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     strip_jsonp,
  15     int_or_none,
  16     float_or_none,
  17     determine_ext,
  18     remove_end,
  19 )
  20
  21
  22 class ORFTVthekIE(InfoExtractor):
  23     IE_NAME = 'orf:tvthek'
  24     IE_DESC = 'ORF TVthek'
  25     _VALID_URL = r'https?://tvthek\.orf\.at/(?:programs/.+?/episodes|topics?/.+?|program/[^/]+)/(?P<id>\d+)'
  26
  27     _TESTS = [{
  28         'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
  29         'playlist': [{
  30             'md5': '2942210346ed779588f428a92db88712',
  31             'info_dict': {
  32                 'id': '8896777',
  33                 'ext': 'mp4',
  34                 'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
  35                 'description': 'md5:c1272f0245537812d4e36419c207b67d',
  36                 'duration': 2668,
  37                 'upload_date': '20141208',
  38             },
  39         }],
  40         'skip': 'Blocked outside of Austria / Germany',
  41     }, {
  42         'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
  43         'info_dict': {
  44             'id': '7982259',
  45             'ext': 'mp4',
  46             'title': 'Best of Ingrid Thurnher',
  47             'upload_date': '20140527',
  48             '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".',
  49         },
  50         'params': {
  51             'skip_download': True,  # rtsp downloads
  52         },
  53         '_skip': 'Blocked outside of Austria / Germany',
  54     }]
  55
  56     def _real_extract(self, url):
  57         playlist_id = self._match_id(url)
  58         webpage = self._download_webpage(url, playlist_id)
  59
  60         data_json = self._search_regex(
  61             r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
  62         all_data = json.loads(data_json)
  63
  64         def get_segments(all_data):
  65             for data in all_data:
  66                 if data['name'] in (
  67                         'Tracker::EPISODE_DETAIL_PAGE_OVER_PROGRAM',
  68                         'Tracker::EPISODE_DETAIL_PAGE_OVER_TOPIC'):
  69                     return data['values']['segments']
  70
  71         sdata = get_segments(all_data)
  72         if not sdata:
  73             raise ExtractorError('Unable to extract segments')
  74
  75         def quality_to_int(s):
  76             m = re.search('([0-9]+)', s)
  77             if m is None:
  78                 return -1
  79             return int(m.group(1))
  80
  81         entries = []
  82         for sd in sdata:
  83             video_id = sd['id']
  84             formats = [{
  85                 'preference': -10 if fd['delivery'] == 'hls' else None,
  86                 'format_id': '%s-%s-%s' % (
  87                     fd['delivery'], fd['quality'], fd['quality_string']),
  88                 'url': fd['src'],
  89                 'protocol': fd['protocol'],
  90                 'quality': quality_to_int(fd['quality']),
  91             } for fd in sd['playlist_item_array']['sources']]
  92
  93             # Check for geoblocking.
  94             # There is a property is_geoprotection, but that's always false
  95             geo_str = sd.get('geoprotection_string')
  96             if geo_str:
  97                 try:
  98                     http_url = next(
  99                         f['url']
 100                         for f in formats
 101                         if re.match(r'^https?://.*\.mp4$', f['url']))
 102                 except StopIteration:
 103                     pass
 104                 else:
 105                     req = HEADRequest(http_url)
 106                     self._request_webpage(
 107                         req, video_id,
 108                         note='Testing for geoblocking',
 109                         errnote=((
 110                             'This video seems to be blocked outside of %s. '
 111                             'You may want to try the streaming-* formats.')
 112                             % geo_str),
 113                         fatal=False)
 114
 115             self._check_formats(formats, video_id)
 116             self._sort_formats(formats)
 117
 118             upload_date = unified_strdate(sd['created_date'])
 119             entries.append({
 120                 '_type': 'video',
 121                 'id': video_id,
 122                 'title': sd['header'],
 123                 'formats': formats,
 124                 'description': sd.get('description'),
 125                 'duration': int(sd['duration_in_seconds']),
 126                 'upload_date': upload_date,
 127                 'thumbnail': sd.get('image_full_url'),
 128             })
 129
 130         return {
 131             '_type': 'playlist',
 132             'entries': entries,
 133             'id': playlist_id,
 134         }
 135
 136
 137 class ORFOE1IE(InfoExtractor):
 138     IE_NAME = 'orf:oe1'
 139     IE_DESC = 'Radio Österreich 1'
 140     _VALID_URL = r'https?://oe1\.orf\.at/(?:programm/|konsole\?.*?\btrack_id=)(?P<id>[0-9]+)'
 141
 142     # Audios on ORF radio are only available for 7 days, so we can't add tests.
