_ 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     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         'md5': '68f543909aea49d621dfc7703a11cfaf',
  44         'info_dict': {
  45             'id': '7982259',
  46             'ext': 'mp4',
  47             'title': 'Best of Ingrid Thurnher',
  48             'upload_date': '20140527',
  49             '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".',
  50         },
  51         'params': {
  52             'skip_download': True,  # rtsp downloads
  53         },
  54         '_skip': 'Blocked outside of Austria / Germany',
  55     }]
  56
  57     def _real_extract(self, url):
  58         playlist_id = self._match_id(url)
  59         webpage = self._download_webpage(url, playlist_id)
  60
  61         data_json = self._search_regex(
  62             r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
  63         all_data = json.loads(data_json)
  64
  65         def get_segments(all_data):
  66             for data in all_data:
  67                 if data['name'] in (
  68                         'Tracker::EPISODE_DETAIL_PAGE_OVER_PROGRAM',
  69                         'Tracker::EPISODE_DETAIL_PAGE_OVER_TOPIC'):
  70                     return data['values']['segments']
  71
  72         sdata = get_segments(all_data)
  73         if not sdata:
  74             raise ExtractorError('Unable to extract segments')
  75
  76         def quality_to_int(s):
  77             m = re.search('([0-9]+)', s)
  78             if m is None:
  79                 return -1
  80             return int(m.group(1))
  81
  82         entries = []
  83         for sd in sdata:
  84             video_id = sd['id']
  85             formats = [{
  86                 'preference': -10 if fd['delivery'] == 'hls' else None,
  87                 'format_id': '%s-%s-%s' % (
  88                     fd['delivery'], fd['quality'], fd['quality_string']),
  89                 'url': fd['src'],
  90                 'protocol': fd['protocol'],
  91                 'quality': quality_to_int(fd['quality']),
  92             } for fd in sd['playlist_item_array']['sources']]
  93
  94             # Check for geoblocking.
  95             # There is a property is_geoprotection, but that's always false
  96             geo_str = sd.get('geoprotection_string')
  97             if geo_str:
  98                 try:
  99                     http_url = next(
 100                         f['url']
 101                         for f in formats
 102                         if re.match(r'^https?://.*\.mp4$', f['url']))
 103                 except StopIteration:
 104                     pass
 105                 else:
 106                     req = HEADRequest(http_url)
 107                     self._request_webpage(
 108                         req, video_id,
 109                         note='Testing for geoblocking',
 110                         errnote=((
 111                             'This video seems to be blocked outside of %s. '
 112                             'You may want to try the streaming-* formats.')
 113                             % geo_str),
 114                         fatal=False)
 115
 116             self._check_formats(formats, video_id)
 117             self._sort_formats(formats)
 118
 119             upload_date = unified_strdate(sd['created_date'])
 120             entries.append({
 121                 '_type': 'video',
 122                 'id': video_id,
 123                 'title': sd['header'],
 124                 'formats': formats,
 125                 'description': sd.get('description'),
 126                 'duration': int(sd['duration_in_seconds']),
 127                 'upload_date': upload_date,
 128                 'thumbnail': sd.get('image_full_url'),
 129             })
 130
 131         return {
 132             '_type': 'playlist',
 133             'entries': entries,
 134             'id': playlist_id,
 135         }
 136
 137
 138 class ORFOE1IE(InfoExtractor):
 139     IE_NAME = 'orf:oe1'
 140     IE_DESC = 'Radio Österreich 1'
 141     _VALID_URL = r'https?://oe1\.orf\.at/(?:programm/|konsole.*?#\?track_id=)(?P<id>[0-9]+)'
 142
 143     # Audios on ORF radio are only available for 7 days, so we can't add tests.
