mirror of
https://github.com/l1ving/youtube-dl
synced 2025-02-15 07:43:27 +08:00
552 lines
21 KiB
Python
552 lines
21 KiB
Python
# coding: utf-8
|
|
from __future__ import unicode_literals
|
|
|
|
import re
|
|
|
|
from .common import InfoExtractor
|
|
from .generic import GenericIE
|
|
from ..utils import (
|
|
determine_ext,
|
|
ExtractorError,
|
|
int_or_none,
|
|
parse_duration,
|
|
qualities,
|
|
str_or_none,
|
|
try_get,
|
|
unified_strdate,
|
|
unified_timestamp,
|
|
update_url_query,
|
|
url_or_none,
|
|
xpath_text,
|
|
)
|
|
from ..compat import compat_etree_fromstring
|
|
|
|
|
|
class ARDBaseIE(InfoExtractor):
|
|
# Format information is valid for all media saved on ARD's Akamai servers.
|
|
# IMPORTANT: Not valid for media from WDR/BR/etc. servers
|
|
|
|
_ALL_FORMATS = {
|
|
'vcodec': 'H.264',
|
|
'acodec': 'LC-AAC',
|
|
}
|
|
|
|
_FORMATS = {
|
|
320: {
|
|
'width': 320,
|
|
'height': 188,
|
|
'vbr': 128,
|
|
'abr': 61,
|
|
},
|
|
480: {
|
|
'width': 480,
|
|
'height': 270,
|
|
'vbr': 256,
|
|
'abr': 61,
|
|
},
|
|
512: {
|
|
'width': 512,
|
|
'height': 288,
|
|
'vbr': 512,
|
|
'abr': 94,
|
|
},
|
|
640: {
|
|
'width': 640,
|
|
'height': 360,
|
|
'vbr': 1024,
|
|
'abr': 192,
|
|
},
|
|
960: {
|
|
'width': 960,
|
|
'height': 540,
|
|
'vbr': 1800,
|
|
'abr': 192,
|
|
},
|
|
1280: {
|
|
'width': 1280,
|
|
'height': 720,
|
|
'vbr': 3584,
|
|
'abr': 192,
|
|
},
|
|
}
|
|
|
|
def _check_additional_formats(self, formats, video_id):
|
|
urls = {}
|
|
for format in formats:
|
|
url = format['url']
|
|
if not url.endswith('.mp4'):
|
|
continue
|
|
base_url = url.rsplit('/', 1)[0]
|
|
m = re.search(r'.*/([0-9]+)-[0-9]\..*$', url)
|
|
if not m:
|
|
continue
|
|
width = int_or_none(m.group(1))
|
|
if base_url in urls and width in urls[base_url]:
|
|
continue
|
|
if base_url not in urls:
|
|
urls[base_url] = [width]
|
|
elif width not in urls[base_url]:
|
|
urls[base_url].append(width)
|
|
|
|
for base_url, ignore in urls.items():
|
|
for width in (x for x in self._FORMATS if x not in ignore):
|
|
url = '%s/%s-1.mp4' % (base_url, width)
|
|
if self._is_valid_url(url, video_id):
|
|
format_info = self._FORMATS[width]
|
|
format_id = '%s-%s' % (determine_ext(url), format_info['vbr'] + format_info['abr'])
|
|
format_info.update({
|
|
'url': url,
|
|
'format_id': format_id,
|
|
})
|
|
format_info.update(self._ALL_FORMATS)
|
|
formats.append(format_info)
|
|
|
|
|
|
class ARDMediathekIE(ARDBaseIE):
|
|
IE_NAME = 'ARD:mediathek'
|
|
_VALID_URL = r'^https?://(?:(?:(?:www|classic)\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de|one\.ard\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
|
|
|
|
_TESTS = [{
|
|
# available till 06.10.2023
|
|
'url': 'http://mediathek.daserste.de/Das-Wort-zum-Sonntag/Christian-Rommert-Vielen-Dank-/Video?bcastId=442936&documentId=56727614',
|
|
'md5': 'da2c6b8643cdd4a46f6446bf2786f5b6',
|
|
'info_dict': {
|
|
'id': '56727614',
|
|
'ext': 'mp4',
|
|
'title': 'Christian Rommert: Vielen Dank!?',
|
|
'description': 'md5:ef94f0f576290c7c85137174229a21ca',
|
|
'duration': 233,
|
|
},
|
|
}, {
|
|
# available till 26.07.2022
|
|
'url': 'http://www.ardmediathek.de/tv/S%C3%9CDLICHT/Was-ist-die-Kunst-der-Zukunft-liebe-Ann/BR-Fernsehen/Video?bcastId=34633636&documentId=44726822',
|
|
'md5': '0ab612119ade6214395380723abbfd11',
|
|
'info_dict': {
|
|
'id': '44726822',
|
|
'ext': 'mp4',
|
