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         'playlist': [{
  44             'md5': '68f543909aea49d621dfc7703a11cfaf',
  45             'info_dict': {
  46                 'id': '7982259',
  47                 'ext': 'mp4',
  48                 'title': 'Best of Ingrid Thurnher',
  49                 'upload_date': '20140527',
  50                 '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".',
  51             }
  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._sort_formats(formats)
 116
 117             upload_date = unified_strdate(sd['created_date'])
 118             entries.append({
 119                 '_type': 'video',
 120                 'id': video_id,
 121                 'title': sd['header'],
 122                 'formats': formats,
 123                 'description': sd.get('description'),
 124                 'duration': int(sd['duration_in_seconds']),
 125                 'upload_date': upload_date,
 126                 'thumbnail': sd.get('image_full_url'),
 127             })
 128
 129         return {
 130             '_type': 'playlist',
 131             'entries': entries,
 132             'id': playlist_id,
 133         }
 134
 135
 136 class ORFOE1IE(InfoExtractor):
 137     IE_NAME = 'orf:oe1'
 138     IE_DESC = 'Radio Österreich 1'
 139     _VALID_URL = r'http://oe1\.orf\.at/(?:programm/|konsole.*?#\?track_id=)(?P<id>[0-9]+)'
 140
 141     # Audios on ORF radio are only available for 7 days, so we can't add tests.
 142     _TEST = {
 143         'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
 144         'only_matching': True,
 145     }
 146
 147     def _real_extract(self, url):
 148         show_id = self._match_id(url)
 149         data = self._download_json(
 150             'http://oe1.orf.at/programm/%s/konsole' % show_id,
 151             show_id
 152         )
 153
 154         timestamp = datetime.datetime.strptime('%s %s' % (
 155             data['item']['day_label'],
 156             data['item']['time']
 157         ), '%d.%m.%Y %H:%M')
 158         unix_timestamp = calendar.timegm(timestamp.utctimetuple())
 159
 160         return {
 161             'id': show_id,
 162             'title': data['item']['title'],
 163             'url': data['item']['url_stream'],
 164             'ext': 'mp3',
 165             'description': data['item'].get('info'),
 166             'timestamp': unix_timestamp
 167         }
 168
 169
 170 class ORFFM4IE(InfoExtractor):
 171     IE_NAME = 'orf:fm4'
 172     IE_DESC = 'radio FM4'
 173     _VALID_URL = r'http://fm4\.orf\.at/(?:7tage/?#|player/)(?P<date>[0-9]+)/(?P<show>\w+)'
 174
 175     _TEST = {
 176         'url': 'http://fm4.orf.at/player/20160110/IS/',
 177         'md5': '01e736e8f1cef7e13246e880a59ad298',
 178         'info_dict': {
 179             'id': '2016-01-10_2100_tl_54_7DaysSun13_11244',
 180             'ext': 'mp3',
 181             'title': 'Im Sumpf',
 182             'description': 'md5:384c543f866c4e422a55f66a62d669cd',
 183             'duration': 7173,
 184             'timestamp': 1452456073,
 185             'upload_date': '20160110',
 186         },
 187     }
 188
 189     def _real_extract(self, url):
 190         mobj = re.match(self._VALID_URL, url)
 191         show_date = mobj.group('date')
 192         show_id = mobj.group('show')
 193
 194         data = self._download_json(
 195             'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
 196             show_id
 197         )
 198
 199         def extract_entry_dict(info, title, subtitle):
 200             return {
 201                 'id': info['loopStreamId'].replace('.mp3', ''),
 202                 'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
 203                 'title': title,
 204                 'description': subtitle,
 205                 'duration': (info['end'] - info['start']) / 1000,
 206                 'timestamp': info['start'] / 1000,
 207                 'ext': 'mp3'
 208             }
 209
 210         entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
 211
 212         return {
 213             '_type': 'playlist',
 214             'id': show_id,
 215             'title': data['title'],
 216             'description': data['subtitle'],
 217             'entries': entries
 218         }
 219
 220
 221 class ORFIPTVIE(InfoExtractor):
 222     IE_NAME = 'orf:iptv'
 223     IE_DESC = 'iptv.ORF.at'
 224     _VALID_URL = r'http://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
 225
 226     _TEST = {
 227         'url': 'http://iptv.orf.at/stories/2275236/',
 228         'md5': 'c8b22af4718a4b4af58342529453e3e5',
 229         'info_dict': {
 230             'id': '350612',
 231             'ext': 'flv',
 232             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
 233             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
 234             'duration': 68.197,
 235             'thumbnail': 're:^https?://.*\.jpg$',
 236             'upload_date': '20150425',
 237         },
 238     }
 239
 240     def _real_extract(self, url):
 241         story_id = self._match_id(url)
 242
 243         webpage = self._download_webpage(
 244             'http://iptv.orf.at/stories/%s' % story_id, story_id)
 245
 246         video_id = self._search_regex(
 247             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
 248
 249         data = self._download_json(
 250             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 251             video_id)[0]
 252
 253         duration = float_or_none(data['duration'], 1000)
 254
 255         video = data['sources']['default']
 256         load_balancer_url = video['loadBalancerUrl']
 257         abr = int_or_none(video.get('audioBitrate'))
 258         vbr = int_or_none(video.get('bitrate'))
 259         fps = int_or_none(video.get('videoFps'))
 260         width = int_or_none(video.get('videoWidth'))
 261         height = int_or_none(video.get('videoHeight'))
 262         thumbnail = video.get('preview')
 263
 264         rendition = self._download_json(
 265             load_balancer_url, video_id, transform_source=strip_jsonp)
 266
 267         f = {
 268             'abr': abr,
 269             'vbr': vbr,
 270             'fps': fps,
 271             'width': width,
 272             'height': height,
 273         }
 274
 275         formats = []
 276         for format_id, format_url in rendition['redirect'].items():
 277             if format_id == 'rtmp':
 278                 ff = f.copy()
 279                 ff.update({
 280                     'url': format_url,
 281                     'format_id': format_id,
 282                 })
 283                 formats.append(ff)
 284             elif determine_ext(format_url) == 'f4m':
 285                 formats.extend(self._extract_f4m_formats(
 286                     format_url, video_id, f4m_id=format_id))
 287             elif determine_ext(format_url) == 'm3u8':
 288                 formats.extend(self._extract_m3u8_formats(
 289                     format_url, video_id, 'mp4', m3u8_id=format_id))
 290             else:
 291                 continue
 292         self._sort_formats(formats)
 293
 294         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
 295         description = self._og_search_description(webpage)
 296         upload_date = unified_strdate(self._html_search_meta(
 297             'dc.date', webpage, 'upload date'))
 298
 299         return {
 300             'id': video_id,
 301             'title': title,
 302             'description': description,
 303             'duration': duration,
 304             'thumbnail': thumbnail,
 305             'upload_date': upload_date,
 306             'formats': formats,
 307         }