youtube_dl/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     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/(?:[^/]+/){1,2}(?P<id>\d+)'
  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': 'db2a5369238b51f9811ad815b69dc086',
  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
  63     def _real_extract(self, url):
  64         video_id = self._match_id(url)
  65         webpage = self._download_webpage(url, video_id)
  66
  67         mobj = re.search(
  68             r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
  69             webpage)
  70         if mobj is None:
  71             expired = self._html_search_regex(r'(?s)class="expired-(?:video|audio)".+?<span>(.+?)</span>', webpage, 'expired', None)
  72             if expired:
  73                 raise ExtractorError('%s said: %s' % (self.IE_NAME, expired), expected=True)
  74             raise ExtractorError('Unable to extract video urls')
  75
  76         urls_info = self._parse_json(
  77             mobj.group('json_data'), video_id, transform_source=js_to_json)
  78
  79         if not isinstance(urls_info, list):
  80             urls_info = [urls_info]
  81
  82         if mobj.group('type') == 'YouTube':
  83             return self.playlist_result([
  84                 self.url_result(url_info['url']) for url_info in urls_info])
  85
  86         formats = [{
  87             'url': url_info['url'],
  88             'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none',
  89             'width': int_or_none(url_info.get('width')),
  90             'height': int_or_none(url_info.get('height')),
  91             'tbr': int_or_none(url_info.get('bitrate')),
  92             'filesize': int_or_none(url_info.get('filesize')),
  93         } for url_info in urls_info]
  94
  95         self._sort_formats(formats)
  96
  97         return {
  98             'id': video_id,
  99             'title': self._og_search_title(webpage),
 100             'formats': formats,
 101             'description': self._og_search_description(webpage),
 102             'thumbnail': self._og_search_thumbnail(webpage),
 103         }
 104
 105
 106 class ABCIViewIE(InfoExtractor):
 107     IE_NAME = 'abc.net.au:iview'
 108     _VALID_URL = r'https?://iview\.abc\.net\.au/(?:[^/]+/)*video/(?P<id>[^/?#]+)'
 109     _GEO_COUNTRIES = ['AU']
 110
 111     # ABC iview programs are normally available for 14 days only.
 112     _TESTS = [{
 113         'url': 'https://iview.abc.net.au/show/ben-and-hollys-little-kingdom/series/0/video/ZX9371A050S00',
 114         'md5': 'cde42d728b3b7c2b32b1b94b4a548afc',
 115         'info_dict': {
 116             'id': 'ZX9371A050S00',
 117             'ext': 'mp4',
 118             'title': "Gaston's Birthday",
 119             'series': "Ben And Holly's Little Kingdom",
 120             'description': 'md5:f9de914d02f226968f598ac76f105bcf',
 121             'upload_date': '20180604',
 122             'uploader_id': 'abc4kids',
 123             'timestamp': 1528140219,
 124         },
 125         'params': {
 126             'skip_download': True,
 127         },
 128     }]
 129
 130     def _real_extract(self, url):
 131         video_id = self._match_id(url)
 132         video_params = self._download_json(
 133             'https://iview.abc.net.au/api/programs/' + video_id, video_id)
 134         title = unescapeHTML(video_params.get('title') or video_params['seriesTitle'])
 135         stream = next(s for s in video_params['playlist'] if s.get('type') in ('program', 'livestream'))
 136
 137         house_number = video_params.get('episodeHouseNumber') or video_id
 138         path = '/auth/hls/sign?ts={0}&hn={1}&d=android-tablet'.format(
 139             int(time.time()), house_number)
 140         sig = hmac.new(
 141             b'android.content.res.Resources',
 142             path.encode('utf-8'), hashlib.sha256).hexdigest()
 143         token = self._download_webpage(
 144             'http://iview.abc.net.au{0}&sig={1}'.format(path, sig), video_id)
 145
 146         def tokenize_url(url, token):
 147             return update_url_query(url, {
 148                 'hdnea': token,
 149             })
 150
 151         for sd in ('sd', 'sd-low'):
 152             sd_url = try_get(
 153                 stream, lambda x: x['streams']['hls'][sd], compat_str)
 154             if not sd_url:
 155                 continue
 156             formats = self._extract_m3u8_formats(
 157                 tokenize_url(sd_url, token), video_id, 'mp4',
 158                 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
 159             if formats:
 160                 break
 161         self._sort_formats(formats)
 162
 163         subtitles = {}
 164         src_vtt = stream.get('captions', {}).get('src-vtt')
 165         if src_vtt:
 166             subtitles['en'] = [{
 167                 'url': src_vtt,
 168                 'ext': 'vtt',
 169             }]
 170
 171         is_live = video_params.get('livestream') == '1'
 172         if is_live:
 173             title = self._live_title(title)
 174
 175         return {
 176             'id': video_id,
 177             'title': title,
 178             'description': video_params.get('description'),
 179             'thumbnail': video_params.get('thumbnail'),
 180             'duration': int_or_none(video_params.get('eventDuration')),
 181             'timestamp': parse_iso8601(video_params.get('pubDate'), ' '),
 182             'series': unescapeHTML(video_params.get('seriesTitle')),
 183             'series_id': video_params.get('seriesHouseNumber') or video_id[:7],
 184             'season_number': int_or_none(self._search_regex(
 185                 r'\bSeries\s+(\d+)\b', title, 'season number', default=None)),
 186             'episode_number': int_or_none(self._search_regex(
 187                 r'\bEp\s+(\d+)\b', title, 'episode number', default=None)),
 188             'episode_id': house_number,
 189             'uploader_id': video_params.get('channel'),
 190             'formats': formats,
 191             'subtitles': subtitles,
 192             'is_live': is_live,
 193         }