X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/d88adeed4f0a22f14db3a2b3fd3c59b266b97a78..1d04e265122c7ed6edf8f3c75a0619931b9368b9:/youtube_dl/extractor/prosiebensat1.py?ds=sidebyside
diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
index da64a1a..d8a4bd2 100644
--- a/youtube_dl/extractor/prosiebensat1.py
+++ b/youtube_dl/extractor/prosiebensat1.py
@@ -1,36 +1,158 @@
-# encoding: utf-8
+# coding: utf-8
from __future__ import unicode_literals
import re
from hashlib import sha1
from .common import InfoExtractor
+from ..compat import compat_str
from ..utils import (
- compat_urllib_parse,
+ ExtractorError,
+ determine_ext,
+ float_or_none,
+ int_or_none,
unified_strdate,
)
-class ProSiebenSat1IE(InfoExtractor):
+class ProSiebenSat1BaseIE(InfoExtractor):
+ def _extract_video_info(self, url, clip_id):
+ client_location = url
+
+ video = self._download_json(
+ 'http://vas.sim-technik.de/vas/live/v2/videos',
+ clip_id, 'Downloading videos JSON', query={
+ 'access_token': self._TOKEN,
+ 'client_location': client_location,
+ 'client_name': self._CLIENT_NAME,
+ 'ids': clip_id,
+ })[0]
+
+ if video.get('is_protected') is True:
+ raise ExtractorError('This video is DRM protected.', expected=True)
+
+ duration = float_or_none(video.get('duration'))
+ source_ids = [compat_str(source['id']) for source in video['sources']]
+
+ client_id = self._SALT[:2] + sha1(''.join([clip_id, self._SALT, self._TOKEN, client_location, self._SALT, self._CLIENT_NAME]).encode('utf-8')).hexdigest()
+
+ sources = self._download_json(
+ 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources' % clip_id,
+ clip_id, 'Downloading sources JSON', query={
+ 'access_token': self._TOKEN,
+ 'client_id': client_id,
+ 'client_location': client_location,
+ 'client_name': self._CLIENT_NAME,
+ })
+ server_id = sources['server_id']
+
+ def fix_bitrate(bitrate):
+ bitrate = int_or_none(bitrate)
+ if not bitrate:
+ return None
+ return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
+
+ formats = []
+ for source_id in source_ids:
+ client_id = self._SALT[:2] + sha1(''.join([self._SALT, clip_id, self._TOKEN, server_id, client_location, source_id, self._SALT, self._CLIENT_NAME]).encode('utf-8')).hexdigest()
+ urls = self._download_json(
+ 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url' % clip_id,
+ clip_id, 'Downloading urls JSON', fatal=False, query={
+ 'access_token': self._TOKEN,
+ 'client_id': client_id,
+ 'client_location': client_location,
+ 'client_name': self._CLIENT_NAME,
+ 'server_id': server_id,
+ 'source_ids': source_id,
+ })
+ if not urls:
+ continue
+ if urls.get('status_code') != 0:
+ raise ExtractorError('This video is unavailable', expected=True)
+ urls_sources = urls['sources']
+ if isinstance(urls_sources, dict):
+ urls_sources = urls_sources.values()
+ for source in urls_sources:
+ source_url = source.get('url')
+ if not source_url:
+ continue
+ protocol = source.get('protocol')
+ mimetype = source.get('mimetype')
+ if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
+ formats.extend(self._extract_f4m_formats(
+ source_url, clip_id, f4m_id='hds', fatal=False))
+ elif mimetype == 'application/x-mpegURL':
+ formats.extend(self._extract_m3u8_formats(
+ source_url, clip_id, 'mp4', 'm3u8_native',
+ m3u8_id='hls', fatal=False))
+ elif mimetype == 'application/dash+xml':
+ formats.extend(self._extract_mpd_formats(
+ source_url, clip_id, mpd_id='dash', fatal=False))
+ else:
+ tbr = fix_bitrate(source['bitrate'])
+ if protocol in ('rtmp', 'rtmpe'):
+ mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
+ if not mobj:
+ continue
+ path = mobj.group('path')
+ mp4colon_index = path.rfind('mp4:')
+ app = path[:mp4colon_index]
+ play_path = path[mp4colon_index:]
+ formats.append({
+ 'url': '%s/%s' % (mobj.group('url'), app),
+ 'app': app,
+ 'play_path': play_path,
+ 'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
+ 'page_url': 'http://www.prosieben.de',
+ 'tbr': tbr,
+ 'ext': 'flv',
+ 'format_id': 'rtmp%s' % ('-%d' % tbr if tbr else ''),
+ })
+ else:
+ formats.append({
+ 'url': source_url,
+ 'tbr': tbr,
+ 'format_id': 'http%s' % ('-%d' % tbr if tbr else ''),
+ })
+ self._sort_formats(formats)
+
+ return {
+ 'duration': duration,
+ 'formats': formats,
+ }
+
+
+class ProSiebenSat1IE(ProSiebenSat1BaseIE):
IE_NAME = 'prosiebensat1'
IE_DESC = 'ProSiebenSat.1 Digital'
- _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|ran|the-voice-of-germany)\.de|fem\.com)/(?P<id>.+)'
+ _VALID_URL = r'''(?x)
+ https?://
+ (?:www\.)?
