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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     xpath_text,
   9     find_xpath_attr,
  10     determine_ext,
  11     int_or_none,
  12     unified_strdate,
  13     xpath_element,
  14     ExtractorError,
  15     determine_protocol,
  16     unsmuggle_url,
  17 )
  18
  19
  20 class RadioCanadaIE(InfoExtractor):
  21     IE_NAME = 'radiocanada'
  22     _VALID_URL = r'(?:radiocanada:|https?://ici\.radio-canada\.ca/widgets/mediaconsole/)(?P<app_code>[^:/]+)[:/](?P<id>[0-9]+)'
  23     _TEST = {
  24         'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7184272',
  25         'info_dict': {
  26             'id': '7184272',
  27             'ext': 'mp4',
  28             'title': 'Le parcours du tireur capté sur vidéo',
  29             'description': 'Images des caméras de surveillance fournies par la GRC montrant le parcours du tireur d\'Ottawa',
  30             'upload_date': '20141023',
  31         },
  32         'params': {
  33             # m3u8 download
  34             'skip_download': True,
  35         },
  36     }
  37
  38     def _real_extract(self, url):
  39         url, smuggled_data = unsmuggle_url(url, {})
  40         app_code, video_id = re.match(self._VALID_URL, url).groups()
  41
  42         metadata = self._download_xml(
  43             'http://api.radio-canada.ca/metaMedia/v1/index.ashx',
  44             video_id, note='Downloading metadata XML', query={
  45                 'appCode': app_code,
  46                 'idMedia': video_id,
  47             })
  48
  49         def get_meta(name):
  50             el = find_xpath_attr(metadata, './/Meta', 'name', name)
  51             return el.text if el is not None else None
  52
  53         if get_meta('protectionType'):
  54             raise ExtractorError('This video is DRM protected.', expected=True)
  55
  56         device_types = ['ipad']
  57         if not smuggled_data:
  58             device_types.append('flash')
  59             device_types.append('android')
  60
  61         formats = []
  62         # TODO: extract f4m formats
  63         # f4m formats can be extracted using flashhd device_type but they produce unplayable file
  64         for device_type in device_types:
  65             validation_url = 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx'
  66             query = {
  67                 'appCode': app_code,
  68                 'idMedia': video_id,
  69                 'connectionType': 'broadband',
  70                 'multibitrate': 'true',
  71                 'deviceType': device_type,
  72             }
  73             if smuggled_data:
  74                 validation_url = 'https://services.radio-canada.ca/media/validation/v2/'
  75                 query.update(smuggled_data)
  76             else:
  77                 query.update({
  78                     # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
  79                     'paysJ391wsHjbOJwvCs26toz': 'CA',
  80                     'bypasslock': 'NZt5K62gRqfc',
  81                 })
  82             v_data = self._download_xml(validation_url, video_id, note='Downloading %s XML' % device_type, query=query, fatal=False)
  83             v_url = xpath_text(v_data, 'url')
  84             if not v_url:
  85                 continue
  86             if v_url == 'null':
  87                 raise ExtractorError('%s said: %s' % (
  88                     self.IE_NAME, xpath_text(v_data, 'message')), expected=True)
  89             ext = determine_ext(v_url)
  90             if ext == 'm3u8':
  91                 formats.extend(self._extract_m3u8_formats(
  92                     v_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
  93             elif ext == 'f4m':
  94                 formats.extend(self._extract_f4m_formats(
  95                     v_url, video_id, f4m_id='hds', fatal=False))
  96             else:
  97                 ext = determine_ext(v_url)
  98                 bitrates = xpath_element(v_data, 'bitrates')
  99                 for url_e in bitrates.findall('url'):
 100                     tbr = int_or_none(url_e.get('bitrate'))
 101                     if not tbr:
 102                         continue
 103                     f_url = re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url)
 104                     protocol = determine_protocol({'url': f_url})
 105                     f = {
 106                         'format_id': '%s-%d' % (protocol, tbr),
 107                         'url': f_url,
 108                         'ext': 'flv' if protocol == 'rtmp' else ext,
 109                         'protocol': protocol,
 110                         'width': int_or_none(url_e.get('width')),
 111                         'height': int_or_none(url_e.get('height')),
 112                         'tbr': tbr,
 113                     }
 114                     mobj = re.match(r'(?P<url>rtmp://[^/]+/[^/]+)/(?P<playpath>[^?]+)(?P<auth>\?.+)', f_url)
 115                     if mobj:
 116                         f.update({
 117                             'url': mobj.group('url') + mobj.group('auth'),
 118                             'play_path': mobj.group('playpath'),
 119                         })
 120                     formats.append(f)
 121                     if protocol == 'rtsp':
 122                         base_url = self._search_regex(
 123                             r'rtsp://([^?]+)', f_url, 'base url', default=None)
 124                         if base_url:
 125                             base_url = 'http://' + base_url
 126                             formats.extend(self._extract_m3u8_formats(
 127                                 base_url + '/playlist.m3u8', video_id, 'mp4',
 128                                 'm3u8_native', m3u8_id='hls', fatal=False))
 129                             formats.extend(self._extract_f4m_formats(
 130                                 base_url + '/manifest.f4m', video_id,
 131                                 f4m_id='hds', fatal=False))
 132         self._sort_formats(formats)
 133
 134         subtitles = {}
 135         closed_caption_url = get_meta('closedCaption') or get_meta('closedCaptionHTML5')
 136         if closed_caption_url:
 137             subtitles['fr'] = [{
 138                 'url': closed_caption_url,
 139                 'ext': determine_ext(closed_caption_url, 'vtt'),
 140             }]
 141
 142         return {
 143             'id': video_id,
 144             'title': get_meta('Title'),
 145             'description': get_meta('Description') or get_meta('ShortDescription'),
 146             'thumbnail': get_meta('imageHR') or get_meta('imageMR') or get_meta('imageBR'),
 147             'duration': int_or_none(get_meta('length')),
 148             'series': get_meta('Emission'),
 149             'season_number': int_or_none('SrcSaison'),
 150             'episode_number': int_or_none('SrcEpisode'),
 151             'upload_date': unified_strdate(get_meta('Date')),
 152             'subtitles': subtitles,
 153             'formats': formats,
 154         }
 155
 156
 157 class RadioCanadaAudioVideoIE(InfoExtractor):
 158     'radiocanada:audiovideo'
 159     _VALID_URL = r'https?://ici\.radio-canada\.ca/audio-video/media-(?P<id>[0-9]+)'
 160     _TEST = {
 161         'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
 162         'info_dict': {
 163             'id': '7527184',
 164             'ext': 'mp4',
 165             'title': 'Barack Obama au Vietnam',
 166             'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam',
 167             'upload_date': '20160523',
 168         },
 169         'params': {
 170             # m3u8 download
 171             'skip_download': True,
 172         },
 173     }
 174
 175     def _real_extract(self, url):
 176         return self.url_result('radiocanada:medianet:%s' % self._match_id(url))