youtube_dl/extractor/orf.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6 import calendar
   7 import datetime
   8
   9 from .common import InfoExtractor
  10 from ..utils import (
  11     HEADRequest,
  12     unified_strdate,
  13     ExtractorError,
  14 )
  15
  16
  17 class ORFTVthekIE(InfoExtractor):
  18     IE_NAME = 'orf:tvthek'
  19     IE_DESC = 'ORF TVthek'
  20     _VALID_URL = r'https?://tvthek\.orf\.at/(?:programs/.+?/episodes|topics/.+?|program/[^/]+)/(?P<id>\d+)'
  21
  22     _TEST = {
  23         'url': 'http://tvthek.orf.at/program/matinee-Was-Sie-schon-immer-ueber-Klassik-wissen-wollten/7317210/Was-Sie-schon-immer-ueber-Klassik-wissen-wollten/7319746/Was-Sie-schon-immer-ueber-Klassik-wissen-wollten/7319747',
  24         'file': '7319747.mp4',
  25         'md5': 'bd803c5d8c32d3c64a0ea4b4eeddf375',
  26         'info_dict': {
  27             'title': 'Was Sie schon immer über Klassik wissen wollten',
  28             'description': 'md5:0ddf0d5f0060bd53f744edaa5c2e04a4',
  29             'duration': 3508,
  30             'upload_date': '20140105',
  31         },
  32         'skip': 'Blocked outside of Austria',
  33     }
  34
  35     def _real_extract(self, url):
  36         mobj = re.match(self._VALID_URL, url)
  37         playlist_id = mobj.group('id')
  38         webpage = self._download_webpage(url, playlist_id)
  39
  40         data_json = self._search_regex(
  41             r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
  42         all_data = json.loads(data_json)
  43
  44         def get_segments(all_data):
  45             for data in all_data:
  46                 if data['name'] == 'Tracker::EPISODE_DETAIL_PAGE_OVER_PROGRAM':
  47                     return data['values']['segments']
  48
  49         sdata = get_segments(all_data)
  50         if not sdata:
  51             raise ExtractorError('Unable to extract segments')
  52
  53         def quality_to_int(s):
  54             m = re.search('([0-9]+)', s)
  55             if m is None:
  56                 return -1
  57             return int(m.group(1))
  58
  59         entries = []
  60         for sd in sdata:
  61             video_id = sd['id']
  62             formats = [{
  63                 'preference': -10 if fd['delivery'] == 'hls' else None,
  64                 'format_id': '%s-%s-%s' % (
  65                     fd['delivery'], fd['quality'], fd['quality_string']),
  66                 'url': fd['src'],
  67                 'protocol': fd['protocol'],
  68                 'quality': quality_to_int(fd['quality']),
  69             } for fd in sd['playlist_item_array']['sources']]
  70
  71             # Check for geoblocking.
  72             # There is a property is_geoprotection, but that's always false
  73             geo_str = sd.get('geoprotection_string')
  74             if geo_str:
  75                 try:
  76                     http_url = next(
  77                         f['url']
  78                         for f in formats
  79                         if re.match(r'^https?://.*\.mp4$', f['url']))
  80                 except StopIteration:
  81                     pass
  82                 else:
  83                     req = HEADRequest(http_url)
  84                     self._request_webpage(
  85                         req, video_id,
  86                         note='Testing for geoblocking',
  87                         errnote=((
  88                             'This video seems to be blocked outside of %s. '
  89                             'You may want to try the streaming-* formats.')
  90                             % geo_str),
  91                         fatal=False)
  92
  93             self._sort_formats(formats)
  94
  95             upload_date = unified_strdate(sd['created_date'])
  96             entries.append({
  97                 '_type': 'video',
  98                 'id': video_id,
  99                 'title': sd['header'],
 100                 'formats': formats,
 101                 'description': sd.get('description'),
 102                 'duration': int(sd['duration_in_seconds']),
 103                 'upload_date': upload_date,
 104                 'thumbnail': sd.get('image_full_url'),
 105             })
 106
 107         return {
 108             '_type': 'playlist',
 109             'entries': entries,
 110             'id': playlist_id,
 111         }
 112
 113
 114 # Audios on ORF radio are only available for 7 days, so we can't add tests.
 115
 116
 117 class ORFOE1IE(InfoExtractor):
 118     IE_NAME = 'orf:oe1'
 119     IE_DESC = 'Radio Österreich 1'
 120     _VALID_URL = r'http://oe1\.orf\.at/programm/(?P<id>[0-9]+)'
 121
 122     def _real_extract(self, url):
 123         mobj = re.match(self._VALID_URL, url)
 124         show_id = mobj.group('id')
 125
 126         data = self._download_json(
 127             'http://oe1.orf.at/programm/%s/konsole' % show_id,
 128             show_id
 129         )
 130
 131         timestamp = datetime.datetime.strptime('%s %s' % (
 132             data['item']['day_label'],
 133             data['item']['time']
 134         ), '%d.%m.%Y %H:%M')
 135         unix_timestamp = calendar.timegm(timestamp.utctimetuple())
 136
 137         return {
 138             'id': show_id,
 139             'title': data['item']['title'],
 140             'url': data['item']['url_stream'],
 141             'ext': 'mp3',
 142             'description': data['item'].get('info'),
 143             'timestamp': unix_timestamp
 144         }
 145
 146
 147 class ORFFM4IE(InfoExtractor):
 148     IE_DESC = 'orf:fm4'
 149     IE_DESC = 'radio FM4'
 150     _VALID_URL = r'http://fm4\.orf\.at/7tage/?#(?P<date>[0-9]+)/(?P<show>\w+)'
 151
 152     def _real_extract(self, url):
 153         mobj = re.match(self._VALID_URL, url)
 154         show_date = mobj.group('date')
 155         show_id = mobj.group('show')
 156
 157         data = self._download_json(
 158             'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
 159             show_id
 160         )
 161
 162         def extract_entry_dict(info, title, subtitle):
 163             return {
 164                 'id': info['loopStreamId'].replace('.mp3', ''),
 165                 'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
 166                 'title': title,
 167                 'description': subtitle,
 168                 'duration': (info['end'] - info['start']) / 1000,
 169                 'timestamp': info['start'] / 1000,
 170                 'ext': 'mp3'
 171             }
 172
 173         entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
 174
 175         return {
 176             '_type': 'playlist',
 177             'id': show_id,
 178             'title': data['title'],
 179             'description': data['subtitle'],
 180             'entries': entries
 181         }