yt_dlp/extractor/rte.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..networking.exceptions import HTTPError
   5 from ..utils import (
   6     ExtractorError,
   7     float_or_none,
   8     parse_iso8601,
   9     str_or_none,
  10     try_get,
  11     unescapeHTML,
  12     url_or_none,
  13 )
  14
  15
  16 class RteBaseIE(InfoExtractor):
  17     def _real_extract(self, url):
  18         item_id = self._match_id(url)
  19
  20         info_dict = {}
  21         formats = []
  22
  23         ENDPOINTS = (
  24             'https://feeds.rasset.ie/rteavgen/player/playlist?type=iptv&format=json&showId=',
  25             'http://www.rte.ie/rteavgen/getplaylist/?type=web&format=json&id=',
  26         )
  27
  28         for num, ep_url in enumerate(ENDPOINTS, start=1):
  29             try:
  30                 data = self._download_json(ep_url + item_id, item_id)
  31             except ExtractorError as ee:
  32                 if num < len(ENDPOINTS) or formats:
  33                     continue
  34                 if isinstance(ee.cause, HTTPError) and ee.cause.status == 404:
  35                     error_info = self._parse_json(ee.cause.response.read().decode(), item_id, fatal=False)
  36                     if error_info:
  37                         raise ExtractorError(
  38                             '%s said: %s' % (self.IE_NAME, error_info['message']),
  39                             expected=True)
  40                 raise
  41
  42             # NB the string values in the JSON are stored using XML escaping(!)
  43             show = try_get(data, lambda x: x['shows'][0], dict)
  44             if not show:
  45                 continue
  46
  47             if not info_dict:
  48                 title = unescapeHTML(show['title'])
  49                 description = unescapeHTML(show.get('description'))
  50                 thumbnail = show.get('thumbnail')
  51                 duration = float_or_none(show.get('duration'), 1000)
  52                 timestamp = parse_iso8601(show.get('published'))
  53                 info_dict = {
  54                     'id': item_id,
  55                     'title': title,
  56                     'description': description,
  57                     'thumbnail': thumbnail,
  58                     'timestamp': timestamp,
  59                     'duration': duration,
  60                 }
  61
  62             mg = try_get(show, lambda x: x['media:group'][0], dict)
  63             if not mg:
  64                 continue
  65
  66             if mg.get('url'):
  67                 m = re.match(r'(?P<url>rtmpe?://[^/]+)/(?P<app>.+)/(?P<playpath>mp4:.*)', mg['url'])
  68                 if m:
  69                     m = m.groupdict()
  70                     formats.append({
  71                         'url': m['url'] + '/' + m['app'],
  72                         'app': m['app'],
  73                         'play_path': m['playpath'],
  74                         'player_url': url,
  75                         'ext': 'flv',
  76                         'format_id': 'rtmp',
  77                     })
  78
  79             if mg.get('hls_server') and mg.get('hls_url'):
  80                 formats.extend(self._extract_m3u8_formats(
  81                     mg['hls_server'] + mg['hls_url'], item_id, 'mp4',
  82                     entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
  83
  84             if mg.get('hds_server') and mg.get('hds_url'):
  85                 formats.extend(self._extract_f4m_formats(
  86                     mg['hds_server'] + mg['hds_url'], item_id,
  87                     f4m_id='hds', fatal=False))
  88
  89             mg_rte_server = str_or_none(mg.get('rte:server'))
  90             mg_url = str_or_none(mg.get('url'))
  91             if mg_rte_server and mg_url:
  92                 hds_url = url_or_none(mg_rte_server + mg_url)
  93                 if hds_url:
  94                     formats.extend(self._extract_f4m_formats(
  95                         hds_url, item_id, f4m_id='hds', fatal=False))
  96
  97         info_dict['formats'] = formats
  98         return info_dict
  99
 100
 101 class RteIE(RteBaseIE):
 102     IE_NAME = 'rte'
 103     IE_DESC = 'Raidió Teilifís Éireann TV'
 104     _VALID_URL = r'https?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/[^/]+/(?P<id>[0-9]+)'
 105     _TEST = {
 106         'url': 'http://www.rte.ie/player/ie/show/iwitness-862/10478715/',
 107         'md5': '4a76eb3396d98f697e6e8110563d2604',
 108         'info_dict': {
 109             'id': '10478715',
 110             'ext': 'mp4',
 111             'title': 'iWitness',
 112             'thumbnail': r're:^https?://.*\.jpg$',
 113             'description': 'The spirit of Ireland, one voice and one minute at a time.',
 114             'duration': 60.046,
 115             'upload_date': '20151012',
 116             'timestamp': 1444694160,
 117         },
 118     }
 119
 120
 121 class RteRadioIE(RteBaseIE):
 122     IE_NAME = 'rte:radio'
 123     IE_DESC = 'Raidió Teilifís Éireann radio'
 124     # Radioplayer URLs have two distinct specifier formats,
 125     # the old format #!rii=<channel_id>:<id>:<playable_item_id>:<date>:
 126     # the new format #!rii=b<channel_id>_<id>_<playable_item_id>_<date>_
 127     # where the IDs are int/empty, the date is DD-MM-YYYY, and the specifier may be truncated.
 128     # An <id> uniquely defines an individual recording, and is the only part we require.
 129     _VALID_URL = r'https?://(?:www\.)?rte\.ie/radio/utils/radioplayer/rteradioweb\.html#!rii=(?:b?[0-9]*)(?:%3A|:|%5F|_)(?P<id>[0-9]+)'
 130
 131     _TESTS = [{
 132         # Old-style player URL; HLS and RTMPE formats
 133         'url': 'http://www.rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=16:10507902:2414:27-12-2015:',
 134         'md5': 'c79ccb2c195998440065456b69760411',
 135         'info_dict': {
 136             'id': '10507902',
 137             'ext': 'mp4',
 138             'title': 'Gloria',
 139             'thumbnail': r're:^https?://.*\.jpg$',
 140             'description': 'md5:9ce124a7fb41559ec68f06387cabddf0',
 141             'timestamp': 1451203200,
 142             'upload_date': '20151227',
 143             'duration': 7230.0,
 144         },
 145     }, {
 146         # New-style player URL; RTMPE formats only
 147         'url': 'http://rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=b16_3250678_8861_06-04-2012_',
 148         'info_dict': {
 149             'id': '3250678',
 150             'ext': 'flv',
 151             'title': 'The Lyric Concert with Paul Herriott',
 152             'thumbnail': r're:^https?://.*\.jpg$',
 153             'description': '',
 154             'timestamp': 1333742400,
 155             'upload_date': '20120406',
 156             'duration': 7199.016,
 157         },
 158         'params': {
 159             # rtmp download
 160             'skip_download': True,
 161         },
 162     }]