+ (?:
+ (?:
+ prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|7tv|advopedia
+ )\.(?:de|at|ch)|
+ ran\.de|fem\.com|advopedia\.de
+ )
+ /(?P<id>.+)
+ '''
_TESTS = [
{
+ # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242
+ # in response to fixing https://github.com/rg3/youtube-dl/issues/6215:
+ # - malformed f4m manifest support
+ # - proper handling of URLs starting with `https?://` in 2.0 manifests
+ # - recursive child f4m manifests extraction
'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
'info_dict': {
'id': '2104602',
'ext': 'mp4',
- 'title': 'Staffel 2, Episode 18 - Jahresrückblick',
+ 'title': 'Episode 18 - Staffel 2',
'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
'upload_date': '20131231',
'duration': 5845.04,
},
- 'params': {
- # rtmp download
- 'skip_download': True,
- },
},
{
'url': 'http://www.prosieben.de/videokatalog/Gesellschaft/Leben/Trends/video-Lady-Umstyling-f%C3%BCr-Audrina-Rebekka-Audrina-Fergen-billig-aussehen-Battal-Modica-700544.html',
@@ -62,6 +184,7 @@ class ProSiebenSat1IE(InfoExtractor):
# rtmp download
'skip_download': True,
},
+ 'skip': 'This video is unavailable',
},
{
'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
@@ -77,6 +200,7 @@ class ProSiebenSat1IE(InfoExtractor):
# rtmp download
'skip_download': True,
},
+ 'skip': 'This video is unavailable',
},
{
'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
@@ -85,13 +209,14 @@ class ProSiebenSat1IE(InfoExtractor):
'ext': 'mp4',
'title': 'Im Interview: Kai Wiesinger',
'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
- 'upload_date': '20140225',
+ 'upload_date': '20140203',
'duration': 522.56,
},
'params': {
# rtmp download
'skip_download': True,
},
+ 'skip': 'This video is unavailable',
},
{
'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
@@ -100,13 +225,14 @@ class ProSiebenSat1IE(InfoExtractor):
'ext': 'mp4',
'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
'description': 'md5:2669cde3febe9bce13904f701e774eb6',
- 'upload_date': '20140225',
+ 'upload_date': '20141014',
'duration': 2410.44,
},
'params': {
# rtmp download
'skip_download': True,
},
+ 'skip': 'This video is unavailable',
},
{
'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
@@ -122,6 +248,7 @@ class ProSiebenSat1IE(InfoExtractor):
# rtmp download
'skip_download': True,
},
+ 'skip': 'This video is unavailable',
},
{
'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
@@ -134,7 +261,6 @@ class ProSiebenSat1IE(InfoExtractor):
'duration': 469.88,
},
'params': {
- # rtmp download
'skip_download': True,
},
},
@@ -144,33 +270,98 @@ class ProSiebenSat1IE(InfoExtractor):
'id': '2156342',
'ext': 'mp4',
'title': 'Kurztrips zum Valentinstag',
- 'description': 'md5:8ba6301e70351ae0bedf8da00f7ba528',
- 'upload_date': '20130206',
+ 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
'duration': 307.24,
},
'params': {
- # rtmp download
'skip_download': True,
},
},
+ {
+ 'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
+ 'info_dict': {
+ 'id': '439664',
+ 'title': 'Episode 8 - Ganze Folge - Playlist',
+ 'description': 'md5:63b8963e71f481782aeea877658dec84',
+ },
+ 'playlist_count': 2,
+ 'skip': 'This video is unavailable',
+ },
+ {
+ 'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
+ 'info_dict': {
+ 'id': '4187506',
+ 'ext': 'mp4',
+ 'title': 'Best of Circus HalliGalli',
+ 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
+ 'upload_date': '20151229',
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # title in <h2 class="subtitle">
+ 'url': 'http://www.prosieben.de/stars/oscar-award/videos/jetzt-erst-enthuellt-das-geheimnis-von-emma-stones-oscar-robe-clip',
