youtube_dl/extractor/aol.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_parse_qs,
   9     compat_urllib_parse_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     url_or_none,
  15 )
  16
  17
  18 class AolIE(InfoExtractor):
  19     IE_NAME = 'aol.com'
  20     _VALID_URL = r'(?:aol-video:|https?://(?:www\.)?aol\.com/video/(?:[^/]+/)*)(?P<id>[0-9a-f]+)'
  21
  22     _TESTS = [{
  23         # video with 5min ID
  24         'url': 'https://www.aol.com/video/view/u-s--official-warns-of-largest-ever-irs-phone-scam/518167793/',
  25         'md5': '18ef68f48740e86ae94b98da815eec42',
  26         'info_dict': {
  27             'id': '518167793',
  28             'ext': 'mp4',
  29             'title': 'U.S. Official Warns Of \'Largest Ever\' IRS Phone Scam',
  30             'description': 'A major phone scam has cost thousands of taxpayers more than $1 million, with less than a month until income tax returns are due to the IRS.',
  31             'timestamp': 1395405060,
  32             'upload_date': '20140321',
  33             'uploader': 'Newsy Studio',
  34         },
  35         'params': {
  36             # m3u8 download
  37             'skip_download': True,
  38         }
  39     }, {
  40         # video with vidible ID
  41         'url': 'https://www.aol.com/video/view/netflix-is-raising-rates/5707d6b8e4b090497b04f706/',
  42         'info_dict': {
  43             'id': '5707d6b8e4b090497b04f706',
  44             'ext': 'mp4',
  45             'title': 'Netflix is Raising Rates',
  46             'description': 'Netflix is rewarding millions of it’s long-standing members with an increase in cost. Veuer’s Carly Figueroa has more.',
  47             'upload_date': '20160408',
  48             'timestamp': 1460123280,
  49             'uploader': 'Veuer',
  50         },
  51         'params': {
  52             # m3u8 download
  53             'skip_download': True,
  54         }
  55     }, {
  56         'url': 'https://www.aol.com/video/view/park-bench-season-2-trailer/559a1b9be4b0c3bfad3357a7/',
  57         'only_matching': True,
  58     }, {
  59         'url': 'https://www.aol.com/video/view/donald-trump-spokeswoman-tones-down-megyn-kelly-attacks/519442220/',
  60         'only_matching': True,
  61     }, {
  62         'url': 'aol-video:5707d6b8e4b090497b04f706',
  63         'only_matching': True,
  64     }, {
  65         'url': 'https://www.aol.com/video/playlist/PL8245/5ca79d19d21f1a04035db606/',
  66         'only_matching': True,
  67     }]
  68
  69     def _real_extract(self, url):
  70         video_id = self._match_id(url)
  71
  72         response = self._download_json(
  73             'https://feedapi.b2c.on.aol.com/v1.0/app/videos/aolon/%s/details' % video_id,
  74             video_id)['response']
  75         if response['statusText'] != 'Ok':
  76             raise ExtractorError('%s said: %s' % (self.IE_NAME, response['statusText']), expected=True)
  77
  78         video_data = response['data']
  79         formats = []
  80         m3u8_url = url_or_none(video_data.get('videoMasterPlaylist'))
  81         if m3u8_url:
  82             formats.extend(self._extract_m3u8_formats(
  83                 m3u8_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
  84         for rendition in video_data.get('renditions', []):
  85             video_url = url_or_none(rendition.get('url'))
  86             if not video_url:
  87                 continue
  88             ext = rendition.get('format')
  89             if ext == 'm3u8':
  90                 formats.extend(self._extract_m3u8_formats(
  91                     video_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
  92             else:
  93                 f = {
  94                     'url': video_url,
  95                     'format_id': rendition.get('quality'),
  96                 }
  97                 mobj = re.search(r'(\d+)x(\d+)', video_url)
  98                 if mobj:
  99                     f.update({
 100                         'width': int(mobj.group(1)),
 101                         'height': int(mobj.group(2)),
 102                     })
 103                 else:
 104                     qs = compat_parse_qs(compat_urllib_parse_urlparse(video_url).query)
 105                     f.update({
 106                         'width': int_or_none(qs.get('w', [None])[0]),
 107                         'height': int_or_none(qs.get('h', [None])[0]),
 108                     })
 109                 formats.append(f)
 110         self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
 111
 112         return {
 113             'id': video_id,
 114             'title': video_data['title'],
 115             'duration': int_or_none(video_data.get('duration')),
 116             'timestamp': int_or_none(video_data.get('publishDate')),
 117             'view_count': int_or_none(video_data.get('views')),
 118             'description': video_data.get('description'),
 119             'uploader': video_data.get('videoOwner'),
 120             'formats': formats,
 121         }