|
'title': 'Was ist die Kunst der Zukunft, liebe Anna McCarthy?',
|
|
'description': 'md5:4ada28b3e3b5df01647310e41f3a62f5',
|
|
'duration': 1740,
|
|
},
|
|
}, {
|
|
'url': 'https://one.ard.de/tv/Mord-mit-Aussicht/Mord-mit-Aussicht-6-39-T%C3%B6dliche-Nach/ONE/Video?bcastId=46384294&documentId=55586872',
|
|
'only_matching': True,
|
|
}, {
|
|
# audio
|
|
'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
|
|
'only_matching': True,
|
|
}, {
|
|
'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
|
|
'only_matching': True,
|
|
}, {
|
|
# audio
|
|
'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
|
|
'only_matching': True,
|
|
}, {
|
|
'url': 'https://classic.ardmediathek.de/tv/Panda-Gorilla-Co/Panda-Gorilla-Co-Folge-274/Das-Erste/Video?bcastId=16355486&documentId=58234698',
|
|
'only_matching': True,
|
|
}]
|
|
|
|
@classmethod
|
|
def suitable(cls, url):
|
|
return False if ARDBetaMediathekIE.suitable(url) else super(ARDMediathekIE, cls).suitable(url)
|
|
|
|
def _extract_media_info(self, media_info_url, webpage, video_id):
|
|
media_info = self._download_json(
|
|
media_info_url, video_id, 'Downloading media JSON')
|
|
|
|
formats = self._extract_formats(media_info, video_id)
|
|
self._check_additional_formats(formats, video_id)
|
|
|
|
if not formats:
|
|
if '"fsk"' in webpage:
|
|
raise ExtractorError(
|
|
'This video is only available after 20:00', expected=True)
|
|
elif media_info.get('_geoblocked'):
|
|
raise ExtractorError('This video is not available due to geo restriction', expected=True)
|
|
|
|
self._sort_formats(formats)
|
|
|
|
duration = int_or_none(media_info.get('_duration'))
|
|
thumbnail = media_info.get('_previewImage')
|
|
is_live = media_info.get('_isLive') is True
|
|
|
|
subtitles = {}
|
|
subtitle_url = media_info.get('_subtitleUrl')
|
|
if subtitle_url:
|
|
subtitles['de'] = [{
|
|
'ext': 'ttml',
|
|
'url': subtitle_url,
|
|
}]
|
|
|
|
return {
|
|
'id': video_id,
|
|
'duration': duration,
|
|
'thumbnail': thumbnail,
|
|
'is_live': is_live,
|
|
'formats': formats,
|
|
'subtitles': subtitles,
|
|
}
|
|
|
|
def _extract_formats(self, media_info, video_id):
|
|
type_ = media_info.get('_type')
|
|
media_array = media_info.get('_mediaArray', [])
|
|
formats = []
|
|
for media in media_array:
|
|
for stream in media.get('_mediaStreamArray', []):
|
|
stream_urls = stream.get('_stream')
|
|
if not stream_urls:
|
|
continue
|
|
if not isinstance(stream_urls, list):
|
|
stream_urls = [stream_urls]
|
|
quality = stream.get('_quality')
|
|
server = stream.get('_server')
|
|
for stream_url in stream_urls:
|
|
if not url_or_none(stream_url):
|
|
continue
|
|
ext = determine_ext(stream_url)
|
|
if quality != 'auto' and ext in ('f4m', 'm3u8'):
|
|
continue
|
|
|
|
if ext == 'f4m':
|
|
formats.extend(self._extract_f4m_formats(
|
|
update_url_query(stream_url, {
|
|
'hdcore': '3.1.1',
|
|
'plugin': 'aasp-3.1.1.69.124'
|
|
}),
|
|
video_id, f4m_id='hds', fatal=False))
|
|
elif ext == 'm3u8':
|
|
formats.extend(self._extract_m3u8_formats(
|
|
stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
|
|
else:
|
|
if server and server.startswith('rtmp'):
|
|
f = {
|
|
'url': server,
|
|
'play_path': stream_url,
|
|
'format_id': 'rtmp-%s' % quality,
|
|
}
|
|
else:
|
|
f = {
|
|
'url': stream_url,
|
|
'format_id': '%s-%s' % (ext, quality),
|
|
}
|
|
|
|
m = re.search(r'.*/([0-9]+)-[0-9]\..*$', stream_url)
