_ 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     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     mediaselector_url = 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s'
  18
  19     _TESTS = [{
  20         'url': 'http://www.bbc.com/news/world-europe-32668511',
  21         'info_dict': {
  22             'id': 'world-europe-32668511',
  23             'title': 'Russia stages massive WW2 parade despite Western boycott',
  24         },
  25         'playlist_count': 2,
  26     },{
  27         'url': 'http://www.bbc.com/news/business-28299555',
  28         'info_dict': {
  29             'id': 'business-28299555',
  30             'title': 'Farnborough Airshow: Video highlights',
  31         },
  32         'playlist_count': 9,
  33     },{
  34         'url': 'http://www.bbc.com/news/world-europe-32041533',
  35         'note': 'Video',
  36         'info_dict': {
  37             'id': 'p02mprgb',
  38             'ext': 'mp4',
  39             'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
  40             'description': 'Germanwings plane crash site in aerial video - Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
  41             'duration': 47,
  42         },
  43         'params': {
  44             'skip_download': True,
  45         }
  46     }]
  47
  48     def _duration_str2int(self, str):
  49         if not str:
  50             return None
  51         ret = re.match(r'^\d+$', str)
  52         if ret:
  53             return int(ret.group(0))
  54         ret = re.match(r'PT((?P<h>\d+)H)?((?P<m>\d+)M)?(?P<s>\d+)S$', str)
  55         if ret:
  56             total=int(ret.group('s'))
  57             if ret.group('m'):
  58                 total+=(int(ret.group('m'))*60)
  59             if ret.group('h'):
  60                 total+=(int(ret.group('h'))*3600)
  61             return total
  62         return None
  63
  64     def _real_extract(self, url):
  65         list_id = self._match_id(url)
  66         webpage = self._download_webpage(url, list_id)
  67
  68         list_title = self._html_search_regex(r'<title>(.*?)(?:\s*-\s*BBC News)?</title>', webpage, 'list title')
  69
  70         pubdate = self._html_search_regex(r'"datePublished":\s*"(\d+-\d+-\d+)', webpage, 'date', default=None)
  71         if pubdate:
  72            pubdate = pubdate.replace('-','')
  73
  74         ret = []
  75         # works with bbc.com/news/something-something-123456 articles
  76         matches = re.findall(r"data-media-meta='({[^']+})'", webpage)
  77         if not matches:
  78            # stubbornly generic extractor for {json with "image":{allvideoshavethis},etc}
  79            # in http://www.bbc.com/news/video_and_audio/international
  80            matches = re.findall(r"({[^{}]+image\":{[^}]+}[^}]+})", webpage)
  81         if not matches:
  82            raise ExtractorError('No video found', expected=True)
  83
  84         for ent in matches:
  85             jent = self._parse_json(ent,list_id)
  86
  87             programme_id = jent.get('externalId',None)
  88             xml_url = jent.get('href', None)
  89
  90             title = jent['caption']
  91             duration = self._duration_str2int(jent.get('duration',None))
  92             description = list_title + ' - ' + jent.get('caption','')
  93             thumbnail = None
  94             if jent.has_key('image'):
  95                thumbnail=jent['image'].get('href',None)
  96
  97             if programme_id:
  98                formats, subtitles = self._download_media_selector(programme_id)
  99             elif xml_url:
 100                # Cheap fallback
 101                # http://playlists.bbc.co.uk/news/(list_id)[ABC..]/playlist.sxml
 102                xml = self._download_webpage(xml_url, programme_id, 'Downloading playlist.sxml for externalId (fallback)')
 103                programme_id = self._search_regex(r'<mediator [^>]*identifier="(.+?)"', xml, 'playlist.sxml (externalId fallback)')
 104                formats, subtitles = self._download_media_selector(programme_id)
 105             else:
 106                raise ExtractorError('data-media-meta entry has no externalId or href value.')
 107
 108             self._sort_formats(formats)
 109
 110             ret.append( {
 111                 'id': programme_id,
 112                 'uploader': 'BBC News',
 113                 'upload_date': pubdate,
 114                 'title': title,
 115                 'description': description,
 116                 'thumbnail': thumbnail,
 117                 'duration': duration,
 118                 'formats': formats,
 119                 'subtitles': subtitles,
 120             } )
 121
 122         if len(ret) > 0:
 123            return self.playlist_result(ret, list_id, list_title)
 124         raise ExtractorError('No video found', expected=True)