]> jfr.im git - yt-dlp.git/blame - youtube_dl/extractor/amp.py
[youtube] Make --write-annotations non fatal (closes #21452)
[yt-dlp.git] / youtube_dl / extractor / amp.py
CommitLineData
3793090b 1# coding: utf-8
2from __future__ import unicode_literals
3
4from .common import InfoExtractor
5from ..utils import (
ab49d7a9 6 determine_ext,
8abc7dca 7 ExtractorError,
3052a30d
S
8 int_or_none,
9 mimetype2ext,
10 parse_iso8601,
11 url_or_none,
3793090b 12)
13
14
15class AMPIE(InfoExtractor):
3793090b 16 # parse Akamai Adaptive Media Player feed
17 def _extract_feed_info(self, url):
8abc7dca 18 feed = self._download_json(
c7fa5fa4 19 url, None, 'Downloading Akamai AMP feed',
8abc7dca
RA
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']))
3793090b 24
25 video_id = item['guid']
c7fa5fa4 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
3793090b 32 thumbnails = []
c7fa5fa4 33 media_thumbnail = get_media_node('thumbnail')
3793090b 34 if media_thumbnail:
35 if isinstance(media_thumbnail, dict):
36 media_thumbnail = [media_thumbnail]
37 for thumbnail_data in media_thumbnail:
7fc87519 38 thumbnail = thumbnail_data.get('@attributes', {})
3052a30d 39 thumbnail_url = url_or_none(thumbnail.get('url'))
7fc87519
RA
40 if not thumbnail_url:
41 continue
3793090b 42 thumbnails.append({
7fc87519 43 'url': self._proto_relative_url(thumbnail_url, 'http:'),
3793090b 44 'width': int_or_none(thumbnail.get('width')),
45 'height': int_or_none(thumbnail.get('height')),
46 })
47
48 subtitles = {}
c7fa5fa4 49 media_subtitle = get_media_node('subTitle')
3793090b 50 if media_subtitle:
51 if isinstance(media_subtitle, dict):
52 media_subtitle = [media_subtitle]
53 for subtitle_data in media_subtitle:
7fc87519 54 subtitle = subtitle_data.get('@attributes', {})
3052a30d 55 subtitle_href = url_or_none(subtitle.get('href'))
7fc87519
RA
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 })
3793090b 62
63 formats = []
c7fa5fa4 64 media_content = get_media_node('content')
3793090b 65 if isinstance(media_content, dict):
66 media_content = [media_content]
67 for media_data in media_content:
ab49d7a9 68 media = media_data.get('@attributes', {})
3052a30d 69 media_url = url_or_none(media.get('url'))
ab49d7a9
RA
70 if not media_url:
71 continue
54b0e909 72 ext = mimetype2ext(media.get('type')) or determine_ext(media_url)
ab49d7a9 73 if ext == 'f4m':
7e5edcfd 74 formats.extend(self._extract_f4m_formats(
ab49d7a9 75 media_url + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
7e5edcfd 76 video_id, f4m_id='hds', fatal=False))
ab49d7a9 77 elif ext == 'm3u8':
7e5edcfd 78 formats.extend(self._extract_m3u8_formats(
ab49d7a9 79 media_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
3793090b 80 else:
81 formats.append({
055f0d3d 82 'format_id': media_data.get('media-category', {}).get('@attributes', {}).get('label'),
3052a30d 83 'url': media_url,
c7fa5fa4 84 'tbr': int_or_none(media.get('bitrate')),
3793090b 85 'filesize': int_or_none(media.get('fileSize')),
ab49d7a9 86 'ext': ext,
3793090b 87 })
88
89 self._sort_formats(formats)
90
87d105ac
S
91 timestamp = parse_iso8601(item.get('pubDate'), ' ') or parse_iso8601(item.get('dc-date'))
92
3793090b 93 return {
94 'id': video_id,
c7fa5fa4 95 'title': get_media_node('title'),
96 'description': get_media_node('description'),
3793090b 97 'thumbnails': thumbnails,
87d105ac 98 'timestamp': timestamp,
3793090b 99 'duration': int_or_none(media_content[0].get('@attributes', {}).get('duration')),
fff79f18 100 'subtitles': subtitles,
3793090b 101 'formats': formats,
102 }