 143     _TESTS = [{
 144         'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
 145         'only_matching': True,
 146     }, {
 147         'url': 'http://oe1.orf.at/konsole?show=ondemand&track_id=443608&load_day=/programm/konsole/tag/20160726',
 148         'only_matching': True,
 149     }]
 150
 151     def _real_extract(self, url):
 152         show_id = self._match_id(url)
 153         data = self._download_json(
 154             'http://oe1.orf.at/programm/%s/konsole' % show_id,
 155             show_id
 156         )
 157
 158         timestamp = datetime.datetime.strptime('%s %s' % (
 159             data['item']['day_label'],
 160             data['item']['time']
 161         ), '%d.%m.%Y %H:%M')
 162         unix_timestamp = calendar.timegm(timestamp.utctimetuple())
 163
 164         return {
 165             'id': show_id,
 166             'title': data['item']['title'],
 167             'url': data['item']['url_stream'],
 168             'ext': 'mp3',
 169             'description': data['item'].get('info'),
 170             'timestamp': unix_timestamp
 171         }
 172
 173
 174 class ORFFM4IE(InfoExtractor):
 175     IE_NAME = 'orf:fm4'
 176     IE_DESC = 'radio FM4'
 177     _VALID_URL = r'https?://fm4\.orf\.at/(?:7tage/?#|player/)(?P<date>[0-9]+)/(?P<show>\w+)'
 178
 179     _TEST = {
 180         'url': 'http://fm4.orf.at/player/20160110/IS/',
 181         'md5': '01e736e8f1cef7e13246e880a59ad298',
 182         'info_dict': {
 183             'id': '2016-01-10_2100_tl_54_7DaysSun13_11244',
 184             'ext': 'mp3',
 185             'title': 'Im Sumpf',
 186             'description': 'md5:384c543f866c4e422a55f66a62d669cd',
 187             'duration': 7173,
 188             'timestamp': 1452456073,
 189             'upload_date': '20160110',
 190         },
 191         'skip': 'Live streams on FM4 got deleted soon',
 192     }
 193
 194     def _real_extract(self, url):
 195         mobj = re.match(self._VALID_URL, url)
 196         show_date = mobj.group('date')
 197         show_id = mobj.group('show')
 198
 199         data = self._download_json(
 200             'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
 201             show_id
 202         )
 203
 204         def extract_entry_dict(info, title, subtitle):
 205             return {
 206                 'id': info['loopStreamId'].replace('.mp3', ''),
 207                 'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
 208                 'title': title,
 209                 'description': subtitle,
 210                 'duration': (info['end'] - info['start']) / 1000,
 211                 'timestamp': info['start'] / 1000,
 212                 'ext': 'mp3'
 213             }
 214
 215         entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
 216
 217         return {
 218             '_type': 'playlist',
 219             'id': show_id,
 220             'title': data['title'],
 221             'description': data['subtitle'],
 222             'entries': entries
 223         }
 224
 225
 226 class ORFIPTVIE(InfoExtractor):
 227     IE_NAME = 'orf:iptv'
 228     IE_DESC = 'iptv.ORF.at'
 229     _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
 230
 231     _TEST = {
 232         'url': 'http://iptv.orf.at/stories/2275236/',
 233         'md5': 'c8b22af4718a4b4af58342529453e3e5',
 234         'info_dict': {
 235             'id': '350612',
 236             'ext': 'flv',
 237             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
 238             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
 239             'duration': 68.197,
 240             'thumbnail': 're:^https?://.*\.jpg$',
 241             'upload_date': '20150425',
 242         },
 243     }
 244
 245     def _real_extract(self, url):
 246         story_id = self._match_id(url)
 247
 248         webpage = self._download_webpage(
 249             'http://iptv.orf.at/stories/%s' % story_id, story_id)
 250
 251         video_id = self._search_regex(
 252             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
 253
 254         data = self._download_json(
 255             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 256             video_id)[0]
 257
 258         duration = float_or_none(data['duration'], 1000)
 259
 260         video = data['sources']['default']
 261         load_balancer_url = video['loadBalancerUrl']
 262         abr = int_or_none(video.get('audioBitrate'))
 263         vbr = int_or_none(video.get('bitrate'))
 264         fps = int_or_none(video.get('videoFps'))
 265         width = int_or_none(video.get('videoWidth'))
 266         height = int_or_none(video.get('videoHeight'))
 267         thumbnail = video.get('preview')
 268
 269         rendition = self._download_json(
 270             load_balancer_url, video_id, transform_source=strip_jsonp)
 271
 272         f = {
 273             'abr': abr,
 274             'vbr': vbr,
 275             'fps': fps,
 276             'width': width,
 277             'height': height,
 278         }
 279
 280         formats = []
 281         for format_id, format_url in rendition['redirect'].items():
 282             if format_id == 'rtmp':
 283                 ff = f.copy()
 284                 ff.update({
 285                     'url': format_url,
 286                     'format_id': format_id,
 287                 })
 288                 formats.append(ff)
 289             elif determine_ext(format_url) == 'f4m':
 290                 formats.extend(self._extract_f4m_formats(
 291                     format_url, video_id, f4m_id=format_id))
 292             elif determine_ext(format_url) == 'm3u8':
 293                 formats.extend(self._extract_m3u8_formats(
 294                     format_url, video_id, 'mp4', m3u8_id=format_id))
 295             else:
 296                 continue
 297         self._sort_formats(formats)
 298
 299         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
 300         description = self._og_search_description(webpage)
 301         upload_date = unified_strdate(self._html_search_meta(
 302             'dc.date', webpage, 'upload date'))
 303
 304         return {
 305             'id': video_id,
 306             'title': title,
 307             'description': description,
 308             'duration': duration,
 309             'thumbnail': thumbnail,
 310             'upload_date': upload_date,
 311             'formats': formats,
 312         }