_ Git - youtube-dl/blob - youtube_dl/extractor/abc.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     js_to_json,
   9     int_or_none,
  10     update_url_query,
  11     parse_iso8601,
  12 )
  13
  14
  15 class ABCIE(InfoExtractor):
  16     IE_NAME = 'abc.net.au'
  17     _VALID_URL = r'https?://www\.abc\.net\.au/news/(?:[^/]+/){1,2}(?P<id>\d+)'
  18
  19     _TESTS = [{
  20         'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334',
  21         'md5': 'cb3dd03b18455a661071ee1e28344d9f',
  22         'info_dict': {
  23             'id': '5868334',
  24             'ext': 'mp4',
  25             'title': 'Australia to help staff Ebola treatment centre in Sierra Leone',
  26             'description': 'md5:809ad29c67a05f54eb41f2a105693a67',
  27         },
  28         'skip': 'this video has expired',
  29     }, {
  30         'url': 'http://www.abc.net.au/news/2015-08-17/warren-entsch-introduces-same-sex-marriage-bill/6702326',
  31         'md5': 'db2a5369238b51f9811ad815b69dc086',
  32         'info_dict': {
  33             'id': 'NvqvPeNZsHU',
  34             'ext': 'mp4',
  35             'upload_date': '20150816',
  36             'uploader': 'ABC News (Australia)',
  37             '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',
  38             'uploader_id': 'NewsOnABC',
  39             'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill',
  40         },
  41         'add_ie': ['Youtube'],
  42         'skip': 'Not accessible from Travis CI server',
  43     }, {
  44         'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080',
  45         'md5': 'b96eee7c9edf4fc5a358a0252881cc1f',
  46         'info_dict': {
  47             'id': '6880080',
  48             'ext': 'mp3',
  49             'title': 'NAB lifts interest rates, following Westpac and CBA',
  50             'description': 'md5:f13d8edc81e462fce4a0437c7dc04728',
  51         },
  52     }, {
  53         'url': 'http://www.abc.net.au/news/2015-10-19/6866214',
  54         'only_matching': True,
  55     }]
  56
  57     def _real_extract(self, url):
  58         video_id = self._match_id(url)
  59         webpage = self._download_webpage(url, video_id)
  60
  61         mobj = re.search(
  62             r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
  63             webpage)
  64         if mobj is None:
  65             expired = self._html_search_regex(r'(?s)class="expired-(?:video|audio)".+?<span>(.+?)</span>', webpage, 'expired', None)
  66             if expired:
  67                 raise ExtractorError('%s said: %s' % (self.IE_NAME, expired), expected=True)
  68             raise ExtractorError('Unable to extract video urls')
  69
  70         urls_info = self._parse_json(
  71             mobj.group('json_data'), video_id, transform_source=js_to_json)
  72
  73         if not isinstance(urls_info, list):
  74             urls_info = [urls_info]
  75
  76         if mobj.group('type') == 'YouTube':
  77             return self.playlist_result([
  78                 self.url_result(url_info['url']) for url_info in urls_info])
  79
  80         formats = [{
  81             'url': url_info['url'],
  82             'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none',
  83             'width': int_or_none(url_info.get('width')),
  84             'height': int_or_none(url_info.get('height')),
  85             'tbr': int_or_none(url_info.get('bitrate')),
  86             'filesize': int_or_none(url_info.get('filesize')),
  87         } for url_info in urls_info]
  88
  89         self._sort_formats(formats)
  90
  91         return {
  92             'id': video_id,
  93             'title': self._og_search_title(webpage),
  94             'formats': formats,
  95             'description': self._og_search_description(webpage),
  96             'thumbnail': self._og_search_thumbnail(webpage),
  97         }
  98
  99
 100 class ABCIViewIE(InfoExtractor):
 101     IE_NAME = 'abc.net.au:iview'
 102     _VALID_URL = r'https?://iview\.abc\.net\.au/programs/[^/]+/(?P<id>[^/?#]+)'
 103
 104     _TESTS = [{
 105         'url': 'http://iview.abc.net.au/programs/gardening-australia/FA1505V024S00',
 106         'md5': '979d10b2939101f0d27a06b79edad536',
 107         'info_dict': {
 108             'id': 'FA1505V024S00',
 109             'ext': 'mp4',
 110             'title': 'Series 27 Ep 24',
 111             'description': 'md5:b28baeae7504d1148e1d2f0e3ed3c15d',
 112             'upload_date': '20160820',
 113             'uploader_id': 'abc1',
 114             'timestamp': 1471719600,
 115         },
 116     }]
 117
 118     def _real_extract(self, url):
 119         video_id = self._match_id(url)
 120         webpage = self._download_webpage(url, video_id)
 121         video_params = self._parse_json(self._search_regex(
 122             r'videoParams\s*=\s*({.+?});', webpage, 'video params'), video_id)
 123         title = video_params['title']
 124         stream = next(s for s in video_params['playlist'] if s.get('type') == 'program')
 125
 126         formats = []
 127         f4m_url = stream.get('hds-unmetered') or stream['hds-metered']
 128         formats.extend(self._extract_f4m_formats(
 129             update_url_query(f4m_url, {'hdcore': '3.7.0'}),
 130             video_id, f4m_id='hds', fatal=False))
 131         formats.extend(self._extract_m3u8_formats(f4m_url.replace(
 132             'akamaihd.net/z/', 'akamaihd.net/i/').replace('/manifest.f4m', '/master.m3u8'),
 133             video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
 134         self._sort_formats(formats)
 135
 136         subtitles = {}
 137         src_vtt = stream.get('captions', {}).get('src-vtt')
 138         if src_vtt:
 139             subtitles['en'] = [{
 140                 'url': src_vtt,
 141                 'ext': 'vtt',
 142             }]
 143
 144         return {
 145             'id': video_id,
 146             'title': title,
 147             'description': self._html_search_meta(['og:description', 'twitter:description'], webpage),
 148             'thumbnail': self._html_search_meta(['og:image', 'twitter:image:src'], webpage),
 149             'duration': int_or_none(video_params.get('eventDuration')),
 150             'timestamp': parse_iso8601(video_params.get('pubDate'), ' '),
 151             'series': video_params.get('seriesTitle'),
 152             'series_id': video_params.get('seriesHouseNumber') or video_id[:7],
 153             'episode_number': int_or_none(self._html_search_meta('episodeNumber', webpage)),
 154             'episode': self._html_search_meta('episode_title', webpage),
 155             'uploader_id': video_params.get('channel'),
 156             'formats': formats,
 157             'subtitles': subtitles,
 158         }