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 ..compat import compat_HTTPError
   8 from ..utils import (
   9     float_or_none,
  10     parse_iso8601,
  11     unescapeHTML,
  12     ExtractorError,
  13 )
  14
  15
  16 class RteBaseIE(InfoExtractor):
  17     def _real_extract(self, url):
  18         item_id = self._match_id(url)
  19
  20         try:
  21             json_string = self._download_json(
  22                 'http://www.rte.ie/rteavgen/getplaylist/?type=web&format=json&id=' + item_id,
  23                 item_id)
  24         except ExtractorError as ee:
  25             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
  26                 error_info = self._parse_json(ee.cause.read().decode(), item_id, fatal=False)
  27                 if error_info:
  28                     raise ExtractorError(
  29                         '%s said: %s' % (self.IE_NAME, error_info['message']),
  30                         expected=True)
  31             raise
  32
  33         # NB the string values in the JSON are stored using XML escaping(!)
  34         show = json_string['shows'][0]
  35         title = unescapeHTML(show['title'])
  36         description = unescapeHTML(show.get('description'))
  37         thumbnail = show.get('thumbnail')
  38         duration = float_or_none(show.get('duration'), 1000)
  39         timestamp = parse_iso8601(show.get('published'))
  40
  41         mg = show['media:group'][0]
  42
  43         formats = []
  44
  45         if mg.get('url'):
  46             m = re.match(r'(?P<url>rtmpe?://[^/]+)/(?P<app>.+)/(?P<playpath>mp4:.*)', mg['url'])
  47             if m:
  48                 m = m.groupdict()
  49                 formats.append({
  50                     'url': m['url'] + '/' + m['app'],
  51                     'app': m['app'],
  52                     'play_path': m['playpath'],
  53                     'player_url': url,
  54                     'ext': 'flv',
  55                     'format_id': 'rtmp',
  56                 })
  57
  58         if mg.get('hls_server') and mg.get('hls_url'):
  59             formats.extend(self._extract_m3u8_formats(
  60                 mg['hls_server'] + mg['hls_url'], item_id, 'mp4',
  61                 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
  62
  63         if mg.get('hds_server') and mg.get('hds_url'):
  64             formats.extend(self._extract_f4m_formats(
  65                 mg['hds_server'] + mg['hds_url'], item_id,
  66                 f4m_id='hds', fatal=False))
  67
  68         self._sort_formats(formats)
  69
  70         return {
  71             'id': item_id,
  72             'title': title,
  73             'description': description,
  74             'thumbnail': thumbnail,
  75             'timestamp': timestamp,
  76             'duration': duration,
  77             'formats': formats,
  78         }
  79
  80
  81 class RteIE(RteBaseIE):
  82     IE_NAME = 'rte'
  83     IE_DESC = 'Raidió Teilifís Éireann TV'
  84     _VALID_URL = r'https?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/[^/]+/(?P<id>[0-9]+)'
  85     _TEST = {
  86         'url': 'http://www.rte.ie/player/ie/show/iwitness-862/10478715/',
  87         'md5': '4a76eb3396d98f697e6e8110563d2604',
  88         'info_dict': {
  89             'id': '10478715',
  90             'ext': 'mp4',
  91             'title': 'iWitness',
  92             'thumbnail': r're:^https?://.*\.jpg$',
  93             'description': 'The spirit of Ireland, one voice and one minute at a time.',
  94             'duration': 60.046,
  95             'upload_date': '20151012',
  96             'timestamp': 1444694160,
  97         },
  98     }
  99
 100
 101 class RteRadioIE(RteBaseIE):
 102     IE_NAME = 'rte:radio'
 103     IE_DESC = 'Raidió Teilifís Éireann radio'
 104     # Radioplayer URLs have two distinct specifier formats,
 105     # the old format #!rii=<channel_id>:<id>:<playable_item_id>:<date>:
 106     # the new format #!rii=b<channel_id>_<id>_<playable_item_id>_<date>_
 107     # where the IDs are int/empty, the date is DD-MM-YYYY, and the specifier may be truncated.
 108     # An <id> uniquely defines an individual recording, and is the only part we require.
 109     _VALID_URL = r'https?://(?:www\.)?rte\.ie/radio/utils/radioplayer/rteradioweb\.html#!rii=(?:b?[0-9]*)(?:%3A|:|%5F|_)(?P<id>[0-9]+)'
 110
 111     _TESTS = [{
 112         # Old-style player URL; HLS and RTMPE formats
 113         'url': 'http://www.rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=16:10507902:2414:27-12-2015:',
 114         'md5': 'c79ccb2c195998440065456b69760411',
 115         'info_dict': {
 116             'id': '10507902',
 117             'ext': 'mp4',
 118             'title': 'Gloria',
 119             'thumbnail': r're:^https?://.*\.jpg$',
 120             'description': 'md5:9ce124a7fb41559ec68f06387cabddf0',
 121             'timestamp': 1451203200,
 122             'upload_date': '20151227',
 123             'duration': 7230.0,
 124         },
 125     }, {
 126         # New-style player URL; RTMPE formats only
 127         'url': 'http://rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=b16_3250678_8861_06-04-2012_',
 128         'info_dict': {
 129             'id': '3250678',
 130             'ext': 'flv',
 131             'title': 'The Lyric Concert with Paul Herriott',
 132             'thumbnail': r're:^https?://.*\.jpg$',
 133             'description': '',
 134             'timestamp': 1333742400,
 135             'upload_date': '20120406',
 136             'duration': 7199.016,
 137         },
 138         'params': {
 139             # rtmp download
 140             'skip_download': True,
 141         },
 142     }]