youtube_dl/extractor/msn.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     determine_ext,
  10     ExtractorError,
  11     int_or_none,
  12     unescapeHTML,
  13 )
  14
  15
  16 class MSNIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?msn\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/[a-z]{2}-(?P<id>[\da-zA-Z]+)'
  18     _TESTS = [{
  19         'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/criminal-minds-shemar-moore-shares-a-touching-goodbye-message/vp-BBqQYNE',
  20         'md5': '8442f66c116cbab1ff7098f986983458',
  21         'info_dict': {
  22             'id': 'BBqQYNE',
  23             'display_id': 'criminal-minds-shemar-moore-shares-a-touching-goodbye-message',
  24             'ext': 'mp4',
  25             'title': 'Criminal Minds - Shemar Moore Shares A Touching Goodbye Message',
  26             'description': 'md5:e8e89b897b222eb33a6b5067a8f1bc25',
  27             'duration': 104,
  28             'uploader': 'CBS Entertainment',
  29             'uploader_id': 'IT0X5aoJ6bJgYerJXSDCgFmYPB1__54v',
  30         },
  31     }, {
  32         'url': 'http://www.msn.com/en-ae/news/offbeat/meet-the-nine-year-old-self-made-millionaire/ar-BBt6ZKf',
  33         'only_matching': True,
  34     }, {
  35         'url': 'http://www.msn.com/en-ae/video/watch/obama-a-lot-of-people-will-be-disappointed/vi-AAhxUMH',
  36         'only_matching': True,
  37     }, {
  38         # geo restricted
  39         'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/the-first-fart-makes-you-laugh-the-last-fart-makes-you-cry/vp-AAhzIBU',
  40         'only_matching': True,
  41     }, {
  42         'url': 'http://www.msn.com/en-ae/entertainment/bollywood/watch-how-salman-khan-reacted-when-asked-if-he-would-apologize-for-his-‘raped-woman’-comment/vi-AAhvzW6',
  43         'only_matching': True,
  44     }]
  45
  46     def _real_extract(self, url):
  47         mobj = re.match(self._VALID_URL, url)
  48         video_id, display_id = mobj.group('id', 'display_id')
  49
  50         webpage = self._download_webpage(url, display_id)
  51
  52         video = self._parse_json(
  53             self._search_regex(
  54                 r'data-metadata\s*=\s*(["\'])(?P<data>.+?)\1',
  55                 webpage, 'video data', default='{}', group='data'),
  56             display_id, transform_source=unescapeHTML)
  57
  58         if not video:
  59             error = unescapeHTML(self._search_regex(
  60                 r'data-error=(["\'])(?P<error>.+?)\1',
  61                 webpage, 'error', group='error'))
  62             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
  63
  64         title = video['title']
  65
  66         formats = []
  67         for file_ in video.get('videoFiles', []):
  68             format_url = file_.get('url')
  69             if not format_url:
  70                 continue
  71             if 'm3u8' in format_url:
  72                 # m3u8_native should not be used here until
  73                 # https://github.com/ytdl-org/youtube-dl/issues/9913 is fixed
  74                 m3u8_formats = self._extract_m3u8_formats(
  75                     format_url, display_id, 'mp4',
  76                     m3u8_id='hls', fatal=False)
  77                 formats.extend(m3u8_formats)
  78             elif determine_ext(format_url) == 'ism':
  79                 formats.extend(self._extract_ism_formats(
  80                     format_url + '/Manifest', display_id, 'mss', fatal=False))
  81             else:
  82                 formats.append({
  83                     'url': format_url,
  84                     'ext': 'mp4',
  85                     'format_id': 'http',
  86                     'width': int_or_none(file_.get('width')),
  87                     'height': int_or_none(file_.get('height')),
  88                 })
  89         self._sort_formats(formats)
  90
  91         subtitles = {}
  92         for file_ in video.get('files', []):
  93             format_url = file_.get('url')
  94             format_code = file_.get('formatCode')
  95             if not format_url or not format_code:
  96                 continue
  97             if compat_str(format_code) == '3100':
  98                 subtitles.setdefault(file_.get('culture', 'en'), []).append({
  99                     'ext': determine_ext(format_url, 'ttml'),
 100                     'url': format_url,
 101                 })
 102
 103         return {
 104             'id': video_id,
 105             'display_id': display_id,
 106             'title': title,
 107             'description': video.get('description'),
 108             'thumbnail': video.get('headlineImage', {}).get('url'),
 109             'duration': int_or_none(video.get('durationSecs')),
 110             'uploader': video.get('sourceFriendly'),
 111             'uploader_id': video.get('providerId'),
 112             'creator': video.get('creator'),
 113             'subtitles': subtitles,
 114             'formats': formats,
 115         }