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)https?://
  14         (?:
  15             video-api\.wsj\.com/api-video/player/iframe\.html\?guid=|
  16             (?:www\.)?wsj\.com/video/[^/]+/
  17         )
  18         (?P<id>[a-zA-Z0-9-]+)'''
  19     IE_DESC = 'Wall Street Journal'
  20     _TESTS = [{
  21         'url': 'http://video-api.wsj.com/api-video/player/iframe.html?guid=1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A',
  22         'md5': 'e230a5bb249075e40793b655a54a02e4',
  23         'info_dict': {
  24             'id': '1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A',
  25             'ext': 'mp4',
  26             'upload_date': '20150202',
  27             'uploader_id': 'jdesai',
  28             'creator': 'jdesai',
  29             'categories': list,  # a long list
  30             'duration': 90,
  31             'title': 'Bills Coach Rex Ryan Updates His Old Jets Tattoo',
  32         },
  33     }, {
  34         'url': 'http://www.wsj.com/video/can-alphabet-build-a-smarter-city/359DDAA8-9AC1-489C-82E6-0429C1E430E0.html',
  35         'only_matching': True,
  36     }]
  37
  38     def _real_extract(self, url):
  39         video_id = self._match_id(url)
  40
  41         api_url = (
  42             'http://video-api.wsj.com/api-video/find_all_videos.asp?'
  43             'type=guid&count=1&query=%s&fields=type,hls,videoMP4List,'
  44             'thumbnailList,author,description,name,duration,videoURL,'
  45             'titletag,formattedCreationDate,keywords,editor' % video_id)
  46         info = self._download_json(api_url, video_id)['items'][0]
  47         title = info.get('name', info.get('titletag'))
  48
  49         formats = []
  50
  51         f4m_url = info.get('videoURL')
  52         if f4m_url:
  53             formats.extend(self._extract_f4m_formats(
  54                 f4m_url, video_id, f4m_id='hds', fatal=False))
  55
  56         m3u8_url = info.get('hls')
  57         if m3u8_url:
  58             formats.extend(self._extract_m3u8_formats(
  59                 info['hls'], video_id, ext='mp4',
  60                 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
  61
  62         for v in info.get('videoMP4List', []):
  63             mp4_url = v.get('url')
  64             if not mp4_url:
  65                 continue
  66             tbr = int_or_none(v.get('bitrate'))
  67             formats.append({
  68                 'url': mp4_url,
  69                 'format_id': 'http' + ('-%d' % tbr if tbr else ''),
  70                 'tbr': tbr,
  71                 'width': int_or_none(v.get('width')),
  72                 'height': int_or_none(v.get('height')),
  73                 'fps': float_or_none(v.get('fps')),
  74             })
  75         self._sort_formats(formats)
  76
  77         return {
  78             'id': video_id,
  79             'formats': formats,
  80             # Thumbnails are conveniently in the correct format already
  81             'thumbnails': info.get('thumbnailList'),
  82             'creator': info.get('author'),
  83             'uploader_id': info.get('editor'),
  84             'duration': int_or_none(info.get('duration')),
  85             'upload_date': unified_strdate(info.get(
  86                 'formattedCreationDate'), day_first=False),
  87             'title': title,
  88             'categories': info.get('keywords'),
  89         }