yt-dlp/yt_dlp/extractor/dctp.py

102 lines
3.4 KiB
Python
Raw Normal View History

2015-01-28 08:21:04 +01:00
from .common import InfoExtractor
from ..utils import (
float_or_none,
int_or_none,
unified_timestamp,
2018-07-21 14:08:28 +02:00
url_or_none,
)
2015-01-28 08:21:04 +01:00
2015-01-28 08:59:58 +01:00
2015-01-28 08:21:04 +01:00
class DctpTvIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?dctp\.tv/(?:#/)?filme/(?P<id>[^/?#&]+)'
_TESTS = [{
# 4x3
2015-01-28 08:59:58 +01:00
'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',
'md5': '3ffbd1556c3fe210724d7088fad723e3',
2015-01-28 08:59:58 +01:00
'info_dict': {
2016-10-01 09:22:48 +02:00
'id': '95eaa4f33dad413aa17b4ee613cccc6c',
2015-01-30 00:35:53 +01:00
'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',
'ext': 'm4v',
2016-10-01 09:22:48 +02:00
'title': 'Videoinstallation für eine Kaufhausfassade',
'description': 'Kurzfilm',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 71.24,
'timestamp': 1302172322,
'upload_date': '20110407',
},
}, {
# 16x9
'url': 'http://www.dctp.tv/filme/sind-youtuber-die-besseren-lehrer/',
'only_matching': True,
}]
_BASE_URL = 'http://dctp-ivms2-restapi.s3.amazonaws.com'
2015-01-28 08:21:04 +01:00
def _real_extract(self, url):
display_id = self._match_id(url)
2016-10-01 09:22:48 +02:00
version = self._download_json(
f'{self._BASE_URL}/version.json', display_id,
'Downloading version JSON')
restapi_base = '{}/{}/restapi'.format(
self._BASE_URL, version['version_name'])
2015-01-28 08:21:04 +01:00
info = self._download_json(
f'{restapi_base}/slugs/{display_id}.json', display_id,
'Downloading video info JSON')
2016-10-01 09:22:48 +02:00
media = self._download_json(
'{}/media/{}.json'.format(restapi_base, str(info['object_id'])),
display_id, 'Downloading media JSON')
uuid = media['uuid']
title = media['title']
is_wide = media.get('is_wide')
formats = []
def add_formats(suffix):
templ = f'https://%s/{uuid}_dctp_{suffix}.m4v'
formats.extend([{
'format_id': 'hls-' + suffix,
'url': templ % 'cdn-segments.dctp.tv' + '/playlist.m3u8',
'protocol': 'm3u8_native',
}, {
'format_id': 's3-' + suffix,
'url': templ % 'completed-media.s3.amazonaws.com',
}, {
'format_id': 'http-' + suffix,
'url': templ % 'cdn-media.dctp.tv',
}])
add_formats('0500_' + ('16x9' if is_wide else '4x3'))
if is_wide:
add_formats('720p')
thumbnails = []
images = media.get('images')
if isinstance(images, list):
for image in images:
if not isinstance(image, dict):
continue
2018-07-21 14:08:28 +02:00
image_url = url_or_none(image.get('url'))
if not image_url:
continue
thumbnails.append({
'url': image_url,
'width': int_or_none(image.get('width')),
'height': int_or_none(image.get('height')),
})
2015-01-28 08:21:04 +01:00
return {
'id': uuid,
'display_id': display_id,
2015-01-28 08:21:04 +01:00
'title': title,
'alt_title': media.get('subtitle'),
'description': media.get('description') or media.get('teaser'),
'timestamp': unified_timestamp(media.get('created')),
'duration': float_or_none(media.get('duration_in_ms'), scale=1000),
'thumbnails': thumbnails,
2016-10-01 09:22:48 +02:00
'formats': formats,
2015-01-28 08:21:04 +01:00
}