youtube_dl/extractor/rte.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     float_or_none,
   7     parse_iso8601,
   8     unescapeHTML,
   9 )
  10
  11
  12 class RteIE(InfoExtractor):
  13     IE_NAME = 'rte'
  14     IE_DESC = 'Raidió Teilifís Éireann TV'
  15     _VALID_URL = r'https?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/[^/]+/(?P<id>[0-9]+)'
  16     _TEST = {
  17         'url': 'http://www.rte.ie/player/ie/show/iwitness-862/10478715/',
  18         'info_dict': {
  19             'id': '10478715',
  20             'ext': 'flv',
  21             'title': 'Watch iWitness  online',
  22             'thumbnail': 're:^https?://.*\.jpg$',
  23             'description': 'iWitness : The spirit of Ireland, one voice and one minute at a time.',
  24             'duration': 60.046,
  25         },
  26         'params': {
  27             'skip_download': 'f4m fails with --test atm'
  28         }
  29     }
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33         webpage = self._download_webpage(url, video_id)
  34
  35         title = self._og_search_title(webpage)
  36         description = self._html_search_meta('description', webpage, 'description')
  37         duration = float_or_none(self._html_search_meta(
  38             'duration', webpage, 'duration', fatal=False), 1000)
  39
  40         thumbnail_id = self._search_regex(
  41             r'<meta name="thumbnail" content="uri:irus:(.*?)" />', webpage, 'thumbnail')
  42         thumbnail = 'http://img.rasset.ie/' + thumbnail_id + '.jpg'
  43
  44         feeds_url = self._html_search_meta("feeds-prefix", webpage, 'feeds url') + video_id
  45         json_string = self._download_json(feeds_url, video_id)
  46
  47         # f4m_url = server + relative_url
  48         f4m_url = json_string['shows'][0]['media:group'][0]['rte:server'] + json_string['shows'][0]['media:group'][0]['url']
  49         f4m_formats = self._extract_f4m_formats(f4m_url, video_id)
  50
  51         return {
  52             'id': video_id,
  53             'title': title,
  54             'formats': f4m_formats,
  55             'description': description,
  56             'thumbnail': thumbnail,
  57             'duration': duration,
  58         }
  59
  60
  61 class RteRadioIE(InfoExtractor):
  62     IE_NAME = 'rte:radio'
  63     IE_DESC = 'Raidió Teilifís Éireann radio'
  64     # Radioplayer URLs have the specifier #!rii=<channel_id>:<id>:<playable_item_id>:<date>:
  65     # where the IDs are int/empty, the date is DD-MM-YYYY, and the specifier may be truncated.
  66     # An <id> uniquely defines an individual recording, and is the only part we require.
  67     _VALID_URL = r'https?://(?:www\.)?rte\.ie/radio/utils/radioplayer/rteradioweb\.html#!rii=(?:[0-9]*)(?:%3A|:)(?P<id>[0-9]+)'
  68
  69     _TEST = {
  70         'url': 'http://www.rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=16:10507902:2414:27-12-2015:',
  71         'info_dict': {
  72             'id': '10507902',
  73             'ext': 'mp4',
  74             'title': 'Gloria',
  75             'thumbnail': 're:^https?://.*\.jpg$',
  76             'description': 'md5:9ce124a7fb41559ec68f06387cabddf0',
  77             'timestamp': 1451203200,
  78             'upload_date': '20151227',
  79             'duration': 7230.0,
  80         },
  81         'params': {
  82             'skip_download': 'f4m fails with --test atm'
  83         }
  84     }
  85
  86     def _real_extract(self, url):
  87         item_id = self._match_id(url)
  88
  89         json_string = self._download_json(
  90             'http://www.rte.ie/rteavgen/getplaylist/?type=web&format=json&id=' + item_id,
  91             item_id)
  92
  93         # NB the string values in the JSON are stored using XML escaping(!)
  94         show = json_string['shows'][0]
  95         title = unescapeHTML(show['title'])
  96         description = unescapeHTML(show.get('description'))
  97         thumbnail = show.get('thumbnail')
  98         duration = float_or_none(show.get('duration'), 1000)
  99         timestamp = parse_iso8601(show.get('published'))
 100
 101         mg = show['media:group'][0]
 102
 103         formats = []
 104
 105         if mg.get('url') and not mg['url'].startswith('rtmpe:'):
 106             formats.append({'url': mg['url']})
 107
 108         if mg.get('hls_server') and mg.get('hls_url'):
 109             formats.extend(self._extract_m3u8_formats(
 110                 mg['hls_server'] + mg['hls_url'], item_id, 'mp4',
 111                 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
 112
 113         if mg.get('hds_server') and mg.get('hds_url'):
 114             formats.extend(self._extract_f4m_formats(
 115                 mg['hds_server'] + mg['hds_url'], item_id,
 116                 f4m_id='hds', fatal=False))
 117
 118         self._sort_formats(formats)
 119
 120         return {
 121             'id': item_id,
 122             'title': title,
 123             'description': description,
 124             'thumbnail': thumbnail,
 125             'timestamp': timestamp,
 126             'duration': duration,
 127             'formats': formats,
 128         }