Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/abc.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     ExtractorError,
   9     js_to_json,
  10     int_or_none,
  11     parse_iso8601,
  12     try_get,
  13 )
  14
  15
  16 class ABCIE(InfoExtractor):
  17     IE_NAME = 'abc.net.au'
  18     _VALID_URL = r'https?://(?:www\.)?abc\.net\.au/news/(?:[^/]+/){1,2}(?P<id>\d+)'
  19
  20     _TESTS = [{
  21         'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334',
  22         'md5': 'cb3dd03b18455a661071ee1e28344d9f',
  23         'info_dict': {
  24             'id': '5868334',
  25             'ext': 'mp4',
  26             'title': 'Australia to help staff Ebola treatment centre in Sierra Leone',
  27             'description': 'md5:809ad29c67a05f54eb41f2a105693a67',
  28         },
  29         'skip': 'this video has expired',
  30     }, {
  31         'url': 'http://www.abc.net.au/news/2015-08-17/warren-entsch-introduces-same-sex-marriage-bill/6702326',
  32         'md5': 'db2a5369238b51f9811ad815b69dc086',
  33         'info_dict': {
  34             'id': 'NvqvPeNZsHU',
  35             'ext': 'mp4',
  36             'upload_date': '20150816',
  37             'uploader': 'ABC News (Australia)',
  38             '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',
  39             'uploader_id': 'NewsOnABC',
  40             'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill',
  41         },
  42         'add_ie': ['Youtube'],
  43         'skip': 'Not accessible from Travis CI server',
  44     }, {
  45         'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080',
  46         'md5': 'b96eee7c9edf4fc5a358a0252881cc1f',
  47         'info_dict': {
  48             'id': '6880080',
  49             'ext': 'mp3',
  50             'title': 'NAB lifts interest rates, following Westpac and CBA',
  51             'description': 'md5:f13d8edc81e462fce4a0437c7dc04728',
  52         },
  53     }, {
  54         'url': 'http://www.abc.net.au/news/2015-10-19/6866214',
  55         'only_matching': True,
  56     }]
  57
  58     def _real_extract(self, url):
  59         video_id = self._match_id(url)
  60         webpage = self._download_webpage(url, video_id)
  61
  62         mobj = re.search(
  63             r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
  64             webpage)
  65         if mobj is None:
  66             expired = self._html_search_regex(r'(?s)class="expired-(?:video|audio)".+?<span>(.+?)</span>', webpage, 'expired', None)
  67             if expired:
  68                 raise ExtractorError('%s said: %s' % (self.IE_NAME, expired), expected=True)
  69             raise ExtractorError('Unable to extract video urls')
  70
  71         urls_info = self._parse_json(
  72             mobj.group('json_data'), video_id, transform_source=js_to_json)
  73
  74         if not isinstance(urls_info, list):
  75             urls_info = [urls_info]
  76
  77         if mobj.group('type') == 'YouTube':
  78             return self.playlist_result([
  79                 self.url_result(url_info['url']) for url_info in urls_info])
  80
  81         formats = [{
  82             'url': url_info['url'],
  83             'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none',
  84             'width': int_or_none(url_info.get('width')),
  85             'height': int_or_none(url_info.get('height')),
  86             'tbr': int_or_none(url_info.get('bitrate')),
  87             'filesize': int_or_none(url_info.get('filesize')),
  88         } for url_info in urls_info]
  89
  90         self._sort_formats(formats)
  91
  92         return {
  93             'id': video_id,
  94             'title': self._og_search_title(webpage),
  95             'formats': formats,
  96             'description': self._og_search_description(webpage),
  97             'thumbnail': self._og_search_thumbnail(webpage),
  98         }
  99
 100
 101 class ABCIViewIE(InfoExtractor):
 102     IE_NAME = 'abc.net.au:iview'
 103     _VALID_URL = r'https?://iview\.abc\.net\.au/programs/[^/]+/(?P<id>[^/?#]+)'
 104
 105     # ABC iview programs are normally available for 14 days only.
 106     _TESTS = [{
 107         'url': 'http://iview.abc.net.au/programs/diaries-of-a-broken-mind/ZX9735A001S00',
 108         'md5': 'cde42d728b3b7c2b32b1b94b4a548afc',
 109         'info_dict': {
 110             'id': 'ZX9735A001S00',
 111             'ext': 'mp4',
 112             'title': 'Diaries Of A Broken Mind',
 113             'description': 'md5:7de3903874b7a1be279fe6b68718fc9e',
 114             'upload_date': '20161010',
 115             'uploader_id': 'abc2',
 116             'timestamp': 1476064920,
 117         },
 118         'skip': 'Video gone',
 119     }]
 120
 121     def _real_extract(self, url):
 122         video_id = self._match_id(url)
 123         webpage = self._download_webpage(url, video_id)
 124         video_params = self._parse_json(self._search_regex(
 125             r'videoParams\s*=\s*({.+?});', webpage, 'video params'), video_id)
 126         title = video_params.get('title') or video_params['seriesTitle']
 127         stream = next(s for s in video_params['playlist'] if s.get('type') == 'program')
 128
 129         format_urls = [
 130             try_get(stream, lambda x: x['hds-unmetered'], compat_str)]
 131
 132         # May have higher quality video
 133         sd_url = try_get(
 134             stream, lambda x: x['streams']['hds']['sd'], compat_str)
 135         if sd_url:
 136             format_urls.append(sd_url.replace('metered', 'um'))
 137
 138         formats = []
 139         for format_url in format_urls:
 140             if format_url:
 141                 formats.extend(
 142                     self._extract_akamai_formats(format_url, video_id))
 143         self._sort_formats(formats)
 144
 145         subtitles = {}
 146         src_vtt = stream.get('captions', {}).get('src-vtt')
 147         if src_vtt:
 148             subtitles['en'] = [{
 149                 'url': src_vtt,
 150                 'ext': 'vtt',
 151             }]
 152
 153         return {
 154             'id': video_id,
 155             'title': title,
 156             'description': self._html_search_meta(['og:description', 'twitter:description'], webpage),
 157             'thumbnail': self._html_search_meta(['og:image', 'twitter:image:src'], webpage),
 158             'duration': int_or_none(video_params.get('eventDuration')),
 159             'timestamp': parse_iso8601(video_params.get('pubDate'), ' '),
 160             'series': video_params.get('seriesTitle'),
 161             'series_id': video_params.get('seriesHouseNumber') or video_id[:7],
 162             'episode_number': int_or_none(self._html_search_meta('episodeNumber', webpage, default=None)),
 163             'episode': self._html_search_meta('episode_title', webpage, default=None),
 164             'uploader_id': video_params.get('channel'),
 165             'formats': formats,
 166             'subtitles': subtitles,
 167         }