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