yt_dlp/extractor/abc.py

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