_ Git - youtube-dl/blob - youtube_dl/extractor/wsj.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     float_or_none,
   8     unified_strdate,
   9 )
  10
  11
  12 class WSJIE(InfoExtractor):
  13     _VALID_URL = r'''(?x)
  14         (?:
  15             https?://video-api\.wsj\.com/api-video/player/iframe\.html\?guid=|
  16             https?://(?:www\.)?wsj\.com/video/[^/]+/|
  17             wsj:
  18         )
  19         (?P<id>[a-zA-Z0-9-]+)'''
  20     IE_DESC = 'Wall Street Journal'
  21     _TESTS = [{
  22         'url': 'http://video-api.wsj.com/api-video/player/iframe.html?guid=1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A',
  23         'md5': 'e230a5bb249075e40793b655a54a02e4',
  24         'info_dict': {
  25             'id': '1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A',
  26             'ext': 'mp4',
  27             'upload_date': '20150202',
  28             'uploader_id': 'jdesai',
  29             'creator': 'jdesai',
  30             'categories': list,  # a long list
  31             'duration': 90,
  32             'title': 'Bills Coach Rex Ryan Updates His Old Jets Tattoo',
  33         },
  34     }, {
  35         'url': 'http://www.wsj.com/video/can-alphabet-build-a-smarter-city/359DDAA8-9AC1-489C-82E6-0429C1E430E0.html',
  36         'only_matching': True,
  37     }]
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41
  42         api_url = (
  43             'http://video-api.wsj.com/api-video/find_all_videos.asp?'
  44             'type=guid&count=1&query=%s&fields=type,hls,videoMP4List,'
  45             'thumbnailList,author,description,name,duration,videoURL,'
  46             'titletag,formattedCreationDate,keywords,editor' % video_id)
  47         info = self._download_json(api_url, video_id)['items'][0]
  48         title = info.get('name', info.get('titletag'))
  49
  50         formats = []
  51
  52         f4m_url = info.get('videoURL')
  53         if f4m_url:
  54             formats.extend(self._extract_f4m_formats(
  55                 f4m_url, video_id, f4m_id='hds', fatal=False))
  56
  57         m3u8_url = info.get('hls')
  58         if m3u8_url:
  59             formats.extend(self._extract_m3u8_formats(
  60                 info['hls'], video_id, ext='mp4',
  61                 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
  62
  63         for v in info.get('videoMP4List', []):
  64             mp4_url = v.get('url')
  65             if not mp4_url:
  66                 continue
  67             tbr = int_or_none(v.get('bitrate'))
  68             formats.append({
  69                 'url': mp4_url,
  70                 'format_id': 'http' + ('-%d' % tbr if tbr else ''),
  71                 'tbr': tbr,
  72                 'width': int_or_none(v.get('width')),
  73                 'height': int_or_none(v.get('height')),
  74                 'fps': float_or_none(v.get('fps')),
  75             })
  76         self._sort_formats(formats)
  77
  78         return {
  79             'id': video_id,
  80             'formats': formats,
  81             # Thumbnails are conveniently in the correct format already
  82             'thumbnails': info.get('thumbnailList'),
  83             'creator': info.get('author'),
  84             'uploader_id': info.get('editor'),
  85             'duration': int_or_none(info.get('duration')),
  86             'upload_date': unified_strdate(info.get(
  87                 'formattedCreationDate'), day_first=False),
  88             'title': title,
  89             'categories': info.get('keywords'),
  90         }
  91
  92
  93 class WSJArticleIE(InfoExtractor):
  94     _VALID_URL = r'(?i)https?://(?:www\.)?wsj\.com/articles/(?P<id>\w[^/]+)'
  95     _TESTS = [{
  96         'url': 'https://www.wsj.com/articles/dont-like-china-no-pandas-for-you-1490366939?',
  97         'info_dict': {
  98             'id': '4B13FA62-1D8C-45DB-8EA1-4105CB20B362',
  99             'ext': 'mp4',
 100             'upload_date': '20170221',
 101             'uploader_id': 'ralcaraz',
 102             'title': 'Bao Bao the Panda Leaves for China',
 103         }
 104     }]
 105
 106     def _real_extract(self, url):
 107         article_id = self._match_id(url)
 108         webpage = self._download_webpage(url, article_id)
 109         video_id = self._search_regex(r'data-src=["\']([A-Z0-9\-]+)',
 110                                       webpage, 'video id')
 111         return self.url_result('wsj:%s' % video_id, WSJIE.ie_key(), video_id)