Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/orf.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import calendar
   6 import datetime
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_str
  10 from ..utils import (
  11     HEADRequest,
  12     unified_strdate,
  13     strip_jsonp,
  14     int_or_none,
  15     float_or_none,
  16     determine_ext,
  17     remove_end,
  18     unescapeHTML,
  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/(?:[^/]+/)+(?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         'url': 'http://tvthek.orf.at/topic/Fluechtlingskrise/10463081/Heimat-Fremde-Heimat/13879132/Senioren-betreuen-Migrantenkinder/13879141',
  56         'skip_download': True,
  57     }, {
  58         'url': 'http://tvthek.orf.at/profile/Universum/35429',
  59         'skip_download': True,
  60     }]
  61
  62     def _real_extract(self, url):
  63         playlist_id = self._match_id(url)
  64         webpage = self._download_webpage(url, playlist_id)
  65
  66         data_jsb = self._parse_json(
  67             self._search_regex(
  68                 r'<div[^>]+class=(["\']).*?VideoPlaylist.*?\1[^>]+data-jsb=(["\'])(?P<json>.+?)\2',
  69                 webpage, 'playlist', group='json'),
  70             playlist_id, transform_source=unescapeHTML)['playlist']['videos']
  71
  72         def quality_to_int(s):
  73             m = re.search('([0-9]+)', s)
  74             if m is None:
  75                 return -1
  76             return int(m.group(1))
  77
  78         entries = []
  79         for sd in data_jsb:
  80             video_id, title = sd.get('id'), sd.get('title')
  81             if not video_id or not title:
  82                 continue
  83             video_id = compat_str(video_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['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             subtitles = {}
 119             for sub in sd.get('subtitles', []):
 120                 sub_src = sub.get('src')
 121                 if not sub_src:
 122                     continue
 123                 subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({
 124                     'url': sub_src,
 125                 })
 126
 127             upload_date = unified_strdate(sd.get('created_date'))
 128             entries.append({
 129                 '_type': 'video',
 130                 'id': video_id,
 131                 'title': title,
 132                 'formats': formats,
 133                 'subtitles': subtitles,
 134                 'description': sd.get('description'),
 135                 'duration': int_or_none(sd.get('duration_in_seconds')),
 136                 'upload_date': upload_date,
 137                 'thumbnail': sd.get('image_full_url'),
 138             })
 139
 140         return {
 141             '_type': 'playlist',
 142             'entries': entries,
 143             'id': playlist_id,
 144         }
 145
 146
 147 class ORFOE1IE(InfoExtractor):
 148     IE_NAME = 'orf:oe1'
 149     IE_DESC = 'Radio Österreich 1'
 150     _VALID_URL = r'https?://oe1\.orf\.at/(?:programm/|konsole\?.*?\btrack_id=)(?P<id>[0-9]+)'
 151
 152     # Audios on ORF radio are only available for 7 days, so we can't add tests.
 153     _TESTS = [{
 154         'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
 155         'only_matching': True,
 156     }, {
 157         'url': 'http://oe1.orf.at/konsole?show=ondemand&track_id=443608&load_day=/programm/konsole/tag/20160726',
 158         'only_matching': True,
 159     }]
 160
 161     def _real_extract(self, url):
 162         show_id = self._match_id(url)
 163         data = self._download_json(
 164             'http://oe1.orf.at/programm/%s/konsole' % show_id,
 165             show_id
 166         )
 167
 168         timestamp = datetime.datetime.strptime('%s %s' % (
 169             data['item']['day_label'],
 170             data['item']['time']
 171         ), '%d.%m.%Y %H:%M')
 172         unix_timestamp = calendar.timegm(timestamp.utctimetuple())
 173
 174         return {
 175             'id': show_id,
 176             'title': data['item']['title'],
 177             'url': data['item']['url_stream'],
 178             'ext': 'mp3',
 179             'description': data['item'].get('info'),
 180             'timestamp': unix_timestamp
 181         }
 182
 183
 184 class ORFFM4IE(InfoExtractor):
 185     IE_NAME = 'orf:fm4'
 186     IE_DESC = 'radio FM4'
 187     _VALID_URL = r'https?://fm4\.orf\.at/(?:7tage/?#|player/)(?P<date>[0-9]+)/(?P<show>\w+)'
 188
 189     _TEST = {
 190         'url': 'http://fm4.orf.at/player/20160110/IS/',
 191         'md5': '01e736e8f1cef7e13246e880a59ad298',
 192         'info_dict': {
 193             'id': '2016-01-10_2100_tl_54_7DaysSun13_11244',
 194             'ext': 'mp3',
 195             'title': 'Im Sumpf',