+ 'info_dict': {
+ 'id': '4895826',
+ 'ext': 'mp4',
+ 'title': 'Jetzt erst enthüllt: Das Geheimnis von Emma Stones Oscar-Robe',
+ 'description': 'md5:e5ace2bc43fadf7b63adc6187e9450b9',
+ 'upload_date': '20170302',
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ 'skip': 'geo restricted to Germany',
+ },
+ {
+ # geo restricted to Germany
+ 'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',
+ 'only_matching': True,
+ },
+ {
+ # geo restricted to Germany
+ 'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
+ 'only_matching': True,
+ },
+ {
+ 'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
+ 'only_matching': True,
+ },
+ {
+ 'url': 'http://www.advopedia.de/videos/lenssen-klaert-auf/lenssen-klaert-auf-folge-8-staffel-3-feiertage-und-freie-tage',
+ 'only_matching': True,
+ },
]
+ _TOKEN = 'prosieben'
+ _SALT = '01!8d8F_)r9]4s[qeuXfP%'
+ _CLIENT_NAME = 'kolibri-2.0.19-splec4'
_CLIPID_REGEXES = [
r'"clip_id"\s*:\s+"(\d+)"',
r'clipid: "(\d+)"',
r'clip[iI]d=(\d+)',
+ r'clip[iI]d\s*=\s*["\'](\d+)',
+ r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
]
_TITLE_REGEXES = [
r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
r'<header class="clearfix">\s*<h3>(.+?)</h3>',
r'<!-- start video -->\s*<h1>(.+?)</h1>',
r'<h1 class="att-name">\s*(.+?)</h1>',
+ r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
+ r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
+ r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
+ r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',
]
_DESCRIPTION_REGEXES = [
r'<p itemprop="description">\s*(.+?)</p>',
r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
r'<p class="att-description">\s*(.+?)\s*</p>',
+ r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
+ r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
]
_UPLOAD_DATE_REGEXES = [
r'<meta property="og:published_time" content="(.+?)">',
@@ -179,108 +370,78 @@ class ProSiebenSat1IE(InfoExtractor):
r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
]
+ _PAGE_TYPE_REGEXES = [
+ r'<meta name="page_type" content="([^"]+)">',
+ r"'itemType'\s*:\s*'([^']*)'",
+ ]
+ _PLAYLIST_ID_REGEXES = [
+ r'content[iI]d=(\d+)',
+ r"'itemId'\s*:\s*'([^']*)'",
+ ]
+ _PLAYLIST_CLIP_REGEXES = [
+ r'(?s)data-qvt=.+?<a href="([^"]+)"',
+ ]
- def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- video_id = mobj.group('id')
-
- page = self._download_webpage(url, video_id, 'Downloading page')
-
- clip_id = self._html_search_regex(self._CLIPID_REGEXES, page, 'clip id')
-
- access_token = 'testclient'
- client_name = 'kolibri-1.2.5'
- client_location = url
-
- videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
- 'access_token': access_token,
- 'client_location': client_location,
- 'client_name': client_name,
- 'ids': clip_id,
- })
-
- videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
-
- duration = float(videos[0]['duration'])
- source_ids = [source['id'] for source in videos[0]['sources']]
- source_ids_str = ','.join(map(str, source_ids))
-
- g = '01!8d8F_)r9]4s[qeuXfP%'
-
- client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
- .encode('utf-8')).hexdigest()
-
- sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse.urlencode({
- 'access_token': access_token,
- 'client_id': client_id,
- 'client_location': client_location,
- 'client_name': client_name,
- }))
-
- sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
- server_id = sources['server_id']
-
- client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
- client_location, source_ids_str, g, client_name])
- .encode('utf-8')).hexdigest()
-
- url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse.urlencode({