|
|
if m:
|
|
width = int_or_none(m.group(1))
|
|
f.update(self._FORMATS.get(width, {}))
|
|
f.update(self._ALL_FORMATS)
|
|
f.update({'format_id': '%s-%s' % (ext, f['vbr'] + f['abr'])})
|
|
else:
|
|
width = stream.get('_width')
|
|
height = stream.get('_height')
|
|
if not width and not height:
|
|
m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
|
|
if m:
|
|
width = int(m.group('width'))
|
|
height = int(m.group('height'))
|
|
if width and height:
|
|
f.update({
|
|
'width': width,
|
|
'height': height,
|
|
})
|
|
|
|
if type_ == 'audio':
|
|
f['vcodec'] = 'none'
|
|
formats.append(f)
|
|
|
|
return formats
|
|
|
|
def _real_extract(self, url):
|
|
# determine video id from url
|
|
m = re.match(self._VALID_URL, url)
|
|
|
|
document_id = None
|
|
|
|
numid = re.search(r'documentId=([0-9]+)', url)
|
|
if numid:
|
|
document_id = video_id = numid.group(1)
|
|
else:
|
|
video_id = m.group('video_id')
|
|
|
|
webpage = self._download_webpage(url, video_id)
|
|
|
|
ERRORS = (
|
|
('>Leider liegt eine Störung vor.', 'Video %s is unavailable'),
|
|
('>Der gewünschte Beitrag ist nicht mehr verfügbar.<',
|
|
'Video %s is no longer available'),
|
|
)
|
|
|
|
for pattern, message in ERRORS:
|
|
if pattern in webpage:
|
|
raise ExtractorError(message % video_id, expected=True)
|
|
|
|
if re.search(r'[\?&]rss($|[=&])', url):
|
|
doc = compat_etree_fromstring(webpage.encode('utf-8'))
|
|
if doc.tag == 'rss':
|
|
return GenericIE()._extract_rss(url, video_id, doc)
|
|
|
|
title = self._html_search_regex(
|
|
[r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
|
|
r'<meta name="dcterms\.title" content="(.*?)"/>',
|
|
r'<h4 class="headline">(.*?)</h4>',
|
|
r'<title[^>]*>(.*?)</title>'],
|
|
webpage, 'title')
|
|
description = self._html_search_meta(
|
|
'dcterms.abstract', webpage, 'description', default=None)
|
|
if description is None:
|
|
description = self._html_search_meta(
|
|
'description', webpage, 'meta description', default=None)
|
|
if description is None:
|
|
description = self._html_search_regex(
|
|
r'<p\s+class="teasertext">(.+?)</p>',
|
|
webpage, 'teaser text', default=None)
|
|
|
|
# Thumbnail is sometimes not present.
|
|
# It is in the mobile version, but that seems to use a different URL
|
|
# structure altogether.
|
|
thumbnail = self._og_search_thumbnail(webpage, default=None)
|
|
|
|
media_streams = re.findall(r'''(?x)
|
|
mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
|
|
"([^"]+)"''', webpage)
|
|
|
|
if media_streams:
|
|
QUALITIES = qualities(['lo', 'hi', 'hq'])
|
|
formats = []
|
|
for furl in set(media_streams):
|
|
if furl.endswith('.f4m'):
|
|
fid = 'f4m'
|
|
else:
|
|
fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
|
|
fid = fid_m.group(1) if fid_m else None
|
|
formats.append({
|
|
'quality': QUALITIES(fid),
|
|
'format_id': fid,
|
|
'url': furl,
|
|
})
|
|
self._check_additional_formats(formats, video_id)
|
|
self._sort_formats(formats)
|
|
info = {
|
|
'formats': formats,
|
|
}
|
|
else: # request JSON file
|
|
if not document_id:
|
|
video_id = self._search_regex(
|
|
r'/play/(?:config|media)/(\d+)', webpage, 'media id')
|
|
info = self._extract_media_info(
|
|
'http://www.ardmediathek.de/play/media/%s' % video_id,
|
|
webpage, video_id)
|
|
|
|
info.update({
|
|
'id': video_id,
|
|
'title': self._live_title(title) if info.get('is_live') else title,
