Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/cbsnews.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from .theplatform import ThePlatformIE
   6 from ..utils import (
   7     parse_duration,
   8     find_xpath_attr,
   9 )
  10
  11
  12 class CBSNewsIE(ThePlatformIE):
  13     IE_DESC = 'CBS News'
  14     _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
  15
  16     _TESTS = [
  17         {
  18             'url': 'http://www.cbsnews.com/news/tesla-and-spacex-elon-musks-industrial-empire/',
  19             'info_dict': {
  20                 'id': 'tesla-and-spacex-elon-musks-industrial-empire',
  21                 'ext': 'flv',
  22                 'title': 'Tesla and SpaceX: Elon Musk\'s industrial empire',
  23                 'thumbnail': 'http://beta.img.cbsnews.com/i/2014/03/30/60147937-2f53-4565-ad64-1bdd6eb64679/60-0330-pelley-640x360.jpg',
  24                 'duration': 791,
  25             },
  26             'params': {
  27                 # rtmp download
  28                 'skip_download': True,
  29             },
  30         },
  31         {
  32             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
  33             'info_dict': {
  34                 'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
  35                 'ext': 'mp4',
  36                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
  37                 'thumbnail': 're:^https?://.*\.jpg$',
  38                 'duration': 205,
  39                 'subtitles': {
  40                     'en': [{
  41                         'ext': 'ttml',
  42                     }],
  43                 },
  44             },
  45             'params': {
  46                 # m3u8 download
  47                 'skip_download': True,
  48             },
  49         },
  50     ]
  51
  52     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
  53         closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL')
  54         return {
  55             'en': [{
  56                 'ext': 'ttml',
  57                 'url': closed_caption_e.attrib['value'],
  58             }]
  59         } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
  60
  61     def _real_extract(self, url):
  62         video_id = self._match_id(url)
  63
  64         webpage = self._download_webpage(url, video_id)
  65
  66         video_info = self._parse_json(self._html_search_regex(
  67             r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
  68             webpage, 'video JSON info'), video_id)
  69
  70         item = video_info['item'] if 'item' in video_info else video_info
  71         title = item.get('articleTitle') or item.get('hed')
  72         duration = item.get('duration')
  73         thumbnail = item.get('mediaImage') or item.get('thumbnail')
  74
  75         subtitles = {}
  76         formats = []
  77         for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
  78             pid = item.get('media' + format_id)
  79             if not pid:
  80                 continue
  81             release_url = 'http://link.theplatform.com/s/dJ5BDC/%s?format=SMIL&mbr=true' % pid
  82             tp_formats, tp_subtitles = self._extract_theplatform_smil(release_url, video_id, 'Downloading %s SMIL data' % pid)
  83             formats.extend(tp_formats)
  84             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
  85         self._sort_formats(formats)
  86
  87         return {
  88             'id': video_id,
  89             'title': title,
  90             'thumbnail': thumbnail,
  91             'duration': duration,
  92             'formats': formats,
  93             'subtitles': subtitles,
  94         }
  95
  96
  97 class CBSNewsLiveVideoIE(InfoExtractor):
  98     IE_DESC = 'CBS News Live Videos'
  99     _VALID_URL = r'http://(?:www\.)?cbsnews\.com/live/video/(?P<id>[\da-z_-]+)'
 100
 101     _TEST = {
 102         'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
 103         'info_dict': {
 104             'id': 'clinton-sanders-prepare-to-face-off-in-nh',
 105             'ext': 'flv',
 106             'title': 'Clinton, Sanders Prepare To Face Off In NH',
 107             'duration': 334,
 108         },
 109     }
 110
 111     def _real_extract(self, url):
 112         video_id = self._match_id(url)
 113
 114         webpage = self._download_webpage(url, video_id)
 115
 116         video_info = self._parse_json(self._html_search_regex(
 117             r'data-story-obj=\'({.+?})\'', webpage, 'video JSON info'), video_id)['story']
 118
 119         hdcore_sign = 'hdcore=3.3.1'
 120         f4m_formats = self._extract_f4m_formats(video_info['url'] + '&' + hdcore_sign, video_id)
 121         if f4m_formats:
 122             for entry in f4m_formats:
 123                 # URLs without the extra param induce an 404 error
 124                 entry.update({'extra_param_to_segment_url': hdcore_sign})
 125
 126         return {
 127             'id': video_id,
 128             'title': video_info['headline'],
 129             'thumbnail': video_info.get('thumbnail_url_hd') or video_info.get('thumbnail_url_sd'),
 130             'duration': parse_duration(video_info.get('segmentDur')),
 131             'formats': f4m_formats,
 132         }