Merge 638ec38eb5
into 4d05f84325
This commit is contained in:
commit
7a4751372a
@ -645,6 +645,7 @@ from .livestream import (
|
||||
)
|
||||
from .lnkgo import LnkGoIE
|
||||
from .localnews8 import LocalNews8IE
|
||||
from .videocdn import VideoCdnIE
|
||||
from .lovehomeporn import LoveHomePornIE
|
||||
from .lrt import LRTIE
|
||||
from .lynda import (
|
||||
|
62
youtube_dl/extractor/videocdn.py
Normal file
62
youtube_dl/extractor/videocdn.py
Normal file
@ -0,0 +1,62 @@
|
||||
# coding: utf-8
|
||||
from __future__ import unicode_literals
|
||||
|
||||
from .common import InfoExtractor
|
||||
|
||||
from ..utils import (
|
||||
determine_ext,
|
||||
urljoin,
|
||||
)
|
||||
|
||||
|
||||
class VideoCdnIE(InfoExtractor):
|
||||
_VALID_URL = r'https?://e\.video-cdn\.net/video?.*video-id=(?P<id>[a-zA-Z0-9-_]+).*'
|
||||
_TESTS = [
|
||||
{
|
||||
'url': 'https://e.video-cdn.net/video?video-id=8eBUrWaMJFS38A5X-j2CgY&player-id=53Tun3ZZpZpVuvaTvsm3jU',
|
||||
'info_dict': {
|
||||
'id': '8eBUrWaMJFS38A5X-j2CgY',
|
||||
'ext': 'mp4',
|
||||
'title': 'RiskBuster FireFighter VI - Adventskranz',
|
||||
'thumbnail': r're:(?i)https://.*\.jpeg',
|
||||
},
|
||||
},
|
||||
{
|
||||
'url': 'https://e.video-cdn.net/video?video-id=91imQ_wKjkTFghe-3mmBAA&player-id=7nCLZ_ESM8rT9YUw6qUGA9',
|
||||
'info_dict': {
|
||||
'id': '91imQ_wKjkTFghe-3mmBAA',
|
||||
'ext': 'mp4',
|
||||
'title': 'SCC2019_Talk_Tychsen_TXL.mp4',
|
||||
'thumbnail': r're:(?i)https://.*\.jpeg',
|
||||
},
|
||||
},
|
||||
]
|
||||
|
||||
def _real_extract(self, url):
|
||||
video_id = self._match_id(url)
|
||||
|
||||
webpage = self._download_webpage(url, video_id)
|
||||
|
||||
thumbnail = self._search_regex(
|
||||
r'\"thumbnailUrl\":\"(?P<thumbnail>[^\"]+)',
|
||||
webpage, 'thumbnail', group='thumbnail',
|
||||
default=None)
|
||||
|
||||
title = self._search_regex(r'"name"\s*:\s*"((?:\\"|[^"])+)', webpage, 'title')
|
||||
|
||||
manifest_url = self._search_regex(r'"contentUrl"\s*:\s*"((?:\\"|[^"])+)', webpage, 'manifest_url')
|
||||
manifest_url = urljoin(url, manifest_url)
|
||||
|
||||
formats = []
|
||||
if manifest_url and determine_ext(manifest_url) == 'm3u8':
|
||||
formats.extend(self._extract_m3u8_formats(
|
||||
manifest_url, video_id, 'mp4',
|
||||
entry_protocol='m3u8_native', m3u8_id='m3u8'))
|
||||
self._sort_formats(formats)
|
||||
|
||||
return {
|
||||
'id': video_id,
|
||||
'title': title,
|
||||
'thumbnail': thumbnail,
|
||||
'formats': formats,
|
||||
}
|
Loading…
Reference in New Issue
Block a user