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

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     parse_duration,
   7     int_or_none,
   8 )
   9 from ..compat import compat_HTTPError
  10 import re
  11 from .bbccouk import BBCCoUkIE
  12
  13 class BBCNewsIE(BBCCoUkIE):
  14     IE_NAME = 'bbc.com'
  15     IE_DESC = 'BBC news'
  16     _VALID_URL = r'https?://(?:www\.)?(?:bbc\.co\.uk|bbc\.com)/news/(?P<id>[^/]+)'
  17
  18     mediaselector_url = 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s'
  19
  20     _TESTS = [{
  21         'url': 'http://www.bbc.com/news/world-europe-32668511',
  22         'info_dict': {
  23             'id': 'world-europe-32668511',
  24             'title': 'Russia stages massive WW2 parade despite Western boycott',
  25         },
  26         'playlist_count': 2,
  27     },{
  28         'url': 'http://www.bbc.com/news/business-28299555',
  29         'info_dict': {
  30             'id': 'business-28299555',
  31             'title': 'Farnborough Airshow: Video highlights',
  32         },
  33         'playlist_count': 9,
  34     },{
  35         'url': 'http://www.bbc.com/news/world-europe-32041533',
  36         'note': 'Video',
  37         'info_dict': {
  38             'id': 'p02mprgb',
  39             'ext': 'mp4',
  40             'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
  41             'description': 'Germanwings plane crash site in aerial video - Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
  42             'duration': 47,
  43         },
  44         'params': {
  45             'skip_download': True,
  46         }
  47     }]
  48
  49     def _real_extract(self, url):
  50         list_id = self._match_id(url)
  51         webpage = self._download_webpage(url, list_id)
  52
  53         list_title = self._html_search_regex(r'<title>(.*?)(?:\s*-\s*BBC News)?</title>', webpage, 'list title')
  54
  55         pubdate = self._html_search_regex(r'"datePublished":\s*"(\d+-\d+-\d+)', webpage, 'date', default=None)
  56         if pubdate:
  57            pubdate = pubdate.replace('-','')
  58
  59         ret = []
  60         # works with bbc.com/news/something-something-123456 articles
  61         matches = re.findall(r"data-media-meta='({[^']+})'", webpage)
  62         if not matches:
  63            # stubbornly generic extractor for {json with "image":{allvideoshavethis},etc}
  64            # in http://www.bbc.com/news/video_and_audio/international
  65            matches = re.findall(r"({[^{}]+image\":{[^}]+}[^}]+})", webpage)
  66         if not matches:
  67            raise ExtractorError('No video found', expected=True)
  68
  69         for ent in matches:
  70             jent = self._parse_json(ent,list_id)
  71
  72             programme_id = jent.get('externalId',None)
  73             xml_url = jent.get('href', None)
  74
  75             title = jent['caption']
  76             duration = parse_duration(jent.get('duration',None))
  77             description = list_title + ' - ' + jent.get('caption','')
  78             thumbnail = None
  79             if jent.has_key('image'):
  80                thumbnail=jent['image'].get('href',None)
  81
  82             if programme_id:
  83                formats, subtitles = self._download_media_selector(programme_id)
  84             elif xml_url:
  85                # Cheap fallback
  86                # http://playlists.bbc.co.uk/news/(list_id)[ABC..]/playlist.sxml
  87                xml = self._download_webpage(xml_url, programme_id, 'Downloading playlist.sxml for externalId (fallback)')
  88                programme_id = self._search_regex(r'<mediator [^>]*identifier="(.+?)"', xml, 'playlist.sxml (externalId fallback)')
  89                formats, subtitles = self._download_media_selector(programme_id)
  90             else:
  91                raise ExtractorError('data-media-meta entry has no externalId or href value.')
  92
  93             self._sort_formats(formats)
  94
  95             ret.append( {
  96                 'id': programme_id,
  97                 'uploader': 'BBC News',
  98                 'upload_date': pubdate,
  99                 'title': title,
 100                 'description': description,
 101                 'thumbnail': thumbnail,
 102                 'duration': duration,
 103                 'formats': formats,
 104                 'subtitles': subtitles,
 105             } )
 106
 107         if len(ret) > 0:
 108            return self.playlist_result(ret, list_id, list_title)
 109         raise ExtractorError('No video found', expected=True)