Merge branch 'master' into GoogleDrive-issue-13619

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.07.09*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.07.15*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2017.07.09**
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2017.07.15**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2017.07.09
+[debug] youtube-dl version 2017.07.15
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

ChangeLog

@@ -1,11 +1,25 @@
-version <unreleased>
+version 2017.07.15
+
+Core
+* [YoutubeDL] Don't expand environment variables in meta fields (#13637)
 
 Extractors
+* [spiegeltv] Delegate extraction to nexx extractor (#13159)
++ [nexx] Add support for nexx.cloud (#10807, #13465)
+* [generic] Fix rutube embeds extraction (#13641)
+* [karrierevideos] Fix title extraction (#13641)
+* [youtube] Don't capture YouTube Red ad for creator meta field (#13621)
+* [slideshare] Fix extraction (#13617)
++ [5tv] Add another video URL pattern (#13354, #13606)
+* [drtv] Make HLS and HDS extraction non fatal
+* [ted] Fix subtitles extraction (#13628, #13629)
 * [vine] Make sure the title won't be empty
 + [twitter] Support HLS streams in vmap URLs
 + [periscope] Support pscp.tv URLs in embedded frames
+* [twitter] Extract mp4 urls via mobile API (#12726)
 * [niconico] Fix authentication error handling (#12486)
 * [giantbomb] Extract m3u8 formats (#13626)
++ [vlive:playlist] Add support for playlists (#13613)
 
 
 version 2017.07.09

docs/supportedsites.md

@@ -521,6 +521,7 @@
  - **NextMedia**: 蘋果日報
  - **NextMediaActionNews**: 蘋果日報 - 動新聞
  - **NextTV**: 壹電視
+ - **Nexx**
  - **nfb**: National Film Board of Canada
  - **nfl.com**
  - **NhkVod**
@@ -942,6 +943,7 @@
  - **vk:wallpost**
  - **vlive**
  - **vlive:channel**
+ - **vlive:playlist**
  - **Vodlocker**
  - **VODPl**
  - **VODPlatform**

youtube_dl/YoutubeDL.py

@@ -20,13 +20,14 @@ import re
 import shutil
 import subprocess
 import socket
-import string
 import sys
 import time
 import tokenize
 import traceback
 import random
 
+from string import ascii_letters
+
 from .compat import (
     compat_basestring,
     compat_cookiejar,
@@ -679,7 +680,7 @@ class YoutubeDL(object):
             # correspondingly that is not what we want since we need to keep
             # '%%' intact for template dict substitution step. Working around
             # with boundary-alike separator hack.
-            sep = ''.join([random.choice(string.ascii_letters) for _ in range(32)])
+            sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
             outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
 
             # outtmpl should be expand_path'ed before template dict substitution

youtube_dl/extractor/adn.py

@@ -107,11 +107,13 @@ class ADNIE(InfoExtractor):
         metas = options.get('metas') or {}
         title = metas.get('title') or video_info['title']
         links = player_config.get('links') or {}
+        error = None
         if not links:
             links_url = player_config['linksurl']
             links_data = self._download_json(urljoin(
                 self._BASE_URL, links_url), video_id)
             links = links_data.get('links') or {}
+            error = links_data.get('error')
 
         formats = []
         for format_id, qualities in links.items():
@@ -130,7 +132,8 @@ class ADNIE(InfoExtractor):
                     for f in m3u8_formats:
                         f['language'] = 'fr'
                 formats.extend(m3u8_formats)
-        error = options.get('error')
+        if not error:
+            error = options.get('error')
         if not formats and error:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
         self._sort_formats(formats)

youtube_dl/extractor/common.py

@@ -730,12 +730,12 @@ class InfoExtractor(object):
             video_info['title'] = video_title
         return video_info
 
-    def playlist_from_matches(self, matches, video_id, video_title, getter=None, ie=None):
+    def playlist_from_matches(self, matches, playlist_id=None, playlist_title=None, getter=None, ie=None):
-        urlrs = orderedSet(
+        urls = orderedSet(
             self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
             for m in matches)
         return self.playlist_result(
-            urlrs, playlist_id=video_id, playlist_title=video_title)
+            urls, playlist_id=playlist_id, playlist_title=playlist_title)
 
