youtube-dl

Another place where youtube-dl lives on
git clone git://git.oshgnacknak.de/youtube-dl.git
Log | Files | Refs | README | LICENSE

aol.py (5648B)


      1 # coding: utf-8
      2 from __future__ import unicode_literals
      3 
      4 import re
      5 
      6 from .yahoo import YahooIE
      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(YahooIE):
     19     IE_NAME = 'aol.com'
     20     _VALID_URL = r'(?:aol-video:|https?://(?:www\.)?aol\.(?:com|ca|co\.uk|de|jp)/video/(?:[^/]+/)*)(?P<id>\d{9}|[0-9a-f]{24}|[0-9a-f]{8}-(?:[0-9a-f]{4}-){3}[0-9a-f]{12})'
     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         'url': 'https://www.aol.ca/video/view/u-s-woman-s-family-arrested-for-murder-first-pinned-on-panhandler-police/5c7ccf45bc03931fa04b2fe1/',
     69         'only_matching': True,
     70     }, {
     71         'url': 'https://www.aol.co.uk/video/view/-one-dead-and-22-hurt-in-bus-crash-/5cb3a6f3d21f1a072b457347/',
     72         'only_matching': True,
     73     }, {
     74         'url': 'https://www.aol.de/video/view/eva-braun-privataufnahmen-von-hitlers-geliebter-werden-digitalisiert/5cb2d49de98ab54c113d3d5d/',
     75         'only_matching': True,
     76     }, {
     77         'url': 'https://www.aol.jp/video/playlist/5a28e936a1334d000137da0c/5a28f3151e642219fde19831/',
     78         'only_matching': True,
     79     }, {
     80         # Yahoo video
     81         'url': 'https://www.aol.com/video/play/991e6700-ac02-11ea-99ff-357400036f61/24bbc846-3e30-3c46-915e-fe8ccd7fcc46/',
     82         'only_matching': True,
     83     }]
     84 
     85     def _real_extract(self, url):
     86         video_id = self._match_id(url)
     87         if '-' in video_id:
     88             return self._extract_yahoo_video(video_id, 'us')
     89 
     90         response = self._download_json(
     91             'https://feedapi.b2c.on.aol.com/v1.0/app/videos/aolon/%s/details' % video_id,
     92             video_id)['response']
     93         if response['statusText'] != 'Ok':
     94             raise ExtractorError('%s said: %s' % (self.IE_NAME, response['statusText']), expected=True)
     95 
     96         video_data = response['data']
     97         formats = []
     98         m3u8_url = url_or_none(video_data.get('videoMasterPlaylist'))
     99         if m3u8_url:
    100             formats.extend(self._extract_m3u8_formats(
    101                 m3u8_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
    102         for rendition in video_data.get('renditions', []):
    103             video_url = url_or_none(rendition.get('url'))
    104             if not video_url:
    105                 continue
    106             ext = rendition.get('format')
    107             if ext == 'm3u8':
    108                 formats.extend(self._extract_m3u8_formats(
    109                     video_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
    110             else:
    111                 f = {
    112                     'url': video_url,
    113                     'format_id': rendition.get('quality'),
    114                 }
    115                 mobj = re.search(r'(\d+)x(\d+)', video_url)
    116                 if mobj:
    117                     f.update({
    118                         'width': int(mobj.group(1)),
    119                         'height': int(mobj.group(2)),
    120                     })
    121                 else:
    122                     qs = compat_parse_qs(compat_urllib_parse_urlparse(video_url).query)
    123                     f.update({
    124                         'width': int_or_none(qs.get('w', [None])[0]),
    125                         'height': int_or_none(qs.get('h', [None])[0]),
    126                     })
    127                 formats.append(f)
    128         self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
    129 
    130         return {
    131             'id': video_id,
    132             'title': video_data['title'],
    133             'duration': int_or_none(video_data.get('duration')),
    134             'timestamp': int_or_none(video_data.get('publishDate')),
    135             'view_count': int_or_none(video_data.get('views')),
    136             'description': video_data.get('description'),
    137             'uploader': video_data.get('videoOwner'),
    138             'formats': formats,
    139         }