.com/play/1020791/',
54             'only_matching': True,
55         },
56         {
57             # Hard-coded hostname
58             'url': 'http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface',
59             'md5': 'a8efb6c31ed06ca8739294960b2dbabd',
60             'info_dict': {
61                 'id': '840376_BQRC',
62                 'ext': 'mp4',
63                 'display_id': 'Tenacious-Design-and-The-Interface',
64                 'title': 'Tenacious Design and The Interface of \'Destiny\'',
65             },
66         },
67         {
68             # Multiple audios
69             'url': 'http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC',
70             'info_dict': {
71                 'id': '12396_1299111843500GMPX',
72                 'ext': 'mp4',
73                 'title': 'How to Create a Good Game - From My Experience of Designing Pac-Man',
74             },
75             # 'params': {
76             #     'skip_download': True,  # Requires rtmpdump
77             #     'format': 'jp',  # The japanese audio
78             # }
79         },
80         {
81             # gdc-player.html
82             'url': 'http://www.gdcvault.com/play/1435/An-American-engine-in-Tokyo',
83             'info_dict': {
84                 'id': '9350_1238021887562UHXB',
85                 'display_id': 'An-American-engine-in-Tokyo',
86                 'ext': 'mp4',
87                 'title': 'An American Engine in Tokyo:/nThe collaboration of Epic Games and Square Enix/nFor THE LAST REMINANT',
88             },
89         },
90         {
91             # Kaltura Embed
92             'url': 'https://www.gdcvault.com/play/1026180/Mastering-the-Apex-of-Scaling',
93             'info_dict': {
94                 'id': '0_h1fg8j3p',
95                 'ext': 'mp4',
96                 'title': 'Mastering the Apex of Scaling Game Servers (Presented by Multiplay)',
97                 'timestamp': 1554401811,
98                 'upload_date': '20190404',
99                 'uploader_id': 'joe@blazestreaming.com',
100             },
101             'params': {
102                 'format': 'mp4-408',
103             },
104         },
105     ]
106
107     def _login(self, webpage_url, display_id):
108         username, password = self._get_login_info()
109         if username is None or password is None:
110             self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
111             return None
112
113         mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
114         login_url = mobj.group('root_url') + 'api/login.php'
115         logout_url = mobj.group('root_url') + 'logout'
116
117         login_form = {
118             'email': username,
119             'password': password,
120         }
121
122         request = sanitized_Request(login_url, urlencode_postdata(login_form))
123         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
124         self._download_webpage(request, display_id, 'Logging in')
125         start_page = self._download_webpage(webpage_url, display_id, 'Getting authenticated video page')
126         self._download_webpage(logout_url, display_id, 'Logging out')
127
128         return start_page
129
130     def _real_extract(self, url):
131         video_id, name = re.match(self._VALID_URL, url).groups()
132         display_id = name or video_id
133
134         webpage_url = 'http://www.gdcvault.com/play/' + video_id
135         start_page = self._download_webpage(webpage_url, display_id)
136
137         direct_url = self._search_regex(
138             r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);',
139             start_page, 'url', default=None)
140         if direct_url:
141             title = self._html_search_regex(
142                 r'<td><strong>Session Name:?</strong></td>\s*<td>(.*?)</td>',
143                 start_page, 'title')
144             video_url = 'http://www.gdcvault.com' + direct_url
145             # resolve the url so that we can detect the correct extension
146             video_url = self._request_webpage(
147                 HEADRequest(video_url), video_id).geturl()
148
149             return {
150                 'id': video_id,
151                 'display_id': display_id,
152                 'url': video_url,
153                 'title': title,
154             }
155
156         embed_url = KalturaIE._extract_url(start_page)
157         if embed_url:
158             embed_url = smuggle_url(embed_url, {'source_url': url})
159             ie_key = 'Kaltura'
160         else:
161             PLAYER_REGEX = r'<iframe src="(?P<xml_root>.+?)/(?:gdc-)?player.*?\.html.*?".*?</iframe>'
162
163             xml_root = self._html_search_regex(
164                 PLAYER_REGEX, start_page, 'xml root', default=None)
165             if xml_root is None:
166                 # Probably need to authenticate
167                 login_res = self._login(webpage_url, display_id)
168                 if login_res is None:
169                     self.report_warning('Could not login.')
170                 else:
171                     start_page = login_res
172                     # Grab the url from the authenticated page
173                     xml_root = self._html_search_regex(
174                         PLAYER_REGEX, start_page, 'xml root')
175
176             xml_name = self._html_search_regex(
177                 r'<iframe src=".*?\?xml(?:=|URL=xml/)(.+?\.xml).*?".*?</iframe>',
178                 start_page, 'xml filename')
179             embed_url = '%s/xml/%s' % (xml_root, xml_name)
180             ie_key = 'DigitallySpeaking'
181
182         return {
183             '_type': 'url_transparent',
184             'id': video_id,
185             'display_id': display_id,
186             'url': embed_url,
187             'ie_key': ie_key,
188         }