     @staticmethod
     def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):

youtube_dl/extractor/crunchyroll.py

@@ -510,7 +510,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
         # webpage provide more accurate data than series_title from XML
         series = self._html_search_regex(
-            r'id=["\']showmedia_about_episode_num[^>]+>\s*<a[^>]+>([^<]+)',
+            r'(?s)<h\d[^>]+\bid=["\']showmedia_about_episode_num[^>]+>(.+?)</h\d',
             webpage, 'series', fatal=False)
         season = xpath_text(metadata, 'series_title')
 
@@ -518,7 +518,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         episode_number = int_or_none(xpath_text(metadata, 'episode_number'))
 
         season_number = int_or_none(self._search_regex(
-            r'(?s)<h4[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h4>\s*<h4>\s*Season (\d+)',
+            r'(?s)<h\d[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h\d>\s*<h4>\s*Season (\d+)',
             webpage, 'season number', default=None))
 
         return {

youtube_dl/extractor/extractors.py

@@ -653,6 +653,10 @@ from .nextmedia import (
     AppleDailyIE,
     NextTVIE,
 )
+from .nexx import (
+    NexxIE,
+    NexxEmbedIE,
+)
 from .nfb import NFBIE
 from .nfl import NFLIE
 from .nhk import NhkVodIE
@@ -761,6 +765,7 @@ from .pandoratv import PandoraTVIE
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE
+from .pearvideo import PearVideoIE
 from .people import PeopleIE
 from .periscope import (
     PeriscopeIE,

youtube_dl/extractor/generic.py

@@ -36,6 +36,10 @@ from .brightcove import (
     BrightcoveLegacyIE,
     BrightcoveNewIE,
 )
+from .nexx import (
+    NexxIE,
+    NexxEmbedIE,
+)
 from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE
@@ -1549,6 +1553,43 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['BrightcoveLegacy'],
         },
+        # Nexx embed
+        {
+            'url': 'https://www.funk.net/serien/5940e15073f6120001657956/items/593efbb173f6120001657503',
+            'info_dict': {
+                'id': '247746',
+                'ext': 'mp4',
+                'title': "Yesterday's Jam (OV)",
+                'description': 'md5:09bc0984723fed34e2581624a84e05f0',
+                'timestamp': 1492594816,
+                'upload_date': '20170419',
+            },
+            'params': {
+                'format': 'bestvideo',
+                'skip_download': True,
+            },
+        },
+        # Nexx iFrame embed
+        {
+            'url': 'http://www.spiegel.de/sptv/spiegeltv/spiegel-tv-ueber-schnellste-katapult-achterbahn-der-welt-taron-a-1137884.html',
+            'info_dict': {
+                'id': '161464',
+                'ext': 'mp4',
+                'title': 'Nervenkitzel Achterbahn',
+                'alt_title': 'Karussellbauer in Deutschland',
+                'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
+                'release_year': 2005,
+                'creator': 'SPIEGEL TV',
+                'thumbnail': r're:^https?://.*\.jpg$',
+                'duration': 2761,
+                'timestamp': 1394021479,
+                'upload_date': '20140305',
+            },
+            'params': {
+                'format': 'bestvideo',
+                'skip_download': True,
+            },
+        },
         # Facebook <iframe> embed
         {
             'url': 'https://www.hostblogger.de/blog/archives/6181-Auto-jagt-Betonmischer.html',
@@ -2133,6 +2174,16 @@ class GenericIE(InfoExtractor):
         if bc_urls:
             return self.playlist_from_matches(bc_urls, video_id, video_title, ie='BrightcoveNew')
 
