]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/lynda.py
Implement a different adult sites checking algorithm
[youtube-dl] / youtube_dl / extractor / lynda.py
1 from __future__ import unicode_literals
2
3 import re
4 import json
5
6 from .subtitles import SubtitlesInfoExtractor
7 from .common import InfoExtractor
8 from ..utils import (
9     compat_urllib_parse,
10     compat_urllib_request,
11     ExtractorError
12 )
13
14
15 class LyndaIE(SubtitlesInfoExtractor):
16     IE_NAME = 'lynda'
17     IE_DESC = 'lynda.com videos'
18     _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html'
19     _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
20     _NETRC_MACHINE = 'lynda'
21
22     _SUCCESSFUL_LOGIN_REGEX = r'<a href="https://www.lynda.com/home/userAccount/ChangeContactInfo.aspx" data-qa="eyebrow_account_menu">My account'
23     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
24
25     ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
26
27     _TEST = {
28         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
29         'file': '114408.mp4',
30         'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
31         'info_dict': {
32             'title': 'Using the exercise files',
33             'duration': 68
34         }
35     }
36
37     def _real_initialize(self):
38         self._login()
39
40     def _real_extract(self, url):
41         mobj = re.match(self._VALID_URL, url)
42         video_id = mobj.group(1)
43
44         page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
45                                       video_id, 'Downloading video JSON')
46         video_json = json.loads(page)
47
48         if 'Status' in video_json:
49             raise ExtractorError('lynda returned error: %s' % video_json['Message'], expected=True)
50
51         if video_json['HasAccess'] is False:
52             raise ExtractorError('Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True)
53
54         video_id = video_json['ID']
55         duration = video_json['DurationInSeconds']
56         title = video_json['Title']
57
58         formats = [{'url': fmt['Url'],
59                     'ext': fmt['Extension'],
60                     'width': fmt['Width'],
61                     'height': fmt['Height'],
62                     'filesize': fmt['FileSize'],
63                     'format_id': str(fmt['Resolution'])
64                     } for fmt in video_json['Formats']]
65
66         self._sort_formats(formats)
67
68         if self._downloader.params.get('listsubtitles', False):
69             self._list_available_subtitles(video_id, page)
70             return
71
72         subtitles = self._fix_subtitles(self.extract_subtitles(video_id, page))
73
74         return {
75             'id': video_id,
76             'title': title,
77             'duration': duration,
78             'subtitles': subtitles,
79             'formats': formats
80         }
81
82     def _login(self):
83         (username, password) = self._get_login_info()
84         if username is None:
85             return
86
87         login_form = {
88             'username': username,
89             'password': password,
90             'remember': 'false',
91             'stayPut': 'false'
92         }        
93         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
94         login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
95
96         # Not (yet) logged in
97         m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page)
98         if m is not None:
99             response = m.group('json')
100             response_json = json.loads(response)            
101             state = response_json['state']
102
103             if state == 'notlogged':
104                 raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
105
106             # This is when we get popup:
107             # > You're already logged in to lynda.com on two devices.
108             # > If you log in here, we'll log you out of another device.
109             # So, we need to confirm this.
110             if state == 'conflicted':
111                 confirm_form = {
112                     'username': '',
113                     'password': '',
114                     'resolve': 'true',
115                     'remember': 'false',
116                     'stayPut': 'false',
117                 }
118                 request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form))
119                 login_page = self._download_webpage(request, None, note='Confirming log in and log out from another device')
120
121         if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
122             raise ExtractorError('Unable to log in')
123
124     def _fix_subtitles(self, subtitles):
125         if subtitles is None:
126             return subtitles  # subtitles not requested
127
128         fixed_subtitles = {}
129         for k, v in subtitles.items():
130             subs = json.loads(v)
131             if len(subs) == 0:
132                 continue
133             srt = ''
134             for pos in range(0, len(subs) - 1):
135                 seq_current = subs[pos]
136                 m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
137                 if m_current is None:
138                     continue
139                 seq_next = subs[pos + 1]
140                 m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
141                 if m_next is None:
142                     continue
143                 appear_time = m_current.group('timecode')
144                 disappear_time = m_next.group('timecode')
145                 text = seq_current['Caption']
146                 srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text)
147             if srt:
148                 fixed_subtitles[k] = srt
149         return fixed_subtitles
150
151     def _get_available_subtitles(self, video_id, webpage):
152         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
153         sub = self._download_webpage(url, None, note=False)
154         sub_json = json.loads(sub)
155         return {'en': url} if len(sub_json) > 0 else {}
156
157
158 class LyndaCourseIE(InfoExtractor):
159     IE_NAME = 'lynda:course'
160     IE_DESC = 'lynda.com online courses'
161
162     # Course link equals to welcome/introduction video link of same course
163     # We will recognize it as course link
164     _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
165
166     def _real_extract(self, url):
167         mobj = re.match(self._VALID_URL, url)
168         course_path = mobj.group('coursepath')
169         course_id = mobj.group('courseid')
170         
171         page = self._download_webpage('http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
172                                       course_id, 'Downloading course JSON')
173         course_json = json.loads(page)
174
175         if 'Status' in course_json and course_json['Status'] == 'NotFound':
176             raise ExtractorError('Course %s does not exist' % course_id, expected=True)
177
178         unaccessible_videos = 0
179         videos = []
180         (username, _) = self._get_login_info()
181
182         for chapter in course_json['Chapters']:
183             for video in chapter['Videos']:
184                 if username is None and video['HasAccess'] is False:
185                     unaccessible_videos += 1
186                     continue
187                 videos.append(video['ID'])
188
189         if unaccessible_videos > 0:
190             self._downloader.report_warning('%s videos are only available for members and will not be downloaded. '
191                                             % unaccessible_videos + LyndaIE.ACCOUNT_CREDENTIALS_HINT)
192
193         entries = [
194             self.url_result('http://www.lynda.com/%s/%s-4.html' %
195                             (course_path, video_id),
196                             'Lynda')
197             for video_id in videos]
198
199         course_title = course_json['Title']
200
201         return self.playlist_result(entries, course_id, course_title)