youtube_dlc/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     ExtractorError,
  12     js_to_json,
  13     int_or_none,
  14     parse_iso8601,
  15     str_or_none,
  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
  91     def _real_extract(self, url):
  92         video_id = self._match_id(url)
  93         webpage = self._download_webpage(url, video_id)
  94
  95         mobj = re.search(r'<a\s+href="(?P<url>[^"]+)"\s+data-duration="\d+"\s+title="Download audio directly">', webpage)
  96         if mobj:
  97             urls_info = mobj.groupdict()
  98             youtube = False
  99             video = False
 100         else:
 101             mobj = re.search(r'<a href="(?P<url>http://www\.youtube\.com/watch\?v=[^"]+)"><span><strong>External Link:</strong>',
 102                              webpage)
 103             if mobj is None:
 104                 mobj = re.search(r'<iframe width="100%" src="(?P<url>//www\.youtube-nocookie\.com/embed/[^?"]+)', webpage)
 105             if mobj:
 106                 urls_info = mobj.groupdict()
 107                 youtube = True
 108                 video = True
 109
 110         if mobj is None:
 111             mobj = re.search(r'(?P<type>)"sources": (?P<json_data>\[[^\]]+\]),', webpage)
 112             if mobj is None:
 113                 mobj = re.search(
 114                     r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
 115                     webpage)
 116                 if mobj is None:
 117                     expired = self._html_search_regex(r'(?s)class="expired-(?:video|audio)".+?<span>(.+?)</span>', webpage, 'expired', None)
 118                     if expired:
 119                         raise ExtractorError('%s said: %s' % (self.IE_NAME, expired), expected=True)
 120                     raise ExtractorError('Unable to extract video urls')
 121
 122             urls_info = self._parse_json(
 123                 mobj.group('json_data'), video_id, transform_source=js_to_json)
 124             youtube = mobj.group('type') == 'YouTube'
 125             video = mobj.group('type') == 'Video' or urls_info[0]['contentType'] == 'video/mp4'
 126
 127         if not isinstance(urls_info, list):
 128             urls_info = [urls_info]
 129
 130         if youtube:
 131             return self.playlist_result([
 132                 self.url_result(url_info['url']) for url_info in urls_info])
 133
 134         formats = []
 135         for url_info in urls_info:
 136             height = int_or_none(url_info.get('height'))
 137             bitrate = int_or_none(url_info.get('bitrate'))
 138             width = int_or_none(url_info.get('width'))
 139             format_id = None
 140             mobj = re.search(r'_(?:(?P<height>\d+)|(?P<bitrate>\d+)k)\.mp4$', url_info['url'])
 141             if mobj:
 142                 height_from_url = mobj.group('height')
 143                 if height_from_url:
 144                     height = height or int_or_none(height_from_url)
 145                     width = width or int_or_none(url_info.get('label'))
 146                 else:
 147                     bitrate = bitrate or int_or_none(mobj.group('bitrate'))
 148                     format_id = str_or_none(url_info.get('label'))
 149             formats.append({
 150                 'url': url_info['url'],
 151                 'vcodec': url_info.get('codec') if video else 'none',
 152                 'width': width,
 153                 'height': height,
 154                 'tbr': bitrate,
 155                 'filesize': int_or_none(url_info.get('filesize')),
 156                 'format_id': format_id
 157             })
 158
 159         self._sort_formats(formats)
 160
 161         return {
 162             'id': video_id,
 163             'title': self._og_search_title(webpage),
 164             'formats': formats,
 165             'description': self._og_search_description(webpage),
 166             'thumbnail': self._og_search_thumbnail(webpage),
 167         }
 168
 169
 170 class ABCIViewIE(InfoExtractor):
 171     IE_NAME = 'abc.net.au:iview'
 172     _VALID_URL = r'https?://iview\.abc\.net\.au/(?:[^/]+/)*video/(?P<id>[^/?#]+)'
 173     _GEO_COUNTRIES = ['AU']
 174
 175     # ABC iview programs are normally available for 14 days only.
 176     _TESTS = [{
 177         'url': 'https://iview.abc.net.au/show/gruen/series/11/video/LE1927H001S00',
 178         'md5': '67715ce3c78426b11ba167d875ac6abf',
 179         'info_dict': {
 180             'id': 'LE1927H001S00',
 181             'ext': 'mp4',
 182             'title': "Series 11 Ep 1",
 183             'series': "Gruen",
 184             'description': 'md5:52cc744ad35045baf6aded2ce7287f67',
 185             'upload_date': '20190925',
 186             'uploader_id': 'abc1',
 187             'timestamp': 1569445289,
 188         },
 189         'params': {
 190             'skip_download': True,
 191         },
 192     }]
 193
 194     def _real_extract(self, url):
 195         video_id = self._match_id(url)
 196         video_params = self._download_json(
 197             'https://iview.abc.net.au/api/programs/' + video_id, video_id)
 198         title = unescapeHTML(video_params.get('title') or video_params['seriesTitle'])
 199         stream = next(s for s in video_params['playlist'] if s.get('type') in ('program', 'livestream'))
 200
 201         house_number = video_params.get('episodeHouseNumber') or video_id
 202         path = '/auth/hls/sign?ts={0}&hn={1}&d=android-tablet'.format(
 203             int(time.time()), house_number)
 204         sig = hmac.new(
 205             b'android.content.res.Resources',
 206             path.encode('utf-8'), hashlib.sha256).hexdigest()
 207         token = self._download_webpage(
 208             'http://iview.abc.net.au{0}&sig={1}'.format(path, sig), video_id)
 209
 210         def tokenize_url(url, token):
 211             return update_url_query(url, {
 212                 'hdnea': token,
 213             })
 214
 215         for sd in ('720', 'sd', 'sd-low'):
 216             sd_url = try_get(
 217                 stream, lambda x: x['streams']['hls'][sd], compat_str)
 218             if not sd_url:
 219                 continue
 220             formats = self._extract_m3u8_formats(
 221                 tokenize_url(sd_url, token), video_id, 'mp4',
 222                 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
 223             if formats:
 224                 break
 225         self._sort_formats(formats)
 226
 227         subtitles = {}
 228         src_vtt = stream.get('captions', {}).get('src-vtt')
 229         if src_vtt:
 230             subtitles['en'] = [{
 231                 'url': src_vtt,
 232                 'ext': 'vtt',
 233             }]
 234
 235         is_live = video_params.get('livestream') == '1'
 236         if is_live:
 237             title = self._live_title(title)
 238
 239         return {
 240             'id': video_id,
 241             'title': title,
 242             'description': video_params.get('description'),
 243             'thumbnail': video_params.get('thumbnail'),
 244             'duration': int_or_none(video_params.get('eventDuration')),
 245             'timestamp': parse_iso8601(video_params.get('pubDate'), ' '),
 246             'series': unescapeHTML(video_params.get('seriesTitle')),
 247             'series_id': video_params.get('seriesHouseNumber') or video_id[:7],
 248             'season_number': int_or_none(self._search_regex(
 249                 r'\bSeries\s+(\d+)\b', title, 'season number', default=None)),
 250             'episode_number': int_or_none(self._search_regex(
 251                 r'\bEp\s+(\d+)\b', title, 'episode number', default=None)),
 252             'episode_id': house_number,
 253             'uploader_id': video_params.get('channel'),
 254             'formats': formats,
 255             'subtitles': subtitles,
 256             'is_live': is_live,
 257         }