youtube_dl/extractor/bbcnews.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     int_or_none,
   7 )
   8 from ..compat import compat_HTTPError
   9 import re
  10 from .bbccouk import BBCCoUkIE
  11
  12 class BBCNewsIE(BBCCoUkIE):
  13     IE_NAME = 'bbc.com'
  14     IE_DESC = 'BBC news'
  15     _VALID_URL = r'https?://(?:www\.)?(?:bbc\.co\.uk|bbc\.com)/news/(?P<id>[^/]+)'
  16
  17     _TESTS = [{
  18         'url': 'http://www.bbc.com/news/world-europe-32668511',
  19         'info_dict': {
  20             'id': 'world-europe-32668511',
  21             'title': 'Russia stages massive WW2 parade despite Western boycott',
  22         },
  23         'playlist_count': 2,
  24     },{
  25         'url': 'http://www.bbc.com/news/business-28299555',
  26         'info_dict': {
  27             'id': 'business-28299555',
  28             'title': 'Farnborough Airshow: Video highlights',
  29         },
  30         'playlist_count': 9,
  31     },{
  32         'url': 'http://www.bbc.com/news/world-europe-32041533',
  33         'note': 'Video',
  34         'info_dict': {
  35             'id': 'p02mprgb',
  36             'ext': 'mp4',
  37             'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
  38             'description': 'Germanwings plane crash site in aerial video - Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
  39             'duration': 47,
  40         },
  41         'params': {
  42             'skip_download': True,
  43         }
  44     }]
  45
  46     def _duration_str2int(self, str):
  47         if not str:
  48             return None
  49         ret = re.match(r'^\d+$', str)
  50         if ret:
  51             return int(ret.group(0))
  52         ret = re.match(r'PT((?P<h>\d+)H)?((?P<m>\d+)M)?(?P<s>\d+)S$', str)
  53         if ret:
  54             total=int(ret.group('s'))
  55             if ret.group('m'):
  56                 total+=(int(ret.group('m'))*60)
  57             if ret.group('h'):
  58                 total+=(int(ret.group('h'))*3600)
  59             return total
  60         return None
  61
  62     def _download_media_selector(self, programme_id):
  63         # bbc news uses http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/ not
  64         # http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/
  65         # Could add third urlspec arg to BBCCoUkIE._download_media_selector instead of duplicating it
  66
  67         try:
  68             media_selection = self._download_xml(
  69                'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s' % programme_id,
  70                 programme_id, 'Downloading media selection XML')
  71         except ExtractorError as ee:
  72             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
  73                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
  74             else:
  75                 raise
  76         formats = []
  77         subtitles = None
  78
  79         for media in self._extract_medias(media_selection):
  80             kind = media.get('kind')
  81             if kind == 'audio':
  82                 formats.extend(self._extract_audio(media, programme_id))
  83             elif kind == 'video':
  84                 formats.extend(self._extract_video(media, programme_id))
  85             elif kind == 'captions':
  86                 subtitles = self.extract_subtitles(media, programme_id)
  87
  88         formats = []
  89         subtitles = None
  90
  91         for media in self._extract_medias(media_selection):
  92             kind = media.get('kind')
  93             if kind == 'audio':
  94                 formats.extend(self._extract_audio(media, programme_id))
  95             elif kind == 'video':
  96                 formats.extend(self._extract_video(media, programme_id))
  97             elif kind == 'captions':
  98                 subtitles = self.extract_subtitles(media, programme_id)
  99
 100         return formats, subtitles
 101
 102     def _real_extract(self, url):
 103         list_id = self._match_id(url)
 104         webpage = self._download_webpage(url, list_id)
 105
 106         list_title = self._html_search_regex(r'<title>(.*?)(?:\s*-\s*BBC News)?</title>', webpage, 'list title')
 107
 108         pubdate = self._html_search_regex(r'"datePublished":\s*"(\d+-\d+-\d+)', webpage, 'date', default=None)
 109         if pubdate:
 110            pubdate = pubdate.replace('-','')
 111
 112         ret = []
 113         # works with bbc.com/news/something-something-123456 articles
 114         matches = re.findall(r"data-media-meta='({[^']+})'", webpage)
 115         if not matches:
 116            # stubbornly generic extractor for {json with "image":{allvideoshavethis},etc}
 117            # in http://www.bbc.com/news/video_and_audio/international
 118            matches = re.findall(r"({[^{}]+image\":{[^}]+}[^}]+})", webpage)
 119         if not matches:
 120            raise ExtractorError('No video found', expected=True)
 121
 122         for ent in matches:
 123             jent = self._parse_json(ent,list_id)
 124
 125             programme_id = jent.get('externalId',None)
 126             xml_url = jent.get('href', None)
 127
 128             title = jent['caption']
 129             duration = self._duration_str2int(jent.get('duration',None))
 130             description = list_title + ' - ' + jent.get('caption','')
 131             thumbnail = None
 132             if jent.has_key('image'):
 133                thumbnail=jent['image'].get('href',None)
 134
 135             if programme_id:
 136                formats, subtitles = self._download_media_selector(programme_id)
 137             elif xml_url:
 138                # Cheap fallback
 139                # http://playlists.bbc.co.uk/news/(list_id)[ABC..]/playlist.sxml
 140                xml = self._download_webpage(xml_url, programme_id, 'Downloading playlist.sxml for externalId (fallback)')
 141                programme_id = self._search_regex(r'<mediator [^>]*identifier="(.+?)"', xml, 'playlist.sxml (externalId fallback)')
 142                formats, subtitles = self._download_media_selector(programme_id)
 143             else:
 144                raise ExtractorError('data-media-meta entry has no externalId or href value.')
 145
 146             self._sort_formats(formats)
 147
 148             ret.append( {
 149                 'id': programme_id,
 150                 'uploader': 'BBC News',
 151                 'upload_date': pubdate,
 152                 'title': title,
 153                 'description': description,
 154                 'thumbnail': thumbnail,
 155                 'duration': duration,
 156                 'formats': formats,
 157                 'subtitles': subtitles,
 158             } )
 159
 160         if len(ret) > 0:
 161            return self.playlist_result(ret, list_id, list_title)
 162         raise ExtractorError('No video found', expected=True)