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 app_code != 'toutv':
  58             device_types.append('flash')
  59         if not smuggled_data:
  60             device_types.append('android')
  61
  62         formats = []
  63         # TODO: extract f4m formats
  64         # f4m formats can be extracted using flashhd device_type but they produce unplayable file
  65         for device_type in device_types:
  66             validation_url = 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx'
  67             query = {
  68                 'appCode': app_code,
  69                 'idMedia': video_id,
  70                 'connectionType': 'broadband',
  71                 'multibitrate': 'true',
  72                 'deviceType': device_type,
  73             }
  74             if smuggled_data:
  75                 validation_url = 'https://services.radio-canada.ca/media/validation/v2/'
  76                 query.update(smuggled_data)
  77             else:
  78                 query.update({
  79                     # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
  80                     'paysJ391wsHjbOJwvCs26toz': 'CA',
  81                     'bypasslock': 'NZt5K62gRqfc',
  82                 })
  83             v_data = self._download_xml(validation_url, video_id, note='Downloading %s XML' % device_type, query=query, fatal=False)
  84             v_url = xpath_text(v_data, 'url')
  85             if not v_url:
  86                 continue
  87             if v_url == 'null':
  88                 raise ExtractorError('%s said: %s' % (
  89                     self.IE_NAME, xpath_text(v_data, 'message')), expected=True)
  90             ext = determine_ext(v_url)
  91             if ext == 'm3u8':
  92                 formats.extend(self._extract_m3u8_formats(
  93                     v_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
  94             elif ext == 'f4m':
  95                 formats.extend(self._extract_f4m_formats(
  96                     v_url, video_id, f4m_id='hds', fatal=False))
  97             else:
  98                 ext = determine_ext(v_url)
  99                 bitrates = xpath_element(v_data, 'bitrates')
 100                 for url_e in bitrates.findall('url'):
 101                     tbr = int_or_none(url_e.get('bitrate'))
 102                     if not tbr:
 103                         continue
 104                     f_url = re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url)
 105                     protocol = determine_protocol({'url': f_url})
 106                     formats.append({
 107                         'format_id': '%s-%d' % (protocol, tbr),
 108                         'url': f_url,
 109                         'ext': 'flv' if protocol == 'rtmp' else ext,
 110                         'protocol': protocol,
 111                         'width': int_or_none(url_e.get('width')),
 112                         'height': int_or_none(url_e.get('height')),
 113                         'tbr': tbr,
 114                     })
 115                     if protocol == 'rtsp':
 116                         base_url = self._search_regex(
 117                             r'rtsp://([^?]+)', f_url, 'base url', default=None)
 118                         if base_url:
 119                             base_url = 'http://' + base_url
 120                             formats.extend(self._extract_m3u8_formats(
 121                                 base_url + '/playlist.m3u8', video_id, 'mp4',
 122                                 'm3u8_native', m3u8_id='hls', fatal=False))
 123                             formats.extend(self._extract_f4m_formats(
 124                                 base_url + '/manifest.f4m', video_id,
 125                                 f4m_id='hds', fatal=False))
 126         self._sort_formats(formats)
 127
 128         subtitles = {}
 129         closed_caption_url = get_meta('closedCaption') or get_meta('closedCaptionHTML5')
 130         if closed_caption_url:
 131             subtitles['fr'] = [{
 132                 'url': closed_caption_url,
 133                 'ext': determine_ext(closed_caption_url, 'vtt'),
 134             }]
 135
 136         return {
 137             'id': video_id,
 138             'title': get_meta('Title'),
 139             'description': get_meta('Description') or get_meta('ShortDescription'),
 140             'thumbnail': get_meta('imageHR') or get_meta('imageMR') or get_meta('imageBR'),
 141             'duration': int_or_none(get_meta('length')),
 142             'series': get_meta('Emission'),
 143             'season_number': int_or_none('SrcSaison'),
 144             'episode_number': int_or_none('SrcEpisode'),
 145             'upload_date': unified_strdate(get_meta('Date')),
 146             'subtitles': subtitles,
 147             'formats': formats,
 148         }
 149
 150
 151 class RadioCanadaAudioVideoIE(InfoExtractor):
 152     'radiocanada:audiovideo'
 153     _VALID_URL = r'https?://ici\.radio-canada\.ca/audio-video/media-(?P<id>[0-9]+)'
 154     _TEST = {
 155         'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
 156         'info_dict': {
 157             'id': '7527184',
 158             'ext': 'mp4',
 159             'title': 'Barack Obama au Vietnam',
 160             'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam',
 161             'upload_date': '20160523',
 162         },
 163         'params': {
 164             # m3u8 download
 165             'skip_download': True,
 166         },
 167     }
 168
 169     def _real_extract(self, url):
 170         return self.url_result('radiocanada:medianet:%s' % self._match_id(url))