_ Git - youtube-dl/blob - 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_video_id(self, urn, course_slug, video_slug):
  38         if urn:
  39             mobj = re.search(r'urn:li:lyndaCourse:\d+,(\d+)', urn)
  40             if mobj:
  41                 return mobj.group(1)
  42         return '%s/%s' % (course_slug, video_slug)
  43
  44     def _real_initialize(self):
  45         email, password = self._get_login_info()
  46         if email is None:
  47             return
  48
  49         login_page = self._download_webpage(
  50             'https://www.linkedin.com/uas/login?trk=learning',
  51             None, 'Downloading login page')
  52         action_url = self._search_regex(
  53             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page, 'post url',
  54             default='https://www.linkedin.com/uas/login-submit', group='url')
  55         data = self._hidden_inputs(login_page)
  56         data.update({
  57             'session_key': email,
  58             'session_password': password,
  59         })
  60         login_submit_page = self._download_webpage(
  61             action_url, None, 'Logging in',
  62             data=urlencode_postdata(data))
  63         error = self._search_regex(
  64             r'<span[^>]+class="error"[^>]*>\s*(.+?)\s*</span>',
  65             login_submit_page, 'error', default=None)
  66         if error:
  67             raise ExtractorError(error, expected=True)
  68
  69
  70 class LinkedInLearningIE(LinkedInLearningBaseIE):
  71     IE_NAME = 'linkedin:learning'
  72     _VALID_URL = r'https?://(?:www\.)?linkedin\.com/learning/(?P<course_slug>[^/]+)/(?P<id>[^/?#]+)'
  73     _TEST = {
  74         'url': 'https://www.linkedin.com/learning/programming-foundations-fundamentals/welcome?autoplay=true',
  75         'md5': 'a1d74422ff0d5e66a792deb996693167',
  76         'info_dict': {
  77             'id': '90426',
  78             'ext': 'mp4',
  79             'title': 'Welcome',
  80             'timestamp': 1430396150.82,
  81             'upload_date': '20150430',
  82         },
  83     }
  84
  85     def _real_extract(self, url):
  86         course_slug, video_slug = re.match(self._VALID_URL, url).groups()
  87
  88         video_data = None
  89         formats = []
  90         for width, height in ((640, 360), (960, 540), (1280, 720)):
  91             video_data = self._call_api(
  92                 course_slug, 'selectedVideo', video_slug, height)['selectedVideo']
  93
  94             video_url_data = video_data.get('url') or {}
  95             progressive_url = video_url_data.get('progressiveUrl')
  96             if progressive_url:
  97                 formats.append({
  98                     'format_id': 'progressive-%dp' % height,
  99                     'url': progressive_url,
 100                     'height': height,
 101                     'width': width,
 102                     'source_preference': 1,
 103                 })
 104
 105         title = video_data['title']
 106
 107         audio_url = video_data.get('audio', {}).get('progressiveUrl')
 108         if audio_url:
 109             formats.append({
 110                 'abr': 64,
 111                 'ext': 'm4a',
 112                 'format_id': 'audio',
 113                 'url': audio_url,
 114                 'vcodec': 'none',
 115             })
 116
 117         streaming_url = video_url_data.get('streamingUrl')
 118         if streaming_url:
 119             formats.extend(self._extract_m3u8_formats(
 120                 streaming_url, video_slug, 'mp4',
 121                 'm3u8_native', m3u8_id='hls', fatal=False))
 122
 123         self._sort_formats(formats, ('width', 'height', 'source_preference', 'tbr', 'abr'))
 124
 125         return {
 126             'id': self._get_video_id(video_data.get('urn'), course_slug, video_slug),
 127             'title': title,
 128             'formats': formats,
 129             'thumbnail': video_data.get('defaultThumbnail'),
 130             'timestamp': float_or_none(video_data.get('publishedOn'), 1000),
 131             'duration': int_or_none(video_data.get('durationInSeconds')),
 132         }
 133
 134
 135 class LinkedInLearningCourseIE(LinkedInLearningBaseIE):
 136     IE_NAME = 'linkedin:learning:course'
 137     _VALID_URL = r'https?://(?:www\.)?linkedin\.com/learning/(?P<id>[^/?#]+)'
 138     _TEST = {
 139         'url': 'https://www.linkedin.com/learning/programming-foundations-fundamentals',
 140         'info_dict': {
 141             'id': 'programming-foundations-fundamentals',
 142             'title': 'Programming Foundations: Fundamentals',
 143             'description': 'md5:76e580b017694eb89dc8e8923fff5c86',
 144         },
 145         'playlist_mincount': 61,
 146     }
 147
 148     @classmethod
 149     def suitable(cls, url):
 150         return False if LinkedInLearningIE.suitable(url) else super(LinkedInLearningCourseIE, cls).suitable(url)
 151
 152     def _real_extract(self, url):
 153         course_slug = self._match_id(url)
 154         course_data = self._call_api(course_slug, 'chapters,description,title')
 155
 156         entries = []
 157         for chapter in course_data.get('chapters', []):
 158             chapter_title = chapter.get('title')
 159             for video in chapter.get('videos', []):
 160                 video_slug = video.get('slug')
 161                 if not video_slug:
 162                     continue
 163                 entries.append({
 164                     '_type': 'url_transparent',
 165                     'id': self._get_video_id(video.get('urn'), course_slug, video_slug),
 166                     'title': video.get('title'),
 167                     'url': 'https://www.linkedin.com/learning/%s/%s' % (course_slug, video_slug),
 168                     'chapter': chapter_title,
 169                     'ie_key': LinkedInLearningIE.ie_key(),
 170                 })
 171
 172         return self.playlist_result(
 173             entries, course_slug,
 174             course_data.get('title'),
 175             course_data.get('description'))