]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/amp.py
[pornhub] Improve locked videos detection (closes #22449, closes #22780)
[youtube-dl] / youtube_dl / extractor / amp.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..utils import (
6     determine_ext,
7     ExtractorError,
8     int_or_none,
9     mimetype2ext,
10     parse_iso8601,
11     url_or_none,
12 )
13
14
15 class AMPIE(InfoExtractor):
16     # parse Akamai Adaptive Media Player feed
17     def _extract_feed_info(self, url):
18         feed = self._download_json(
19             url, None, 'Downloading Akamai AMP feed',
20             'Unable to download Akamai AMP feed')
21         item = feed.get('channel', {}).get('item')
22         if not item:
23             raise ExtractorError('%s said: %s' % (self.IE_NAME, feed['error']))
24
25         video_id = item['guid']
26
27         def get_media_node(name, default=None):
28             media_name = 'media-%s' % name
29             media_group = item.get('media-group') or item
30             return media_group.get(media_name) or item.get(media_name) or item.get(name, default)
31
32         thumbnails = []
33         media_thumbnail = get_media_node('thumbnail')
34         if media_thumbnail:
35             if isinstance(media_thumbnail, dict):
36                 media_thumbnail = [media_thumbnail]
37             for thumbnail_data in media_thumbnail:
38                 thumbnail = thumbnail_data.get('@attributes', {})
39                 thumbnail_url = url_or_none(thumbnail.get('url'))
40                 if not thumbnail_url:
41                     continue
42                 thumbnails.append({
43                     'url': self._proto_relative_url(thumbnail_url, 'http:'),
44                     'width': int_or_none(thumbnail.get('width')),
45                     'height': int_or_none(thumbnail.get('height')),
46                 })
47
48         subtitles = {}
49         media_subtitle = get_media_node('subTitle')
50         if media_subtitle:
51             if isinstance(media_subtitle, dict):
52                 media_subtitle = [media_subtitle]
53             for subtitle_data in media_subtitle:
54                 subtitle = subtitle_data.get('@attributes', {})
55                 subtitle_href = url_or_none(subtitle.get('href'))
56                 if not subtitle_href:
57                     continue
58                 subtitles.setdefault(subtitle.get('lang') or 'en', []).append({
59                     'url': subtitle_href,
60                     'ext': mimetype2ext(subtitle.get('type')) or determine_ext(subtitle_href),
61                 })
62
63         formats = []
64         media_content = get_media_node('content')
65         if isinstance(media_content, dict):
66             media_content = [media_content]
67         for media_data in media_content:
68             media = media_data.get('@attributes', {})
69             media_url = url_or_none(media.get('url'))
70             if not media_url:
71                 continue
72             ext = mimetype2ext(media.get('type')) or determine_ext(media_url)
73             if ext == 'f4m':
74                 formats.extend(self._extract_f4m_formats(
75                     media_url + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
76                     video_id, f4m_id='hds', fatal=False))
77             elif ext == 'm3u8':
78                 formats.extend(self._extract_m3u8_formats(
79                     media_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
80             else:
81                 formats.append({
82                     'format_id': media_data.get('media-category', {}).get('@attributes', {}).get('label'),
83                     'url': media_url,
84                     'tbr': int_or_none(media.get('bitrate')),
85                     'filesize': int_or_none(media.get('fileSize')),
86                     'ext': ext,
87                 })
88
89         self._sort_formats(formats)
90
91         timestamp = parse_iso8601(item.get('pubDate'), ' ') or parse_iso8601(item.get('dc-date'))
92
93         return {
94             'id': video_id,
95             'title': get_media_node('title'),
96             'description': get_media_node('description'),
97             'thumbnails': thumbnails,
98             'timestamp': timestamp,
99             'duration': int_or_none(media_content[0].get('@attributes', {}).get('duration')),
100             'subtitles': subtitles,
101             'formats': formats,
102         }