yt_dlp/extractor/abc.py

   1 import hashlib
   2 import hmac
   3 import re
   4 import time
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     dict_get,
  10     ExtractorError,
  11     js_to_json,
  12     int_or_none,
  13     parse_iso8601,
  14     str_or_none,
  15     traverse_obj,
  16     try_get,
  17     unescapeHTML,
  18     update_url_query,
  19 )
  20
  21
  22 class ABCIE(InfoExtractor):
  23     IE_NAME = 'abc.net.au'
  24     _VALID_URL = r'https?://(?:www\.)?abc\.net\.au/(?:news|btn)/(?:[^/]+/){1,4}(?P<id>\d{5,})'
  25
  26     _TESTS = [{
  27         'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334',
  28         'md5': 'cb3dd03b18455a661071ee1e28344d9f',
  29         'info_dict': {
  30             'id': '5868334',
  31             'ext': 'mp4',
  32             'title': 'Australia to help staff Ebola treatment centre in Sierra Leone',
  33             'description': 'md5:809ad29c67a05f54eb41f2a105693a67',
  34         },
  35         'skip': 'this video has expired',
  36     }, {
  37         'url': 'http://www.abc.net.au/news/2015-08-17/warren-entsch-introduces-same-sex-marriage-bill/6702326',
  38         'md5': '4ebd61bdc82d9a8b722f64f1f4b4d121',
  39         'info_dict': {
  40             'id': 'NvqvPeNZsHU',
  41             'ext': 'mp4',
  42             'upload_date': '20150816',
  43             'uploader': 'ABC News (Australia)',
  44             'description': 'Government backbencher Warren Entsch introduces a cross-party sponsored bill to legalise same-sex marriage, saying the bill is designed to promote "an inclusive Australia, not a divided one.". Read more here: http://ab.co/1Mwc6ef',
  45             'uploader_id': 'NewsOnABC',
  46             'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill',
  47         },
  48         'add_ie': ['Youtube'],
  49         'skip': 'Not accessible from Travis CI server',
  50     }, {
  51         'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080',
  52         'md5': 'b96eee7c9edf4fc5a358a0252881cc1f',
  53         'info_dict': {
  54             'id': '6880080',
  55             'ext': 'mp3',
  56             'title': 'NAB lifts interest rates, following Westpac and CBA',
  57             'description': 'md5:f13d8edc81e462fce4a0437c7dc04728',
  58         },
  59     }, {
  60         'url': 'http://www.abc.net.au/news/2015-10-19/6866214',
  61         'only_matching': True,
  62     }, {
  63         'url': 'https://www.abc.net.au/btn/classroom/wwi-centenary/10527914',
  64         'info_dict': {
  65             'id': '10527914',
  66             'ext': 'mp4',
  67             'title': 'WWI Centenary',
  68             'description': 'md5:c2379ec0ca84072e86b446e536954546',
  69         }
  70     }, {
  71         'url': 'https://www.abc.net.au/news/programs/the-world/2020-06-10/black-lives-matter-protests-spawn-support-for/12342074',
  72         'info_dict': {
  73             'id': '12342074',
  74             'ext': 'mp4',
  75             'title': 'Black Lives Matter protests spawn support for Papuans in Indonesia',
  76             'description': 'md5:2961a17dc53abc558589ccd0fb8edd6f',
  77         }
  78     }, {
  79         'url': 'https://www.abc.net.au/btn/newsbreak/btn-newsbreak-20200814/12560476',
  80         'info_dict': {
  81             'id': 'tDL8Ld4dK_8',
  82             'ext': 'mp4',
  83             'title': 'Fortnite Banned From Apple and Google App Stores',
  84             'description': 'md5:a6df3f36ce8f816b74af4bd6462f5651',
  85             'upload_date': '20200813',
  86             'uploader': 'Behind the News',
  87             'uploader_id': 'behindthenews',
  88         }
  89     }, {
  90         'url': 'https://www.abc.net.au/news/2023-06-25/wagner-boss-orders-troops-back-to-bases-to-avoid-bloodshed/102520540',
  91         'info_dict': {
  92             'id': '102520540',
  93             'title': 'Wagner Group retreating from Russia, leader Prigozhin to move to Belarus',
  94             'ext': 'mp4',
  95             'description': 'Wagner troops leave Rostov-on-Don and\xa0Yevgeny Prigozhin will move to Belarus under a deal brokered by Belarusian President Alexander Lukashenko to end the mutiny.',
  96             'thumbnail': 'https://live-production.wcms.abc-cdn.net.au/0c170f5b57f0105c432f366c0e8e267b?impolicy=wcms_crop_resize&cropH=2813&cropW=5000&xPos=0&yPos=249&width=862&height=485',
  97         }
  98     }]
  99
 100     def _real_extract(self, url):
 101         video_id = self._match_id(url)
 102         webpage = self._download_webpage(url, video_id)
 103
 104         mobj = re.search(r'<a\s+href="(?P<url>[^"]+)"\s+data-duration="\d+"\s+title="Download audio directly">', webpage)
 105         if mobj:
 106             urls_info = mobj.groupdict()
 107             youtube = False
 108             video = False
 109         else:
 110             mobj = re.search(r'<a href="(?P<url>http://www\.youtube\.com/watch\?v=[^"]+)"><span><strong>External Link:</strong>',
