Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/rte.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     float_or_none,
   9     parse_iso8601,
  10     unescapeHTML,
  11 )
  12
  13
  14 class RteIE(InfoExtractor):
  15     IE_NAME = 'rte'
  16     IE_DESC = 'Raidió Teilifís Éireann TV'
  17     _VALID_URL = r'https?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/[^/]+/(?P<id>[0-9]+)'
  18     _TEST = {
  19         'url': 'http://www.rte.ie/player/ie/show/iwitness-862/10478715/',
  20         'info_dict': {
  21             'id': '10478715',
  22             'ext': 'flv',
  23             'title': 'Watch iWitness  online',
  24             'thumbnail': 're:^https?://.*\.jpg$',
  25             'description': 'iWitness : The spirit of Ireland, one voice and one minute at a time.',
  26             'duration': 60.046,
  27         },
  28         'params': {
  29             'skip_download': 'f4m fails with --test atm'
  30         }
  31     }
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35         webpage = self._download_webpage(url, video_id)
  36
  37         title = self._og_search_title(webpage)
  38         description = self._html_search_meta('description', webpage, 'description')
  39         duration = float_or_none(self._html_search_meta(
  40             'duration', webpage, 'duration', fatal=False), 1000)
  41
  42         thumbnail_id = self._search_regex(
  43             r'<meta name="thumbnail" content="uri:irus:(.*?)" />', webpage, 'thumbnail')
  44         thumbnail = 'http://img.rasset.ie/' + thumbnail_id + '.jpg'
  45
  46         feeds_url = self._html_search_meta('feeds-prefix', webpage, 'feeds url') + video_id
  47         json_string = self._download_json(feeds_url, video_id)
  48
  49         # f4m_url = server + relative_url
  50         f4m_url = json_string['shows'][0]['media:group'][0]['rte:server'] + json_string['shows'][0]['media:group'][0]['url']
  51         f4m_formats = self._extract_f4m_formats(f4m_url, video_id)
  52
  53         return {
  54             'id': video_id,
  55             'title': title,
  56             'formats': f4m_formats,
  57             'description': description,
  58             'thumbnail': thumbnail,
  59             'duration': duration,
  60         }
  61
  62
  63 class RteRadioIE(InfoExtractor):
  64     IE_NAME = 'rte:radio'
  65     IE_DESC = 'Raidió Teilifís Éireann radio'
  66     # Radioplayer URLs have two distinct specifier formats,
  67     # the old format #!rii=<channel_id>:<id>:<playable_item_id>:<date>:
  68     # the new format #!rii=b<channel_id>_<id>_<playable_item_id>_<date>_
  69     # where the IDs are int/empty, the date is DD-MM-YYYY, and the specifier may be truncated.
  70     # An <id> uniquely defines an individual recording, and is the only part we require.
  71     _VALID_URL = r'https?://(?:www\.)?rte\.ie/radio/utils/radioplayer/rteradioweb\.html#!rii=(?:b?[0-9]*)(?:%3A|:|%5F|_)(?P<id>[0-9]+)'
  72
  73     _TESTS = [{
  74         # Old-style player URL; HLS and RTMPE formats
  75         'url': 'http://www.rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=16:10507902:2414:27-12-2015:',
  76         'info_dict': {
  77             'id': '10507902',
  78             'ext': 'mp4',
  79             'title': 'Gloria',
  80             'thumbnail': 're:^https?://.*\.jpg$',
  81             'description': 'md5:9ce124a7fb41559ec68f06387cabddf0',
  82             'timestamp': 1451203200,
  83             'upload_date': '20151227',
  84             'duration': 7230.0,
  85         },
  86         'params': {
  87             'skip_download': 'f4m fails with --test atm'
  88         }
  89     }, {
  90         # New-style player URL; RTMPE formats only
  91         'url': 'http://rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=b16_3250678_8861_06-04-2012_',
  92         'info_dict': {
  93             'id': '3250678',
  94             'ext': 'flv',
  95             'title': 'The Lyric Concert with Paul Herriott',
  96             'thumbnail': 're:^https?://.*\.jpg$',
  97             'description': '',
  98             'timestamp': 1333742400,
  99             'upload_date': '20120406',
 100             'duration': 7199.016,
 101         },
 102         'params': {
 103             'skip_download': 'f4m fails with --test atm'
 104         }
 105     }]
 106
 107     def _real_extract(self, url):
 108         item_id = self._match_id(url)
 109
 110         json_string = self._download_json(
 111             'http://www.rte.ie/rteavgen/getplaylist/?type=web&format=json&id=' + item_id,
 112             item_id)
 113
 114         # NB the string values in the JSON are stored using XML escaping(!)
 115         show = json_string['shows'][0]
 116         title = unescapeHTML(show['title'])
 117         description = unescapeHTML(show.get('description'))
 118         thumbnail = show.get('thumbnail')
 119         duration = float_or_none(show.get('duration'), 1000)
 120         timestamp = parse_iso8601(show.get('published'))
 121
 122         mg = show['media:group'][0]
 123
 124         formats = []
 125
 126         if mg.get('url'):
 127             m = re.match(r'(?P<url>rtmpe?://[^/]+)/(?P<app>.+)/(?P<playpath>mp4:.*)', mg['url'])
 128             if m:
 129                 m = m.groupdict()
 130                 formats.append({
 131                     'url': m['url'] + '/' + m['app'],
 132                     'app': m['app'],
 133                     'play_path': m['playpath'],
 134                     'player_url': url,
 135                     'ext': 'flv',
 136                     'format_id': 'rtmp',
 137                 })
 138
 139         if mg.get('hls_server') and mg.get('hls_url'):
 140             formats.extend(self._extract_m3u8_formats(
 141                 mg['hls_server'] + mg['hls_url'], item_id, 'mp4',
 142                 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
 143
 144         if mg.get('hds_server') and mg.get('hds_url'):
 145             formats.extend(self._extract_f4m_formats(
 146                 mg['hds_server'] + mg['hds_url'], item_id,
 147                 f4m_id='hds', fatal=False))
 148
 149         self._sort_formats(formats)
 150
 151         return {
 152             'id': item_id,
 153             'title': title,
 154             'description': description,
 155             'thumbnail': thumbnail,
 156             'timestamp': timestamp,
 157             'duration': duration,
 158             'formats': formats,
 159         }