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 )
  16
  17
  18 class RadioCanadaIE(InfoExtractor):
  19     IE_NAME = 'radiocanada'
  20     _VALID_URL = r'(?:radiocanada:|https?://ici\.radio-canada\.ca/widgets/mediaconsole/)(?P<app_code>[^:/]+)[:/](?P<id>[0-9]+)'
  21     _TEST = {
  22         'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7184272',
  23         'info_dict': {
  24             'id': '7184272',
  25             'ext': 'flv',
  26             'title': 'Le parcours du tireur capté sur vidéo',
  27             'description': 'Images des caméras de surveillance fournies par la GRC montrant le parcours du tireur d\'Ottawa',
  28             'upload_date': '20141023',
  29         },
  30         'params': {
  31             # rtmp download
  32             'skip_download': True,
  33         },
  34     }
  35
  36     def _real_extract(self, url):
  37         app_code, video_id = re.match(self._VALID_URL, url).groups()
  38
  39         formats = []
  40         # TODO: extract m3u8 and f4m formats
  41         # m3u8 formats can be extracted using ipad device_type return 403 error code when ffmpeg try to download segements
  42         # f4m formats can be extracted using flashhd device_type but they produce unplayable file
  43         for device_type in ('flash',):
  44             v_data = self._download_xml(
  45                 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx',
  46                 video_id, note='Downloading %s XML' % device_type, query={
  47                     'appCode': app_code,
  48                     'idMedia': video_id,
  49                     'connectionType': 'broadband',
  50                     'multibitrate': 'true',
  51                     'deviceType': device_type,
  52                     # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
  53                     'paysJ391wsHjbOJwvCs26toz': 'CA',
  54                     'bypasslock': 'NZt5K62gRqfc',
  55                 })
  56             v_url = xpath_text(v_data, 'url')
  57             if not v_url:
  58                 continue
  59             if v_url == 'null':
  60                 raise ExtractorError('%s said: %s' % (
  61                     self.IE_NAME, xpath_text(v_data, 'message')), expected=True)
  62             ext = determine_ext(v_url)
  63             if ext == 'm3u8':
  64                 formats.extend(self._extract_m3u8_formats(
  65                     v_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
  66             elif ext == 'f4m':
  67                 formats.extend(self._extract_f4m_formats(v_url, video_id, f4m_id='hds', fatal=False))
  68             else:
  69                 ext = determine_ext(v_url)
  70                 bitrates = xpath_element(v_data, 'bitrates')
  71                 for url_e in bitrates.findall('url'):
  72                     tbr = int_or_none(url_e.get('bitrate'))
  73                     if not tbr:
  74                         continue
  75                     formats.append({
  76                         'format_id': 'rtmp-%d' % tbr,
  77                         'url': re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url),
  78                         'ext': 'flv',
  79                         'protocol': 'rtmp',
  80                         'width': int_or_none(url_e.get('width')),
  81                         'height': int_or_none(url_e.get('height')),
  82                         'tbr': tbr,
  83                     })
  84         self._sort_formats(formats)
  85
  86         metadata = self._download_xml(
  87             'http://api.radio-canada.ca/metaMedia/v1/index.ashx',
  88             video_id, note='Downloading metadata XML', query={
  89                 'appCode': app_code,
  90                 'idMedia': video_id,
  91             })
  92
  93         def get_meta(name):
  94             el = find_xpath_attr(metadata, './/Meta', 'name', name)
  95             return el.text if el is not None else None
  96
  97         return {
  98             'id': video_id,
  99             'title': get_meta('Title'),
 100             'description': get_meta('Description') or get_meta('ShortDescription'),
 101             'thumbnail': get_meta('imageHR') or get_meta('imageMR') or get_meta('imageBR'),
 102             'duration': int_or_none(get_meta('length')),
 103             'series': get_meta('Emission'),
 104             'season_number': int_or_none('SrcSaison'),
 105             'episode_number': int_or_none('SrcEpisode'),
 106             'upload_date': unified_strdate(get_meta('Date')),
 107             'formats': formats,
 108         }
 109
 110
 111 class RadioCanadaAudioVideoIE(InfoExtractor):
 112     'radiocanada:audiovideo'
 113     _VALID_URL = r'https?://ici\.radio-canada\.ca/audio-video/media-(?P<id>[0-9]+)'
 114     _TEST = {
 115         'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
 116         'info_dict': {
 117             'id': '7527184',
 118             'ext': 'flv',
 119             'title': 'Barack Obama au Vietnam',
 120             'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam',
 121             'upload_date': '20160523',
 122         },
 123         'params': {
 124             # rtmp download
 125             'skip_download': True,
 126         },
 127     }
 128
 129     def _real_extract(self, url):
 130         return self.url_result('radiocanada:medianet:%s' % self._match_id(url))