- 'access_token': access_token,
- 'client_id': client_id,
- 'client_location': client_location,
- 'client_name': client_name,
- 'server_id': server_id,
- 'source_ids': source_ids_str,
- }))
-
- urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
-
- title = self._html_search_regex(self._TITLE_REGEXES, page, 'title')
- description = self._html_search_regex(self._DESCRIPTION_REGEXES, page, 'description', fatal=False)
- thumbnail = self._og_search_thumbnail(page)
-
+ def _extract_clip(self, url, webpage):
+ clip_id = self._html_search_regex(
+ self._CLIPID_REGEXES, webpage, 'clip id')
+ title = self._html_search_regex(
+ self._TITLE_REGEXES, webpage, 'title',
+ default=None) or self._og_search_title(webpage)
+ info = self._extract_video_info(url, clip_id)
+ description = self._html_search_regex(
+ self._DESCRIPTION_REGEXES, webpage, 'description', default=None)
+ if description is None:
+ description = self._og_search_description(webpage)
+ thumbnail = self._og_search_thumbnail(webpage)
upload_date = unified_strdate(self._html_search_regex(
- self._UPLOAD_DATE_REGEXES, page, 'upload date', fatal=False))
-
- formats = []
-
- urls_sources = urls['sources']
- if isinstance(urls_sources, dict):
- urls_sources = urls_sources.values()
-
- def fix_bitrate(bitrate):
- return bitrate / 1000 if bitrate % 1000 == 0 else bitrate
+ self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
- for source in urls_sources:
- protocol = source['protocol']
- if protocol == 'rtmp' or protocol == 'rtmpe':
- mobj = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', source['url'])
- if not mobj:
- continue
- formats.append({
- 'url': mobj.group('url'),
- 'app': mobj.group('app'),
- 'play_path': mobj.group('playpath'),
- 'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
- 'page_url': 'http://www.prosieben.de',
- 'vbr': fix_bitrate(source['bitrate']),
- 'ext': 'mp4',
- 'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
- })
- else:
- formats.append({
- 'url': source['url'],
- 'vbr': fix_bitrate(source['bitrate']),
- })
-
- self._sort_formats(formats)
-
- return {
+ info.update({
'id': clip_id,
'title': title,
'description': description,
'thumbnail': thumbnail,
'upload_date': upload_date,
- 'duration': duration,
- 'formats': formats,
- }
\ No newline at end of file
+ })
+ return info
+
+ def _extract_playlist(self, url, webpage):
+ playlist_id = self._html_search_regex(
+ self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
+ playlist = self._parse_json(
+ self._search_regex(
+ r'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',
+ webpage, 'playlist'),
+ playlist_id)
+ entries = []
+ for item in playlist:
+ clip_id = item.get('id') or item.get('upc')
+ if not clip_id:
+ continue
+ info = self._extract_video_info(url, clip_id)
+ info.update({
+ 'id': clip_id,
+ 'title': item.get('title') or item.get('teaser', {}).get('headline'),
+ 'description': item.get('teaser', {}).get('description'),
+ 'thumbnail': item.get('poster'),
+ 'duration': float_or_none(item.get('duration')),
+ 'series': item.get('tvShowTitle'),
+ 'uploader': item.get('broadcastPublisher'),
+ })
+ entries.append(info)
+ return self.playlist_result(entries, playlist_id)
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+ webpage = self._download_webpage(url, video_id)
+ page_type = self._search_regex(
+ self._PAGE_TYPE_REGEXES, webpage,
+ 'page type', default='clip').lower()
+ if page_type == 'clip':
+ return self._extract_clip(url, webpage)
+ elif page_type == 'playlist':
+ return self._extract_playlist(url, webpage)
+ else:
+ raise ExtractorError(
+ 'Unsupported page type %s' % page_type, expected=True)