mirror of
https://github.com/l1ving/youtube-dl
synced 2024-12-22 12:07:32 +08:00
46 lines
1.8 KiB
Python
46 lines
1.8 KiB
Python
import re
|
|
|
|
from .common import InfoExtractor
|
|
from ..utils import (
|
|
ExtractorError,
|
|
)
|
|
|
|
class ARDIE(InfoExtractor):
|
|
_VALID_URL = r'^(?:https?://)?(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
|
|
_TITLE = r'<h1(?: class="boxTopHeadline")?>(?P<title>.*)</h1>'
|
|
_MEDIA_STREAM = r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)'
|
|
|
|
def _real_extract(self, url):
|
|
# determine video id from url
|
|
m = re.match(self._VALID_URL, url)
|
|
|
|
numid = re.search(r'documentId=([0-9]+)', url)
|
|
if numid:
|
|
video_id = numid.group(1)
|
|
else:
|
|
video_id = m.group('video_id')
|
|
|
|
# determine title and media streams from webpage
|
|
html = self._download_webpage(url, video_id)
|
|
title = re.search(self._TITLE, html).group('title')
|
|
streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
|
|
if not streams:
|
|
assert '"fsk"' in html
|
|
raise ExtractorError(u'This video is only available after 8:00 pm')
|
|
|
|
# choose default media type and highest quality for now
|
|
stream = max([s for s in streams if int(s["media_type"]) == 0],
|
|
key=lambda s: int(s["quality"]))
|
|
|
|
# there's two possibilities: RTMP stream or HTTP download
|
|
info = {'id': video_id, 'title': title, 'ext': 'mp4'}
|
|
if stream['rtmp_url']:
|
|
self.to_screen(u'RTMP download detected')
|
|
assert stream['video_url'].startswith('mp4:')
|
|
info["url"] = stream["rtmp_url"]
|
|
info["play_path"] = stream['video_url']
|
|
else:
|
|
assert stream["video_url"].endswith('.mp4')
|
|
info["url"] = stream["video_url"]
|
|
return [info]
|