 196             'description': 'md5:384c543f866c4e422a55f66a62d669cd',
 197             'duration': 7173,
 198             'timestamp': 1452456073,
 199             'upload_date': '20160110',
 200         },
 201         'skip': 'Live streams on FM4 got deleted soon',
 202     }
 203
 204     def _real_extract(self, url):
 205         mobj = re.match(self._VALID_URL, url)
 206         show_date = mobj.group('date')
 207         show_id = mobj.group('show')
 208
 209         data = self._download_json(
 210             'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
 211             show_id
 212         )
 213
 214         def extract_entry_dict(info, title, subtitle):
 215             return {
 216                 'id': info['loopStreamId'].replace('.mp3', ''),
 217                 'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
 218                 'title': title,
 219                 'description': subtitle,
 220                 'duration': (info['end'] - info['start']) / 1000,
 221                 'timestamp': info['start'] / 1000,
 222                 'ext': 'mp3'
 223             }
 224
 225         entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
 226
 227         return {
 228             '_type': 'playlist',
 229             'id': show_id,
 230             'title': data['title'],
 231             'description': data['subtitle'],
 232             'entries': entries
 233         }
 234
 235
 236 class ORFIPTVIE(InfoExtractor):
 237     IE_NAME = 'orf:iptv'
 238     IE_DESC = 'iptv.ORF.at'
 239     _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
 240
 241     _TEST = {
 242         'url': 'http://iptv.orf.at/stories/2275236/',
 243         'md5': 'c8b22af4718a4b4af58342529453e3e5',
 244         'info_dict': {
 245             'id': '350612',
 246             'ext': 'flv',
 247             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
 248             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
 249             'duration': 68.197,
 250             'thumbnail': r're:^https?://.*\.jpg$',
 251             'upload_date': '20150425',
 252         },
 253     }
 254
 255     def _real_extract(self, url):
 256         story_id = self._match_id(url)
 257
 258         webpage = self._download_webpage(
 259             'http://iptv.orf.at/stories/%s' % story_id, story_id)
 260
 261         video_id = self._search_regex(
 262             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
 263
 264         data = self._download_json(
 265             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 266             video_id)[0]
 267
 268         duration = float_or_none(data['duration'], 1000)
 269
 270         video = data['sources']['default']
 271         load_balancer_url = video['loadBalancerUrl']
 272         abr = int_or_none(video.get('audioBitrate'))
 273         vbr = int_or_none(video.get('bitrate'))
 274         fps = int_or_none(video.get('videoFps'))
 275         width = int_or_none(video.get('videoWidth'))
 276         height = int_or_none(video.get('videoHeight'))
 277         thumbnail = video.get('preview')
 278
 279         rendition = self._download_json(
 280             load_balancer_url, video_id, transform_source=strip_jsonp)
 281
 282         f = {
 283             'abr': abr,
 284             'vbr': vbr,
 285             'fps': fps,
 286             'width': width,
 287             'height': height,
 288         }
 289
 290         formats = []
 291         for format_id, format_url in rendition['redirect'].items():
 292             if format_id == 'rtmp':
 293                 ff = f.copy()
 294                 ff.update({
 295                     'url': format_url,
 296                     'format_id': format_id,
 297                 })
 298                 formats.append(ff)
 299             elif determine_ext(format_url) == 'f4m':
 300                 formats.extend(self._extract_f4m_formats(
 301                     format_url, video_id, f4m_id=format_id))
 302             elif determine_ext(format_url) == 'm3u8':
 303                 formats.extend(self._extract_m3u8_formats(
 304                     format_url, video_id, 'mp4', m3u8_id=format_id))
 305             else:
 306                 continue
 307         self._sort_formats(formats)
 308
 309         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
 310         description = self._og_search_description(webpage)
 311         upload_date = unified_strdate(self._html_search_meta(
 312             'dc.date', webpage, 'upload date'))
 313
 314         return {
 315             'id': video_id,
 316             'title': title,
 317             'description': description,
 318             'duration': duration,
 319             'thumbnail': thumbnail,
 320             'upload_date': upload_date,
 321             'formats': formats,
 322         }