Raphaël G. Git Repositories - youtubedl/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     update_url_query,
  17 )
  18
  19
  20 class ABCIE(InfoExtractor):
  21     IE_NAME = 'abc.net.au'
  22     _VALID_URL = r'https?://(?:www\.)?abc\.net\.au/news/(?:[^/]+/){1,2}(?P<id>\d+)'
  23
  24     _TESTS = [{
  25         'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334',
  26         'md5': 'cb3dd03b18455a661071ee1e28344d9f',
  27         'info_dict': {
  28             'id': '5868334',
  29             'ext': 'mp4',
  30             'title': 'Australia to help staff Ebola treatment centre in Sierra Leone',
  31             'description': 'md5:809ad29c67a05f54eb41f2a105693a67',
  32         },
  33         'skip': 'this video has expired',
  34     }, {
  35         'url': 'http://www.abc.net.au/news/2015-08-17/warren-entsch-introduces-same-sex-marriage-bill/6702326',
  36         'md5': 'db2a5369238b51f9811ad815b69dc086',
  37         'info_dict': {
  38             'id': 'NvqvPeNZsHU',
  39             'ext': 'mp4',
  40             'upload_date': '20150816',
  41             'uploader': 'ABC News (Australia)',
  42             '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',
  43             'uploader_id': 'NewsOnABC',
  44             'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill',
  45         },
  46         'add_ie': ['Youtube'],
  47         'skip': 'Not accessible from Travis CI server',
  48     }, {
  49         'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080',
  50         'md5': 'b96eee7c9edf4fc5a358a0252881cc1f',
  51         'info_dict': {
  52             'id': '6880080',
  53             'ext': 'mp3',
  54             'title': 'NAB lifts interest rates, following Westpac and CBA',
  55             'description': 'md5:f13d8edc81e462fce4a0437c7dc04728',
  56         },
  57     }, {
  58         'url': 'http://www.abc.net.au/news/2015-10-19/6866214',
  59         'only_matching': True,
  60     }]
  61
  62     def _real_extract(self, url):
  63         video_id = self._match_id(url)
  64         webpage = self._download_webpage(url, video_id)
  65
  66         mobj = re.search(
  67             r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
  68             webpage)
  69         if mobj is None:
  70             expired = self._html_search_regex(r'(?s)class="expired-(?:video|audio)".+?<span>(.+?)</span>', webpage, 'expired', None)
  71             if expired:
  72                 raise ExtractorError('%s said: %s' % (self.IE_NAME, expired), expected=True)
  73             raise ExtractorError('Unable to extract video urls')
  74
  75         urls_info = self._parse_json(
  76             mobj.group('json_data'), video_id, transform_source=js_to_json)
  77
  78         if not isinstance(urls_info, list):
  79             urls_info = [urls_info]
  80
  81         if mobj.group('type') == 'YouTube':
  82             return self.playlist_result([
  83                 self.url_result(url_info['url']) for url_info in urls_info])
  84
  85         formats = [{
  86             'url': url_info['url'],
  87             'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none',
  88             'width': int_or_none(url_info.get('width')),
  89             'height': int_or_none(url_info.get('height')),
  90             'tbr': int_or_none(url_info.get('bitrate')),
  91             'filesize': int_or_none(url_info.get('filesize')),
  92         } for url_info in urls_info]
  93
  94         self._sort_formats(formats)
  95
  96         return {
  97             'id': video_id,
  98             'title': self._og_search_title(webpage),
  99             'formats': formats,
 100             'description': self._og_search_description(webpage),
 101             'thumbnail': self._og_search_thumbnail(webpage),
 102         }
 103
 104
 105 class ABCIViewIE(InfoExtractor):
 106     IE_NAME = 'abc.net.au:iview'
 107     _VALID_URL = r'https?://iview\.abc\.net\.au/programs/[^/]+/(?P<id>[^/?#]+)'
 108     _GEO_COUNTRIES = ['AU']
 109
 110     # ABC iview programs are normally available for 14 days only.
 111     _TESTS = [{
 112         'url': 'http://iview.abc.net.au/programs/call-the-midwife/ZW0898A003S00',
 113         'md5': 'cde42d728b3b7c2b32b1b94b4a548afc',
 114         'info_dict': {
 115             'id': 'ZW0898A003S00',
 116             'ext': 'mp4',
 117             'title': 'Series 5 Ep 3',
 118             'description': 'md5:e0ef7d4f92055b86c4f33611f180ed79',
 119             'upload_date': '20171228',
 120             'uploader_id': 'abc1',
 121             'timestamp': 1514499187,
 122         },
 123         'params': {
 124             'skip_download': True,
 125         },
 126     }]
 127
 128     def _real_extract(self, url):
 129         video_id = self._match_id(url)
 130         webpage = self._download_webpage(url, video_id)
 131         video_params = self._parse_json(self._search_regex(
 132             r'videoParams\s*=\s*({.+?});', webpage, 'video params'), video_id)
 133         title = video_params.get('title') or video_params['seriesTitle']
 134         stream = next(s for s in video_params['playlist'] if s.get('type') == 'program')
 135
 136         house_number = video_params.get('episodeHouseNumber')
 137         path = '/auth/hls/sign?ts={0}&hn={1}&d=android-mobile'.format(
 138             int(time.time()), house_number)
 139         sig = hmac.new(
 140             'android.content.res.Resources'.encode('utf-8'),
 141             path.encode('utf-8'), hashlib.sha256).hexdigest()
 142         token = self._download_webpage(
 143             'http://iview.abc.net.au{0}&sig={1}'.format(path, sig), video_id)
 144
 145         def tokenize_url(url, token):
 146             return update_url_query(url, {
 147                 'hdnea': token,
 148             })
 149
 150         for sd in ('sd', 'sd-low'):
 151             sd_url = try_get(
 152                 stream, lambda x: x['streams']['hls'][sd], compat_str)
 153             if not sd_url:
 154                 continue
 155             formats = self._extract_m3u8_formats(
 156                 tokenize_url(sd_url, token), video_id, 'mp4',
 157                 entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
 158             if formats:
 159                 break
 160         self._sort_formats(formats)
 161
 162         subtitles = {}
 163         src_vtt = stream.get('captions', {}).get('src-vtt')
 164         if src_vtt:
 165             subtitles['en'] = [{
 166                 'url': src_vtt,
 167                 'ext': 'vtt',
 168             }]
 169
 170         return {
 171             'id': video_id,
 172             'title': title,
 173             'description': self._html_search_meta(['og:description', 'twitter:description'], webpage),
 174             'thumbnail': self._html_search_meta(['og:image', 'twitter:image:src'], webpage),
 175             'duration': int_or_none(video_params.get('eventDuration')),
 176             'timestamp': parse_iso8601(video_params.get('pubDate'), ' '),
 177             'series': video_params.get('seriesTitle'),
 178             'series_id': video_params.get('seriesHouseNumber') or video_id[:7],
 179             'episode_number': int_or_none(self._html_search_meta('episodeNumber', webpage, default=None)),
 180             'episode': self._html_search_meta('episode_title', webpage, default=None),
 181             'uploader_id': video_params.get('channel'),
 182             'formats': formats,
 183             'subtitles': subtitles,
 184         }