|
|
'description': description,
|
|
})
|
|
if thumbnail:
|
|
info['thumbnail'] = thumbnail
|
|
|
|
return info
|
|
|
|
|
|
class ARDIE(ARDBaseIE):
|
|
_VALID_URL = r'(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos(-folgen-verpasst)?/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
|
|
_TESTS = [{
|
|
# available till 14.02.2019
|
|
'url': 'http://www.daserste.de/information/talk/maischberger/videos/das-groko-drama-zerlegen-sich-die-volksparteien-video-102.html',
|
|
'md5': '19308261237ed95f2293b05eabede2d0',
|
|
'info_dict': {
|
|
'display_id': 'das-groko-drama-zerlegen-sich-die-volksparteien-video',
|
|
'id': '102',
|
|
'ext': 'mp4',
|
|
'duration': 4435.0,
|
|
'title': 'Das GroKo-Drama: Zerlegen sich die Volksparteien?',
|
|
'upload_date': '20180214',
|
|
'thumbnail': r're:^https?://.*\.jpg$',
|
|
},
|
|
}, {
|
|
'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
|
|
'only_matching': True,
|
|
}, {
|
|
'url': 'https://www.daserste.de/unterhaltung/serie/babylon-berlin/videos-folgen-verpasst/Babylon_Berlin_1-folge-100.html',
|
|
'only_matching': True,
|
|
}]
|
|
|
|
def _real_extract(self, url):
|
|
mobj = re.match(self._VALID_URL, url)
|
|
display_id = mobj.group('display_id')
|
|
|
|
player_url = mobj.group('mainurl') + '~playerXml.xml'
|
|
doc = self._download_xml(player_url, display_id)
|
|
video_node = doc.find('./video')
|
|
upload_date = unified_strdate(xpath_text(
|
|
video_node, './broadcastDate'))
|
|
thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
|
|
|
|
formats = []
|
|
format_ids = {}
|
|
for a in video_node.findall('.//asset'):
|
|
if a.find('./serverPrefix').text:
|
|
url = a.find('./serverPrefix').text
|
|
playpath = a.find('./fileName').text
|
|
else:
|
|
url = a.find('./fileName').text
|
|
playpath = None
|
|
|
|
if url.endswith('.f4m'):
|
|
formats.extend(self._extract_f4m_formats(
|
|
url + '?hdcore=3.11.0',
|
|
display_id, f4m_id='hds', fatal=False))
|
|
elif url.endswith('.m3u8'):
|
|
formats.extend(self._extract_m3u8_formats(
|
|
url, display_id, m3u8_id='hls', fatal=False))
|
|
else:
|
|
tbr = int_or_none(a.find('./totalBitrate').text)
|
|
format_id = '%s-%s' % (determine_ext(url), tbr)
|
|
if 'HbbTV' in a.attrib['type']:
|
|
continue
|
|
if format_id in format_ids:
|
|
format_ids[format_id] += 1
|
|
format_id += '-%s' % format_ids[format_id]
|
|
else:
|
|
format_ids[format_id] = 1
|
|
f = {
|
|
'url': url,
|
|
'playpath': playpath,
|
|
'format_id': format_id,
|
|
'width': int_or_none(a.find('./frameWidth').text),
|
|
'height': int_or_none(a.find('./frameHeight').text),
|
|
'vbr': int_or_none(a.find('./bitrateVideo').text),
|
|
'abr': int_or_none(a.find('./bitrateAudio').text),
|
|
'vcodec': a.find('./codecVideo').text,
|
|
'acodec': 'LC-AAC',
|
|
'tbr': tbr,
|
|
}
|
|
formats.append(f)
|
|
|
|
self._check_additional_formats(formats, display_id)
|
|
self._sort_formats(formats)
|
|
|
|
return {
|
|
'id': mobj.group('id'),
|
|
'formats': formats,
|
|
'display_id': display_id,
|
|
'title': video_node.find('./title').text,
|
|
'duration': parse_duration(video_node.find('./duration').text),
|
|
'upload_date': upload_date,
|
|
'thumbnail': thumbnail,
|
|
}
|
|
|
|
|
|
|
|
class ARDBetaMediathekIE(InfoExtractor):
|
|
_VALID_URL = r'https://(?:beta|www)\.ardmediathek\.de/[^/]+/(?:player|live)/(?P<video_id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^/?#]+))?'
