youtube_dl/extractor/linkedin.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     float_or_none,
  10     int_or_none,
  11     urlencode_postdata,
  12 )
  13
  14
  15 class LinkedInLearningBaseIE(InfoExtractor):
  16     _NETRC_MACHINE = 'linkedin'
  17
  18     def _call_api(self, course_slug, fields, video_slug=None, resolution=None):
  19         query = {
  20             'courseSlug': course_slug,
  21             'fields': fields,
  22             'q': 'slugs',
  23         }
  24         sub = ''
  25         if video_slug:
  26             query.update({
  27                 'videoSlug': video_slug,
  28                 'resolution': '_%s' % resolution,
  29             })
  30             sub = ' %dp' % resolution
  31         api_url = 'https://www.linkedin.com/learning-api/detailedCourses'
  32         return self._download_json(
  33             api_url, video_slug, 'Downloading%s JSON metadata' % sub, headers={
  34                 'Csrf-Token': self._get_cookies(api_url)['JSESSIONID'].value,
  35             }, query=query)['elements'][0]
  36
  37     def _get_urn_id(self, video_data):
  38         urn = video_data.get('urn')
  39         if urn:
  40             mobj = re.search(r'urn:li:lyndaCourse:\d+,(\d+)', urn)
  41             if mobj:
  42                 return mobj.group(1)
  43
  44     def _get_video_id(self, video_data, course_slug, video_slug):
  45         return self._get_urn_id(video_data) or '%s/%s' % (course_slug, video_slug)
  46
  47     def _real_initialize(self):
  48         email, password = self._get_login_info()
  49         if email is None:
  50             return
  51
  52         login_page = self._download_webpage(
  53             'https://www.linkedin.com/uas/login?trk=learning',
  54             None, 'Downloading login page')
  55         action_url = self._search_regex(
  56             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page, 'post url',
  57             default='https://www.linkedin.com/uas/login-submit', group='url')
  58         data = self._hidden_inputs(login_page)
  59         data.update({
  60             'session_key': email,
  61             'session_password': password,
  62         })
  63         login_submit_page = self._download_webpage(
  64             action_url, None, 'Logging in',
  65             data=urlencode_postdata(data))
  66         error = self._search_regex(
  67             r'<span[^>]+class="error"[^>]*>\s*(.+?)\s*</span>',
  68             login_submit_page, 'error', default=None)
  69         if error:
  70             raise ExtractorError(error, expected=True)
  71
  72
  73 class LinkedInLearningIE(LinkedInLearningBaseIE):
  74     IE_NAME = 'linkedin:learning'
  75     _VALID_URL = r'https?://(?:www\.)?linkedin\.com/learning/(?P<course_slug>[^/]+)/(?P<id>[^/?#]+)'
  76     _TEST = {
  77         'url': 'https://www.linkedin.com/learning/programming-foundations-fundamentals/welcome?autoplay=true',
  78         'md5': 'a1d74422ff0d5e66a792deb996693167',
  79         'info_dict': {
  80             'id': '90426',
  81             'ext': 'mp4',
  82             'title': 'Welcome',
  83             'timestamp': 1430396150.82,
  84             'upload_date': '20150430',
  85         },
  86     }
  87
  88     def _real_extract(self, url):
  89         course_slug, video_slug = re.match(self._VALID_URL, url).groups()
  90
  91         video_data = None
  92         formats = []
  93         for width, height in ((640, 360), (960, 540), (1280, 720)):
  94             video_data = self._call_api(
  95                 course_slug, 'selectedVideo', video_slug, height)['selectedVideo']
  96
  97             video_url_data = video_data.get('url') or {}
  98             progressive_url = video_url_data.get('progressiveUrl')
  99             if progressive_url:
 100                 formats.append({
 101                     'format_id': 'progressive-%dp' % height,
 102                     'url': progressive_url,
 103                     'height': height,
 104                     'width': width,
 105                     'source_preference': 1,
 106                 })
 107
 108         title = video_data['title']
 109
 110         audio_url = video_data.get('audio', {}).get('progressiveUrl')
 111         if audio_url:
 112             formats.append({
 113                 'abr': 64,
 114                 'ext': 'm4a',
 115                 'format_id': 'audio',
 116                 'url': audio_url,
 117                 'vcodec': 'none',
 118             })
 119
 120         streaming_url = video_url_data.get('streamingUrl')
 121         if streaming_url:
 122             formats.extend(self._extract_m3u8_formats(
 123                 streaming_url, video_slug, 'mp4',
 124                 'm3u8_native', m3u8_id='hls', fatal=False))
 125
 126         self._sort_formats(formats, ('width', 'height', 'source_preference', 'tbr', 'abr'))
 127
 128         return {
 129             'id': self._get_video_id(video_data, course_slug, video_slug),
 130             'title': title,
 131             'formats': formats,
 132             'thumbnail': video_data.get('defaultThumbnail'),
 133             'timestamp': float_or_none(video_data.get('publishedOn'), 1000),
 134             'duration': int_or_none(video_data.get('durationInSeconds')),
 135         }
 136
 137
 138 class LinkedInLearningCourseIE(LinkedInLearningBaseIE):
 139     IE_NAME = 'linkedin:learning:course'
 140     _VALID_URL = r'https?://(?:www\.)?linkedin\.com/learning/(?P<id>[^/?#]+)'
 141     _TEST = {
 142         'url': 'https://www.linkedin.com/learning/programming-foundations-fundamentals',
 143         'info_dict': {
 144             'id': 'programming-foundations-fundamentals',
 145             'title': 'Programming Foundations: Fundamentals',
 146             'description': 'md5:76e580b017694eb89dc8e8923fff5c86',
 147         },
 148         'playlist_mincount': 61,
 149     }
 150
 151     @classmethod
 152     def suitable(cls, url):
 153         return False if LinkedInLearningIE.suitable(url) else super(LinkedInLearningCourseIE, cls).suitable(url)
 154
 155     def _real_extract(self, url):
 156         course_slug = self._match_id(url)
 157         course_data = self._call_api(course_slug, 'chapters,description,title')
 158
 159         entries = []
 160         for chapter_number, chapter in enumerate(course_data.get('chapters', []), 1):
 161             chapter_title = chapter.get('title')
 162             chapter_id = self._get_urn_id(chapter)
 163             for video in chapter.get('videos', []):
 164                 video_slug = video.get('slug')
 165                 if not video_slug:
 166                     continue
 167                 entries.append({
 168                     '_type': 'url_transparent',
 169                     'id': self._get_video_id(video, course_slug, video_slug),
 170                     'title': video.get('title'),
 171                     'url': 'https://www.linkedin.com/learning/%s/%s' % (course_slug, video_slug),
 172                     'chapter': chapter_title,
 173                     'chapter_number': chapter_number,
 174                     'chapter_id': chapter_id,
 175                     'ie_key': LinkedInLearningIE.ie_key(),
 176                 })
 177
 178         return self.playlist_result(
 179             entries, course_slug,
 180             course_data.get('title'),
 181             course_data.get('description'))