_ Git - youtube-dl/blob - 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 ExtractorError
   9
  10
  11 class LyndaIE(SubtitlesInfoExtractor):
  12     IE_NAME = 'lynda'
  13     IE_DESC = 'lynda.com videos'
  14     _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html'
  15
  16     _TEST = {
  17         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
  18         'file': '114408.mp4',
  19         'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
  20         'info_dict': {
  21             'title': 'Using the exercise files',
  22             'duration': 68
  23         }
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28         video_id = mobj.group(1)
  29
  30         page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
  31                                       video_id, 'Downloading video JSON')
  32         video_json = json.loads(page)
  33
  34         if 'Status' in video_json and video_json['Status'] == 'NotFound':
  35             raise ExtractorError('Video %s does not exist' % video_id, expected=True)
  36
  37         if video_json['HasAccess'] is False:
  38             raise ExtractorError('Video %s is only available for members' % video_id, expected=True)
  39
  40         video_id = video_json['ID']
  41         duration = video_json['DurationInSeconds']
  42         title = video_json['Title']
  43
  44         formats = [{'url': fmt['Url'],
  45                     'ext': fmt['Extension'],
  46                     'width': fmt['Width'],
  47                     'height': fmt['Height'],
  48                     'filesize': fmt['FileSize'],
  49                     'format_id': str(fmt['Resolution'])
  50                     } for fmt in video_json['Formats']]
  51
  52         self._sort_formats(formats)
  53
  54         if self._downloader.params.get('listsubtitles', False):
  55             self._list_available_subtitles(video_id, page)
  56             return
  57
  58         subtitles = self._fix_subtitles(self.extract_subtitles(video_id, page))
  59
  60         return {
  61             'id': video_id,
  62             'title': title,
  63             'duration': duration,
  64             'subtitles': subtitles,
  65             'formats': formats
  66         }
  67
  68     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
  69
  70     def _fix_subtitles(self, subtitles):
  71         if subtitles is None:
  72             return subtitles  # subtitles not requested
  73
  74         fixed_subtitles = {}
  75         for k, v in subtitles.items():
  76             subs = json.loads(v)
  77             if len(subs) == 0:
  78                 continue
  79             srt = ''
  80             for pos in range(0, len(subs) - 1):
  81                 seq_current = subs[pos]
  82                 m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
  83                 if m_current is None:
  84                     continue
  85                 seq_next = subs[pos + 1]
  86                 m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
  87                 if m_next is None:
  88                     continue
  89                 appear_time = m_current.group('timecode')
  90                 disappear_time = m_next.group('timecode')
  91                 text = seq_current['Caption']
  92                 srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text)
  93             if srt:
  94                 fixed_subtitles[k] = srt
  95         return fixed_subtitles
  96
  97     def _get_available_subtitles(self, video_id, webpage):
  98         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
  99         sub = self._download_webpage(url, None, note=False)
 100         sub_json = json.loads(sub)
 101         return {'en': url} if len(sub_json) > 0 else {}
 102
 103
 104 class LyndaCourseIE(InfoExtractor):
 105     IE_NAME = 'lynda:course'
 106     IE_DESC = 'lynda.com online courses'
 107
 108     # Course link equals to welcome/introduction video link of same course
 109     # We will recognize it as course link
 110     _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
 111
 112     def _real_extract(self, url):
 113         mobj = re.match(self._VALID_URL, url)
 114         course_path = mobj.group('coursepath')
 115         course_id = mobj.group('courseid')
 116
 117         page = self._download_webpage('http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 118                                       course_id, 'Downloading course JSON')
 119         course_json = json.loads(page)
 120
 121         if 'Status' in course_json and course_json['Status'] == 'NotFound':
 122             raise ExtractorError('Course %s does not exist' % course_id, expected=True)
 123
 124         unaccessible_videos = 0
 125         videos = []
 126
 127         for chapter in course_json['Chapters']:
 128             for video in chapter['Videos']:
 129                 if video['HasAccess'] is not True:
 130                     unaccessible_videos += 1
 131                     continue
 132                 videos.append(video['ID'])
 133
 134         if unaccessible_videos > 0:
 135             self._downloader.report_warning('%s videos are only available for members and will not be downloaded' % unaccessible_videos)
 136
 137         entries = [
 138             self.url_result('http://www.lynda.com/%s/%s-4.html' %
 139                             (course_path, video_id),
 140                             'Lynda')
 141             for video_id in videos]
 142
 143         course_title = course_json['Title']
 144
 145         return self.playlist_result(entries, course_id, course_title)