_ Git - youtube-dl/blob - 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/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',
  36     }
  37
  38     def _real_extract(self, url):
  39         playlist_id = self._match_id(url)
  40         webpage = self._download_webpage(url, playlist_id)
  41
  42         data_json = self._search_regex(
  43             r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
  44         all_data = json.loads(data_json)
  45
  46         def get_segments(all_data):
  47             for data in all_data:
  48                 if data['name'] == 'Tracker::EPISODE_DETAIL_PAGE_OVER_PROGRAM':
  49                     return data['values']['segments']
  50
  51         sdata = get_segments(all_data)
  52         if not sdata:
  53             raise ExtractorError('Unable to extract segments')
  54
  55         def quality_to_int(s):
  56             m = re.search('([0-9]+)', s)
  57             if m is None:
  58                 return -1
  59             return int(m.group(1))
  60
  61         entries = []
  62         for sd in sdata:
  63             video_id = sd['id']
  64             formats = [{
  65                 'preference': -10 if fd['delivery'] == 'hls' else None,
  66                 'format_id': '%s-%s-%s' % (
  67                     fd['delivery'], fd['quality'], fd['quality_string']),
  68                 'url': fd['src'],
  69                 'protocol': fd['protocol'],
  70                 'quality': quality_to_int(fd['quality']),
  71             } for fd in sd['playlist_item_array']['sources']]
  72
  73             # Check for geoblocking.
  74             # There is a property is_geoprotection, but that's always false
  75             geo_str = sd.get('geoprotection_string')
  76             if geo_str:
  77                 try:
  78                     http_url = next(
  79                         f['url']
  80                         for f in formats
  81                         if re.match(r'^https?://.*\.mp4$', f['url']))
  82                 except StopIteration:
  83                     pass
  84                 else:
  85                     req = HEADRequest(http_url)
  86                     self._request_webpage(
  87                         req, video_id,
  88                         note='Testing for geoblocking',
  89                         errnote=((
  90                             'This video seems to be blocked outside of %s. '
  91                             'You may want to try the streaming-* formats.')
  92                             % geo_str),
  93                         fatal=False)
  94
  95             self._sort_formats(formats)
  96
  97             upload_date = unified_strdate(sd['created_date'])
  98             entries.append({
  99                 '_type': 'video',
 100                 'id': video_id,
 101                 'title': sd['header'],
 102                 'formats': formats,
 103                 'description': sd.get('description'),
 104                 'duration': int(sd['duration_in_seconds']),
 105                 'upload_date': upload_date,
 106                 'thumbnail': sd.get('image_full_url'),
 107             })
 108
 109         return {
 110             '_type': 'playlist',
 111             'entries': entries,
 112             'id': playlist_id,
 113         }
 114
 115
 116 # Audios on ORF radio are only available for 7 days, so we can't add tests.
 117
 118
 119 class ORFOE1IE(InfoExtractor):
 120     IE_NAME = 'orf:oe1'
 121     IE_DESC = 'Radio Österreich 1'
 122     _VALID_URL = r'http://oe1\.orf\.at/programm/(?P<id>[0-9]+)'
 123
 124     def _real_extract(self, url):
 125         show_id = self._match_id(url)
 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         }