|
|
_TESTS = [{
|
|
'url': 'https://beta.ardmediathek.de/ard/player/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE/die-robuste-roswita',
|
|
'md5': '7338f01de1ca7af9538cc0bdfc438dd1',
|
|
'info_dict': {
|
|
'display_id': 'die-robuste-roswita',
|
|
'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
|
|
'title': 'Tatort: Die robuste Roswita',
|
|
'description': r're:^Der Mord.*trüber ist als die Ilm.',
|
|
'duration': 5316,
|
|
'thumbnail': r're:^https?://img.ardmediathek.de.*$',
|
|
'upload_date': '20180826',
|
|
'ext': 'mp4',
|
|
},
|
|
}, {
|
|
'url': 'https://www.ardmediathek.de/ard/player/Y3JpZDovL3N3ci5kZS9hZXgvbzEwNzE5MTU/',
|
|
'only_matching': True,
|
|
}, {
|
|
'url': 'https://www.ardmediathek.de/swr/live/Y3JpZDovL3N3ci5kZS8xMzQ4MTA0Mg',
|
|
'only_matching': True,
|
|
}]
|
|
|
|
def _real_extract(self, url):
|
|
mobj = re.match(self._VALID_URL, url)
|
|
video_id = mobj.group('video_id')
|
|
display_id = mobj.group('display_id') or video_id
|
|
|
|
webpage = self._download_webpage(url, display_id)
|
|
data_json = self._search_regex(r'window\.__APOLLO_STATE__\s*=\s*(\{.*);\n', webpage, 'json')
|
|
data = self._parse_json(data_json, display_id)
|
|
|
|
res = {
|
|
'id': video_id,
|
|
'display_id': display_id,
|
|
}
|
|
formats = []
|
|
subtitles = {}
|
|
geoblocked = False
|
|
for widget in data.values():
|
|
|
|
if widget.get('_geoblocked') is True:
|
|
geoblocked = True
|
|
if '_duration' in widget:
|
|
res['duration'] = int_or_none(widget['_duration'])
|
|
if 'clipTitle' in widget:
|
|
res['title'] = widget['clipTitle']
|
|
if '_previewImage' in widget:
|
|
res['thumbnail'] = widget['_previewImage']
|
|
if 'broadcastedOn' in widget:
|
|
res['timestamp'] = unified_timestamp(widget['broadcastedOn'])
|
|
if 'synopsis' in widget:
|
|
res['description'] = widget['synopsis']
|
|
subtitle_url = url_or_none(widget.get('_subtitleUrl'))
|
|
if subtitle_url:
|
|
subtitles.setdefault('de', []).append({
|
|
'ext': 'ttml',
|
|
'url': subtitle_url,
|
|
})
|
|
if '_quality' in widget:
|
|
format_url = url_or_none(try_get(
|
|
widget, lambda x: x['_stream']['json'][0]))
|
|
if not format_url:
|
|
continue
|
|
ext = determine_ext(format_url)
|
|
if ext == 'f4m':
|
|
formats.extend(self._extract_f4m_formats(
|
|
format_url + '?hdcore=3.11.0',
|
|
video_id, f4m_id='hds', fatal=False))
|
|
elif ext == 'm3u8':
|
|
formats.extend(self._extract_m3u8_formats(
|
|
format_url, video_id, 'mp4', m3u8_id='hls',
|
|
fatal=False))
|
|
else:
|
|
# HTTP formats are not available when geoblocked is True,
|
|
# other formats are fine though
|
|
if geoblocked:
|
|
continue
|
|
m = re.search(r'.*/([0-9]+)-[0-9]\..*$', format_url)
|
|
width = int_or_none(m.group(1)) if m else None
|
|
if width and width in self._FORMATS:
|
|
info = self._FORMATS[width]
|
|
info['format_id'] = 'http-%s-%s' % (ext, info['vbr'] + info['abr'])
|
|
info.update(self._ALL_FORMATS)
|
|
else:
|
|
quality = str_or_none(widget.get('_quality'))
|
|
info = {
|
|
'format_id': 'http-%s-%s' % (ext, quality) if quality else 'http-%s' % ext
|
|
}
|
|
ext = determine_ext(format_url)
|
|
info.update({
|
|
'url': format_url,
|
|
'preference': 10 # Plain HTTP, that's nice
|
|
})
|
|
formats.append(info)
|
|
|
|
if not formats and geoblocked:
|
|
self.raise_geo_restricted(
|
|
msg='This video is not available due to geoblocking',
|
|
countries=['DE'])
|
|
|
|
self._remove_duplicate_formats(formats)
|
|
self._check_additional_formats(formats, video_id)
|
|
self._sort_formats(formats)
|
|
res.update({
|
|
'subtitles': subtitles,
|
|
'formats': formats,
|
|
})
|
|
|
|
return res
|