 144     _TEST = {
 145         'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
 146         'only_matching': True,
 147     }
 148
 149     def _real_extract(self, url):
 150         show_id = self._match_id(url)
 151         data = self._download_json(
 152             'http://oe1.orf.at/programm/%s/konsole' % show_id,
 153             show_id
 154         )
 155
 156         timestamp = datetime.datetime.strptime('%s %s' % (
 157             data['item']['day_label'],
 158             data['item']['time']
 159         ), '%d.%m.%Y %H:%M')
 160         unix_timestamp = calendar.timegm(timestamp.utctimetuple())
 161
 162         return {
 163             'id': show_id,
 164             'title': data['item']['title'],
 165             'url': data['item']['url_stream'],
 166             'ext': 'mp3',
 167             'description': data['item'].get('info'),
 168             'timestamp': unix_timestamp
 169         }
 170
 171
 172 class ORFFM4IE(InfoExtractor):
 173     IE_NAME = 'orf:fm4'
 174     IE_DESC = 'radio FM4'
 175     _VALID_URL = r'https?://fm4\.orf\.at/(?:7tage/?#|player/)(?P<date>[0-9]+)/(?P<show>\w+)'
 176
 177     _TEST = {
 178         'url': 'http://fm4.orf.at/player/20160110/IS/',
 179         'md5': '01e736e8f1cef7e13246e880a59ad298',
 180         'info_dict': {
 181             'id': '2016-01-10_2100_tl_54_7DaysSun13_11244',
 182             'ext': 'mp3',
 183             'title': 'Im Sumpf',
 184             'description': 'md5:384c543f866c4e422a55f66a62d669cd',
 185             'duration': 7173,
 186             'timestamp': 1452456073,
 187             'upload_date': '20160110',
 188         },
 189         'skip': 'Live streams on FM4 got deleted soon',
 190     }
 191
 192     def _real_extract(self, url):
 193         mobj = re.match(self._VALID_URL, url)
 194         show_date = mobj.group('date')
 195         show_id = mobj.group('show')
 196
 197         data = self._download_json(
 198             'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
 199             show_id
 200         )
 201
 202         def extract_entry_dict(info, title, subtitle):
 203             return {
 204                 'id': info['loopStreamId'].replace('.mp3', ''),
 205                 'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
 206                 'title': title,
 207                 'description': subtitle,
 208                 'duration': (info['end'] - info['start']) / 1000,
 209                 'timestamp': info['start'] / 1000,
 210                 'ext': 'mp3'
 211             }
 212
 213         entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
 214
 215         return {
 216             '_type': 'playlist',
 217             'id': show_id,
 218             'title': data['title'],
 219             'description': data['subtitle'],
 220             'entries': entries
 221         }
 222
 223
 224 class ORFIPTVIE(InfoExtractor):
 225     IE_NAME = 'orf:iptv'
 226     IE_DESC = 'iptv.ORF.at'
 227     _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
 228
 229     _TEST = {
 230         'url': 'http://iptv.orf.at/stories/2275236/',
 231         'md5': 'c8b22af4718a4b4af58342529453e3e5',
 232         'info_dict': {
 233             'id': '350612',
 234             'ext': 'flv',
 235             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
 236             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
 237             'duration': 68.197,
 238             'thumbnail': 're:^https?://.*\.jpg$',
 239             'upload_date': '20150425',
 240         },
 241     }
 242
 243     def _real_extract(self, url):
 244         story_id = self._match_id(url)
 245
 246         webpage = self._download_webpage(
 247             'http://iptv.orf.at/stories/%s' % story_id, story_id)
 248
 249         video_id = self._search_regex(
 250             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
 251
 252         data = self._download_json(
 253             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 254             video_id)[0]
 255
 256         duration = float_or_none(data['duration'], 1000)
 257
 258         video = data['sources']['default']
 259         load_balancer_url = video['loadBalancerUrl']
 260         abr = int_or_none(video.get('audioBitrate'))
 261         vbr = int_or_none(video.get('bitrate'))
 262         fps = int_or_none(video.get('videoFps'))
 263         width = int_or_none(video.get('videoWidth'))
 264         height = int_or_none(video.get('videoHeight'))
 265         thumbnail = video.get('preview')
 266
 267         rendition = self._download_json(
 268             load_balancer_url, video_id, transform_source=strip_jsonp)
 269
 270         f = {
 271             'abr': abr,
 272             'vbr': vbr,
 273             'fps': fps,
 274             'width': width,
 275             'height': height,
 276         }
 277
 278         formats = []
 279         for format_id, format_url in rendition['redirect'].items():
 280             if format_id == 'rtmp':
 281                 ff = f.copy()
 282                 ff.update({
 283                     'url': format_url,
 284                     'format_id': format_id,
 285                 })
 286                 formats.append(ff)
 287             elif determine_ext(format_url) == 'f4m':
 288                 formats.extend(self._extract_f4m_formats(
 289                     format_url, video_id, f4m_id=format_id))
 290             elif determine_ext(format_url) == 'm3u8':
 291                 formats.extend(self._extract_m3u8_formats(
 292                     format_url, video_id, 'mp4', m3u8_id=format_id))
 293             else:
 294                 continue
 295         self._sort_formats(formats)
 296
 297         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
 298         description = self._og_search_description(webpage)
 299         upload_date = unified_strdate(self._html_search_meta(
 300             'dc.date', webpage, 'upload date'))
 301
 302         return {
 303             'id': video_id,
 304             'title': title,
 305             'description': description,
 306             'duration': duration,
 307             'thumbnail': thumbnail,
 308             'upload_date': upload_date,
 309             'formats': formats,
 310         }