_ Git - youtube-dl/blob - 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 )
  17
  18
  19 class ABCIE(InfoExtractor):
  20     IE_NAME = 'abc.net.au'
  21     _VALID_URL = r'https?://(?:www\.)?abc\.net\.au/news/(?:[^/]+/){1,2}(?P<id>\d+)'
  22
  23     _TESTS = [{
  24         'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334',
  25         'md5': 'cb3dd03b18455a661071ee1e28344d9f',
  26         'info_dict': {
  27             'id': '5868334',
  28             'ext': 'mp4',
  29             'title': 'Australia to help staff Ebola treatment centre in Sierra Leone',
  30             'description': 'md5:809ad29c67a05f54eb41f2a105693a67',
  31         },
  32         'skip': 'this video has expired',
  33     }, {
  34         'url': 'http://www.abc.net.au/news/2015-08-17/warren-entsch-introduces-same-sex-marriage-bill/6702326',
  35         'md5': 'db2a5369238b51f9811ad815b69dc086',
  36         'info_dict': {
  37             'id': 'NvqvPeNZsHU',
  38             'ext': 'mp4',
  39             'upload_date': '20150816',
  40             'uploader': 'ABC News (Australia)',
  41             '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',
  42             'uploader_id': 'NewsOnABC',
  43             'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill',
  44         },
  45         'add_ie': ['Youtube'],
  46         'skip': 'Not accessible from Travis CI server',
  47     }, {
  48         'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080',
  49         'md5': 'b96eee7c9edf4fc5a358a0252881cc1f',
  50         'info_dict': {
  51             'id': '6880080',
  52             'ext': 'mp3',
  53             'title': 'NAB lifts interest rates, following Westpac and CBA',
  54             'description': 'md5:f13d8edc81e462fce4a0437c7dc04728',
  55         },
  56     }, {
  57         'url': 'http://www.abc.net.au/news/2015-10-19/6866214',
  58         'only_matching': True,
  59     }]
  60
  61     def _real_extract(self, url):
  62         video_id = self._match_id(url)
  63         webpage = self._download_webpage(url, video_id)
  64
  65         mobj = re.search(
  66             r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
  67             webpage)
  68         if mobj is None:
  69             expired = self._html_search_regex(r'(?s)class="expired-(?:video|audio)".+?<span>(.+?)</span>', webpage, 'expired', None)
  70             if expired:
  71                 raise ExtractorError('%s said: %s' % (self.IE_NAME, expired), expected=True)
  72             raise ExtractorError('Unable to extract video urls')
  73
  74         urls_info = self._parse_json(
  75             mobj.group('json_data'), video_id, transform_source=js_to_json)
  76
  77         if not isinstance(urls_info, list):
  78             urls_info = [urls_info]
  79
  80         if mobj.group('type') == 'YouTube':
  81             return self.playlist_result([
  82                 self.url_result(url_info['url']) for url_info in urls_info])
  83
  84         formats = [{
  85             'url': url_info['url'],
  86             'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none',
  87             'width': int_or_none(url_info.get('width')),
  88             'height': int_or_none(url_info.get('height')),
  89             'tbr': int_or_none(url_info.get('bitrate')),
  90             'filesize': int_or_none(url_info.get('filesize')),
  91         } for url_info in urls_info]
  92
  93         self._sort_formats(formats)
  94
  95         return {
  96             'id': video_id,
  97             'title': self._og_search_title(webpage),
  98             'formats': formats,
  99             'description': self._og_search_description(webpage),
 100             'thumbnail': self._og_search_thumbnail(webpage),
 101         }
 102
 103
 104 class ABCIViewIE(InfoExtractor):
 105     IE_NAME = 'abc.net.au:iview'
 106     _VALID_URL = r'https?://iview\.abc\.net\.au/programs/[^/]+/(?P<id>[^/?#]+)'
 107
 108     # ABC iview programs are normally available for 14 days only.
 109     _TESTS = [{
 110         'url': 'http://iview.abc.net.au/programs/diaries-of-a-broken-mind/ZX9735A001S00',
 111         'md5': 'cde42d728b3b7c2b32b1b94b4a548afc',
 112         'info_dict': {
 113             'id': 'ZX9735A001S00',
 114             'ext': 'mp4',
 115             'title': 'Diaries Of A Broken Mind',
 116             'description': 'md5:7de3903874b7a1be279fe6b68718fc9e',
 117             'upload_date': '20161010',
 118             'uploader_id': 'abc2',
 119             'timestamp': 1476064920,
 120         },
 121         'skip': 'Video gone',
 122     }]
 123
 124     def _real_extract(self, url):
 125         video_id = self._match_id(url)
 126         webpage = self._download_webpage(url, video_id)
 127         video_params = self._parse_json(self._search_regex(
 128             r'videoParams\s*=\s*({.+?});', webpage, 'video params'), video_id)
 129         title = video_params.get('title') or video_params['seriesTitle']
 130         stream = next(s for s in video_params['playlist'] if s.get('type') == 'program')
 131
 132         key = 'android.content.res.Resources'.encode('utf-8')
 133         time_str = str(int(time.time()))
 134         house_number = video_params.get('episodeHouseNumber')
 135         path = '/auth/hls/sign?ts={0}&hn={1}&d=android-mobile'.format(
 136             time_str, house_number)
 137         sig = hmac.new(key, path.encode('utf-8'), hashlib.sha256).hexdigest()
 138         auth_url = 'http://iview.abc.net.au{0}&sig={1}'.format(path, sig)
 139         token = self._download_webpage(auth_url, video_id)
 140
 141         format_urls = []
 142
 143         def tokenize_url(url, token):
 144             return ''.join([url, '?hdnea=', token])
 145
 146         # May have higher quality video
 147         sd_url = try_get(
 148             stream, lambda x: x['streams']['hls']['sd'], compat_str)
 149         if sd_url:
 150             format_urls.append(tokenize_url(sd_url, token))
 151         else:
 152             sd_low_url = try_get(
 153                 stream, lambda x: x['streams']['hls']['sd-low'], compat_str)
 154             format_urls.append(tokenize_url(sd_low_url, token))
 155
 156         formats = []
 157         for format_url in format_urls:
 158             if format_url:
 159                 formats.extend(
 160                     self._extract_m3u8_formats(format_url, video_id, 'mp4'))
 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         return {
 172             'id': video_id,
 173             'title': title,
 174             'description': self._html_search_meta(['og:description', 'twitter:description'], webpage),
 175             'thumbnail': self._html_search_meta(['og:image', 'twitter:image:src'], webpage),
 176             'duration': int_or_none(video_params.get('eventDuration')),
 177             'timestamp': parse_iso8601(video_params.get('pubDate'), ' '),
 178             'series': video_params.get('seriesTitle'),
 179             'series_id': video_params.get('seriesHouseNumber') or video_id[:7],
 180             'episode_number': int_or_none(self._html_search_meta('episodeNumber', webpage, default=None)),
 181             'episode': self._html_search_meta('episode_title', webpage, default=None),
 182             'uploader_id': video_params.get('channel'),
 183             'formats': formats,
 184             'subtitles': subtitles,
 185         }