 111                              webpage)
 112             if mobj is None:
 113                 mobj = re.search(r'<iframe width="100%" src="(?P<url>//www\.youtube-nocookie\.com/embed/[^?"]+)', webpage)
 114             if mobj:
 115                 urls_info = mobj.groupdict()
 116                 youtube = True
 117                 video = True
 118
 119         if mobj is None:
 120             mobj = re.search(r'(?P<type>)"(?:sources|files|renditions)":\s*(?P<json_data>\[[^\]]+\])', webpage)
 121             if mobj is None:
 122                 mobj = re.search(
 123                     r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
 124                     webpage)
 125                 if mobj is None:
 126                     expired = self._html_search_regex(r'(?s)class="expired-(?:video|audio)".+?<span>(.+?)</span>', webpage, 'expired', None)
 127                     if expired:
 128                         raise ExtractorError('%s said: %s' % (self.IE_NAME, expired), expected=True)
 129                     raise ExtractorError('Unable to extract video urls')
 130
 131             urls_info = self._parse_json(
 132                 mobj.group('json_data'), video_id, transform_source=js_to_json)
 133             youtube = mobj.group('type') == 'YouTube'
 134             video = mobj.group('type') == 'Video' or traverse_obj(
 135                 urls_info, (0, ('contentType', 'MIMEType')), get_all=False) == 'video/mp4'
 136
 137         if not isinstance(urls_info, list):
 138             urls_info = [urls_info]
 139
 140         if youtube:
 141             return self.playlist_result([
 142                 self.url_result(url_info['url']) for url_info in urls_info])
 143
 144         formats = []
 145         for url_info in urls_info:
 146             height = int_or_none(url_info.get('height'))
 147             bitrate = int_or_none(url_info.get('bitrate'))
 148             width = int_or_none(url_info.get('width'))
 149             format_id = None
 150             mobj = re.search(r'_(?:(?P<height>\d+)|(?P<bitrate>\d+)k)\.mp4$', url_info['url'])
 151             if mobj:
 152                 height_from_url = mobj.group('height')
 153                 if height_from_url:
 154                     height = height or int_or_none(height_from_url)
 155                     width = width or int_or_none(url_info.get('label'))
 156                 else:
 157                     bitrate = bitrate or int_or_none(mobj.group('bitrate'))
 158                     format_id = str_or_none(url_info.get('label'))
 159             formats.append({
 160                 'url': url_info['url'],
 161                 'vcodec': url_info.get('codec') if video else 'none',
 162                 'width': width,
 163                 'height': height,
 164                 'tbr': bitrate,
 165                 'filesize': int_or_none(url_info.get('filesize')),
 166                 'format_id': format_id
 167             })
 168
 169         return {
 170             'id': video_id,
 171             'title': self._og_search_title(webpage),
 172             'formats': formats,
 173             'description': self._og_search_description(webpage),
 174             'thumbnail': self._og_search_thumbnail(webpage),
 175         }
 176
 177
 178 class ABCIViewIE(InfoExtractor):
 179     IE_NAME = 'abc.net.au:iview'
 180     _VALID_URL = r'https?://iview\.abc\.net\.au/(?:[^/]+/)*video/(?P<id>[^/?#]+)'
 181     _GEO_COUNTRIES = ['AU']
 182
 183     _TESTS = [{
 184         'url': 'https://iview.abc.net.au/show/gruen/series/11/video/LE1927H001S00',
 185         'md5': '67715ce3c78426b11ba167d875ac6abf',
 186         'info_dict': {
 187             'id': 'LE1927H001S00',
 188             'ext': 'mp4',
 189             'title': "Series 11 Ep 1",
 190             'series': "Gruen",
 191             'description': 'md5:52cc744ad35045baf6aded2ce7287f67',
 192             'upload_date': '20190925',
 193             'uploader_id': 'abc1',
 194             'timestamp': 1569445289,
 195         },
 196         'params': {
 197             'skip_download': True,
 198         },
 199     }]
 200
 201     def _real_extract(self, url):
 202         video_id = self._match_id(url)
 203         video_params = self._download_json(
 204             'https://iview.abc.net.au/api/programs/' + video_id, video_id)
 205         title = unescapeHTML(video_params.get('title') or video_params['seriesTitle'])
 206         stream = next(s for s in video_params['playlist'] if s.get('type') in ('program', 'livestream'))
 207
 208         house_number = video_params.get('episodeHouseNumber') or video_id
 209         path = '/auth/hls/sign?ts={0}&hn={1}&d=android-tablet'.format(
 210             int(time.time()), house_number)
 211         sig = hmac.new(
 212             b'android.content.res.Resources',
 213             path.encode('utf-8'), hashlib.sha256).hexdigest()
