commit 864a4576b70bfe9abc3c4f72b0b5e8173b686875
parent bbb7c3f7e92111d0d5060297db007ecb1047c2c8
Author: Sergey M․ <dstftw@gmail.com>
Date: Fri, 26 Jan 2018 23:49:47 +0700
[dplay] Add support for disco-api videos (closes #15396)
Diffstat:
1 file changed, 96 insertions(+), 4 deletions(-)
diff --git a/youtube_dl/extractor/dplay.py b/youtube_dl/extractor/dplay.py
@@ -12,25 +12,28 @@ from ..compat import (
compat_urlparse,
)
from ..utils import (
+ determine_ext,
ExtractorError,
+ float_or_none,
int_or_none,
remove_end,
try_get,
unified_strdate,
+ unified_timestamp,
update_url_query,
USER_AGENTS,
)
class DPlayIE(InfoExtractor):
- _VALID_URL = r'https?://(?P<domain>www\.dplay\.(?:dk|se|no))/[^/]+/(?P<id>[^/?#]+)'
+ _VALID_URL = r'https?://(?P<domain>www\.(?P<host>dplay\.(?:dk|se|no)))/(?:videoer/)?(?P<id>[^/]+/[^/?#]+)'
_TESTS = [{
# non geo restricted, via secure api, unsigned download hls URL
'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/',
'info_dict': {
'id': '3172',
- 'display_id': 'season-1-svensken-lar-sig-njuta-av-livet',
+ 'display_id': 'nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet',
'ext': 'mp4',
'title': 'Svensken lär sig njuta av livet',
'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8',
@@ -48,7 +51,7 @@ class DPlayIE(InfoExtractor):
'url': 'http://www.dplay.dk/mig-og-min-mor/season-6-episode-12/',
'info_dict': {
'id': '70816',
- 'display_id': 'season-6-episode-12',
+ 'display_id': 'mig-og-min-mor/season-6-episode-12',
'ext': 'mp4',
'title': 'Episode 12',
'description': 'md5:9c86e51a93f8a4401fc9641ef9894c90',
@@ -65,6 +68,26 @@ class DPlayIE(InfoExtractor):
# geo restricted, via direct unsigned hls URL
'url': 'http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/',
'only_matching': True,
+ }, {
+ # disco-api
+ 'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7',
+ 'info_dict': {
+ 'id': '40206',
+ 'display_id': 'i-kongens-klr/sesong-1-episode-7',
+ 'ext': 'mp4',
+ 'title': 'Episode 7',
+ 'description': 'md5:e3e1411b2b9aebeea36a6ec5d50c60cf',
+ 'duration': 2611.16,
+ 'timestamp': 1516726800,
+ 'upload_date': '20180123',
+ 'series': 'I kongens klær',
+ 'season_number': 1,
+ 'episode_number': 7,
+ },
+ 'params': {
+ 'format': 'bestvideo',
+ 'skip_download': True,
+ },
}]
def _real_extract(self, url):
@@ -75,7 +98,76 @@ class DPlayIE(InfoExtractor):
webpage = self._download_webpage(url, display_id)
video_id = self._search_regex(
- r'data-video-id=["\'](\d+)', webpage, 'video id')
+ r'data-video-id=["\'](\d+)', webpage, 'video id', default=None)
+
+ if not video_id:
+ host = mobj.group('host')
+ disco_base = 'https://disco-api.%s' % host
+ self._download_json(
+ '%s/token' % disco_base, display_id, 'Downloading token',
+ query={
+ 'realm': host.replace('.', ''),
+ })
+ video = self._download_json(
+ '%s/content/videos/%s' % (disco_base, display_id), display_id,
+ headers={
+ 'Referer': url,
+ 'x-disco-client': 'WEB:UNKNOWN:dplay-client:0.0.1',
+ }, query={
+ 'include': 'show'
+ })
+ video_id = video['data']['id']
+ info = video['data']['attributes']
+ title = info['name']
+ formats = []
+ for format_id, format_dict in self._download_json(
+ '%s/playback/videoPlaybackInfo/%s' % (disco_base, video_id),
+ display_id)['data']['attributes']['streaming'].items():
+ if not isinstance(format_dict, dict):
+ continue
+ format_url = format_dict.get('url')
+ if not format_url:
+ continue
+ ext = determine_ext(format_url)
+ if format_id == 'dash' or ext == 'mpd':
+ formats.extend(self._extract_mpd_formats(
+ format_url, display_id, mpd_id='dash', fatal=False))
+ elif format_id == 'hls' or ext == 'm3u8':
+ formats.extend(self._extract_m3u8_formats(
+ format_url, display_id, 'mp4',
+ entry_protocol='m3u8_native', m3u8_id='hls',
+ fatal=False))
+ else:
+ formats.append({
+ 'url': format_url,
+ 'format_id': format_id,
+ })
+ self._sort_formats(formats)
+
+ series = None
+ try:
+ included = video.get('included')
+ if isinstance(included, list):
+ show = next(e for e in included if e.get('type') == 'show')
+ series = try_get(
+ show, lambda x: x['attributes']['name'], compat_str)
+ except StopIteration:
+ pass
+
+ return {
+ 'id': video_id,
+ 'display_id': display_id,
+ 'title': title,
+ 'description': info.get('description'),
+ 'duration': float_or_none(
+ info.get('videoDuration'), scale=1000),
+ 'timestamp': unified_timestamp(info.get('publishStart')),
+ 'series': series,
+ 'season_number': int_or_none(info.get('seasonNumber')),
+ 'episode_number': int_or_none(info.get('episodeNumber')),
+ 'age_limit': int_or_none(info.get('minimum_age')),
+ 'formats': formats,
+ }
info = self._download_json(
'http://%s/api/v2/ajax/videos?video_id=%s' % (domain, video_id),
