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     _TESTS = [{
  23         'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
  24         'playlist': [{
  25             'md5': '2942210346ed779588f428a92db88712',
  26             'info_dict': {
  27                 'id': '8896777',
  28                 'ext': 'mp4',
  29                 'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
  30                 'description': 'md5:c1272f0245537812d4e36419c207b67d',
  31                 'duration': 2668,
  32                 'upload_date': '20141208',
  33             },
  34         }],
  35         'skip': 'Blocked outside of Austria / Germany',
  36     }, {
  37         'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
  38         'playlist': [{
  39             'md5': '68f543909aea49d621dfc7703a11cfaf',
  40             'info_dict': {
  41                 'id': '7982259',
  42                 'ext': 'mp4',
  43                 'title': 'Best of Ingrid Thurnher',
  44                 'upload_date': '20140527',
  45                 'description': 'Viele Jahre war Ingrid Thurnher das "Gesicht" der ZIB 2. Vor ihrem Wechsel zur ZIB 2 im jahr 1995 moderierte sie unter anderem "Land und Leute", "Österreich-Bild" und "Niederösterreich heute".',
  46             }
  47         }],
  48         '_skip': 'Blocked outside of Austria / Germany',
  49     }]
  50
  51     def _real_extract(self, url):
  52         playlist_id = self._match_id(url)
  53         webpage = self._download_webpage(url, playlist_id)
  54
  55         data_json = self._search_regex(
  56             r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
  57         all_data = json.loads(data_json)
  58
  59         def get_segments(all_data):
  60             for data in all_data:
  61                 if data['name'] in (
  62                         'Tracker::EPISODE_DETAIL_PAGE_OVER_PROGRAM',
  63                         'Tracker::EPISODE_DETAIL_PAGE_OVER_TOPIC'):
  64                     return data['values']['segments']
  65
  66         sdata = get_segments(all_data)
  67         if not sdata:
  68             raise ExtractorError('Unable to extract segments')
  69
  70         def quality_to_int(s):
  71             m = re.search('([0-9]+)', s)
  72             if m is None:
  73                 return -1
  74             return int(m.group(1))
  75
  76         entries = []
  77         for sd in sdata:
  78             video_id = sd['id']
  79             formats = [{
  80                 'preference': -10 if fd['delivery'] == 'hls' else None,
  81                 'format_id': '%s-%s-%s' % (
  82                     fd['delivery'], fd['quality'], fd['quality_string']),
  83                 'url': fd['src'],
  84                 'protocol': fd['protocol'],
  85                 'quality': quality_to_int(fd['quality']),
  86             } for fd in sd['playlist_item_array']['sources']]
  87
  88             # Check for geoblocking.
  89             # There is a property is_geoprotection, but that's always false
  90             geo_str = sd.get('geoprotection_string')
  91             if geo_str:
  92                 try:
  93                     http_url = next(
  94                         f['url']
  95                         for f in formats
  96                         if re.match(r'^https?://.*\.mp4$', f['url']))
  97                 except StopIteration:
  98                     pass
  99                 else:
 100                     req = HEADRequest(http_url)
 101                     self._request_webpage(
 102                         req, video_id,
 103                         note='Testing for geoblocking',
 104                         errnote=((
 105                             'This video seems to be blocked outside of %s. '
 106                             'You may want to try the streaming-* formats.')
 107                             % geo_str),
 108                         fatal=False)
 109
 110             self._sort_formats(formats)
 111
 112             upload_date = unified_strdate(sd['created_date'])
 113             entries.append({
 114                 '_type': 'video',
 115                 'id': video_id,
 116                 'title': sd['header'],
 117                 'formats': formats,
 118                 'description': sd.get('description'),
 119                 'duration': int(sd['duration_in_seconds']),
 120                 'upload_date': upload_date,
 121                 'thumbnail': sd.get('image_full_url'),
 122             })
 123
 124         return {
 125             '_type': 'playlist',
 126             'entries': entries,
 127             'id': playlist_id,
 128         }
 129
 130
 131 # Audios on ORF radio are only available for 7 days, so we can't add tests.
 132
 133
 134 class ORFOE1IE(InfoExtractor):
 135     IE_NAME = 'orf:oe1'
 136     IE_DESC = 'Radio Österreich 1'
 137     _VALID_URL = r'http://oe1\.orf\.at/programm/(?P<id>[0-9]+)'
 138
 139     def _real_extract(self, url):
 140         show_id = self._match_id(url)
 141         data = self._download_json(
 142             'http://oe1.orf.at/programm/%s/konsole' % show_id,
 143             show_id
 144         )
 145
 146         timestamp = datetime.datetime.strptime('%s %s' % (
 147             data['item']['day_label'],
 148             data['item']['time']
 149         ), '%d.%m.%Y %H:%M')
 150         unix_timestamp = calendar.timegm(timestamp.utctimetuple())
 151
 152         return {
 153             'id': show_id,
 154             'title': data['item']['title'],
 155             'url': data['item']['url_stream'],
 156             'ext': 'mp3',
 157             'description': data['item'].get('info'),
 158             'timestamp': unix_timestamp
 159         }
 160
 161
 162 class ORFFM4IE(InfoExtractor):
 163     IE_NAME = 'orf:fm4'
 164     IE_DESC = 'radio FM4'
 165     _VALID_URL = r'http://fm4\.orf\.at/7tage/?#(?P<date>[0-9]+)/(?P<show>\w+)'
 166
 167     def _real_extract(self, url):
 168         mobj = re.match(self._VALID_URL, url)
 169         show_date = mobj.group('date')
 170         show_id = mobj.group('show')
 171
 172         data = self._download_json(
 173             'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
 174             show_id
 175         )
 176
 177         def extract_entry_dict(info, title, subtitle):
 178             return {
 179                 'id': info['loopStreamId'].replace('.mp3', ''),
 180                 'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
 181                 'title': title,
 182                 'description': subtitle,
 183                 'duration': (info['end'] - info['start']) / 1000,
 184                 'timestamp': info['start'] / 1000,
 185                 'ext': 'mp3'
 186             }
 187
 188         entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
 189
 190         return {
 191             '_type': 'playlist',
 192             'id': show_id,
 193             'title': data['title'],
 194             'description': data['subtitle'],
 195             'entries': entries
 196         }