 214         token = self._download_webpage(
 215             'http://iview.abc.net.au{0}&sig={1}'.format(path, sig), video_id)
 216
 217         def tokenize_url(url, token):
 218             return update_url_query(url, {
 219                 'hdnea': token,
 220             })
 221
 222         for sd in ('1080', '720', 'sd', 'sd-low'):
 223             sd_url = try_get(
 224                 stream, lambda x: x['streams']['hls'][sd], compat_str)
 225             if not sd_url:
 226                 continue
 227             formats = self._extract_m3u8_formats(
 228                 tokenize_url(sd_url, token), video_id, 'mp4',
 229                 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
 230             if formats:
 231                 break
 232
 233         subtitles = {}
 234         src_vtt = stream.get('captions', {}).get('src-vtt')
 235         if src_vtt:
 236             subtitles['en'] = [{
 237                 'url': src_vtt,
 238                 'ext': 'vtt',
 239             }]
 240
 241         is_live = video_params.get('livestream') == '1'
 242
 243         return {
 244             'id': video_id,
 245             'title': title,
 246             'description': video_params.get('description'),
 247             'thumbnail': video_params.get('thumbnail'),
 248             'duration': int_or_none(video_params.get('eventDuration')),
 249             'timestamp': parse_iso8601(video_params.get('pubDate'), ' '),
 250             'series': unescapeHTML(video_params.get('seriesTitle')),
 251             'series_id': video_params.get('seriesHouseNumber') or video_id[:7],
 252             'season_number': int_or_none(self._search_regex(
 253                 r'\bSeries\s+(\d+)\b', title, 'season number', default=None)),
 254             'episode_number': int_or_none(self._search_regex(
 255                 r'\bEp\s+(\d+)\b', title, 'episode number', default=None)),
 256             'episode_id': house_number,
 257             'uploader_id': video_params.get('channel'),
 258             'formats': formats,
 259             'subtitles': subtitles,
 260             'is_live': is_live,
 261         }
 262
 263
 264 class ABCIViewShowSeriesIE(InfoExtractor):
 265     IE_NAME = 'abc.net.au:iview:showseries'
 266     _VALID_URL = r'https?://iview\.abc\.net\.au/show/(?P<id>[^/]+)(?:/series/\d+)?$'
 267     _GEO_COUNTRIES = ['AU']
 268
 269     _TESTS = [{
 270         'url': 'https://iview.abc.net.au/show/upper-middle-bogan',
 271         'info_dict': {
 272             'id': '124870-1',
 273             'title': 'Series 1',
 274             'description': 'md5:93119346c24a7c322d446d8eece430ff',
 275             'series': 'Upper Middle Bogan',
 276             'season': 'Series 1',
 277             'thumbnail': r're:^https?://cdn\.iview\.abc\.net\.au/thumbs/.*\.jpg$'
 278         },
 279         'playlist_count': 8,
 280     }, {
 281         'url': 'https://iview.abc.net.au/show/upper-middle-bogan',
 282         'info_dict': {
 283             'id': 'CO1108V001S00',
 284             'ext': 'mp4',
 285             'title': 'Series 1 Ep 1 I\'m A Swan',
 286             'description': 'md5:7b676758c1de11a30b79b4d301e8da93',
 287             'series': 'Upper Middle Bogan',
 288             'uploader_id': 'abc1',
 289             'upload_date': '20210630',
 290             'timestamp': 1625036400,
 291         },
 292         'params': {
 293             'noplaylist': True,
 294             'skip_download': 'm3u8',
 295         },
 296     }]
 297
 298     def _real_extract(self, url):
 299         show_id = self._match_id(url)
 300         webpage = self._download_webpage(url, show_id)
 301         webpage_data = self._search_regex(
 302             r'window\.__INITIAL_STATE__\s*=\s*[\'"](.+?)[\'"]\s*;',
 303             webpage, 'initial state')
 304         video_data = self._parse_json(
 305             unescapeHTML(webpage_data).encode('utf-8').decode('unicode_escape'), show_id)
 306         video_data = video_data['route']['pageData']['_embedded']
 307
 308         highlight = try_get(video_data, lambda x: x['highlightVideo']['shareUrl'])
 309         if not self._yes_playlist(show_id, bool(highlight), video_label='highlight video'):
 310             return self.url_result(highlight, ie=ABCIViewIE.ie_key())
 311
 312         series = video_data['selectedSeries']
 313         return {
 314             '_type': 'playlist',
 315             'entries': [self.url_result(episode['shareUrl'])
 316                         for episode in series['_embedded']['videoEpisodes']],
 317             'id': series.get('id'),
 318             'title': dict_get(series, ('title', 'displaySubtitle')),
 319             'description': series.get('description'),
 320             'series': dict_get(series, ('showTitle', 'displayTitle')),
 321             'season': dict_get(series, ('title', 'displaySubtitle')),
 322             'thumbnail': series.get('thumbnail'),
 323         }