+        # Look for Nexx embeds
+        nexx_urls = NexxIE._extract_urls(webpage)
+        if nexx_urls:
+            return self.playlist_from_matches(nexx_urls, video_id, video_title, ie=NexxIE.ie_key())
+
+        # Look for Nexx iFrame embeds
+        nexx_embed_urls = NexxEmbedIE._extract_urls(webpage)
+        if nexx_embed_urls:
+            return self.playlist_from_matches(nexx_embed_urls, video_id, video_title, ie=NexxEmbedIE.ie_key())
+
         # Look for ThePlatform embeds
         tp_urls = ThePlatformIE._extract_urls(webpage)
         if tp_urls:

youtube_dl/extractor/nexx.py

@@ -0,0 +1,271 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import hashlib
+import random
+import re
+import time
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    parse_duration,
+    try_get,
+    urlencode_postdata,
+)
+
+
+class NexxIE(InfoExtractor):
+    _VALID_URL = r'https?://api\.nexx(?:\.cloud|cdn\.com)/v3/(?P<domain_id>\d+)/videos/byid/(?P<id>\d+)'
+    _TESTS = [{
+        # movie
+        'url': 'https://api.nexx.cloud/v3/748/videos/byid/128907',
+        'md5': '16746bfc28c42049492385c989b26c4a',
+        'info_dict': {
+            'id': '128907',
+            'ext': 'mp4',
+            'title': 'Stiftung Warentest',
+            'alt_title': 'Wie ein Test abläuft',
+            'description': 'md5:d1ddb1ef63de721132abd38639cc2fd2',
+            'release_year': 2013,
+            'creator': 'SPIEGEL TV',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2509,
+            'timestamp': 1384264416,
+            'upload_date': '20131112',
+        },
+        'params': {
+            'format': 'bestvideo',
+        },
+    }, {
+        # episode
+        'url': 'https://api.nexx.cloud/v3/741/videos/byid/247858',
+        'info_dict': {
+            'id': '247858',
+            'ext': 'mp4',
+            'title': 'Return of the Golden Child (OV)',
+            'description': 'md5:5d969537509a92b733de21bae249dc63',
+            'release_year': 2017,
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 1397,
+            'timestamp': 1495033267,
+            'upload_date': '20170517',
+            'episode_number': 2,
+            'season_number': 2,
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://api.nexxcdn.com/v3/748/videos/byid/128907',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        # Reference:
+        # 1. https://nx-s.akamaized.net/files/201510/44.pdf
+
+        entries = []
+
+        # JavaScript Integration
+        mobj = re.search(
+            r'<script\b[^>]+\bsrc=["\']https?://require\.nexx(?:\.cloud|cdn\.com)/(?P<id>\d+)',
+            webpage)
+        if mobj:
+            domain_id = mobj.group('id')
+            for video_id in re.findall(
+                    r'(?is)onPLAYReady.+?_play\.init\s*\(.+?\s*,\s*["\']?(\d+)',
+                    webpage):
+                entries.append(
+                    'https://api.nexx.cloud/v3/%s/videos/byid/%s'
+                    % (domain_id, video_id))
+
+        # TODO: support more embed formats
+
+        return entries
+
+    @staticmethod
+    def _extract_url(webpage):
+        return NexxIE._extract_urls(webpage)[0]
+
+    def _handle_error(self, response):
+        status = int_or_none(try_get(
+            response, lambda x: x['metadata']['status']) or 200)
+        if 200 <= status < 300:
+            return
+        raise ExtractorError(
+            '%s said: %s' % (self.IE_NAME, response['metadata']['errorhint']),
+            expected=True)
+
+    def _call_api(self, domain_id, path, video_id, data=None, headers={}):
+        headers['Content-Type'] = 'application/x-www-form-urlencoded; charset=UTF-8'
+        result = self._download_json(
+            'https://api.nexx.cloud/v3/%s/%s' % (domain_id, path), video_id,
+            'Downloading %s JSON' % path, data=urlencode_postdata(data),
+            headers=headers)
+        self._handle_error(result)
+        return result['result']
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        domain_id, video_id = mobj.group('domain_id', 'id')
+
+        # Reverse engineered from JS code (see getDeviceID function)
+        device_id = '%d:%d:%d%d' % (
+            random.randint(1, 4), int(time.time()),
+            random.randint(1e4, 99999), random.randint(1, 9))
+
+        result = self._call_api(domain_id, 'session/init', video_id, data={
+            'nxp_devh': device_id,
+            'nxp_userh': '',
+            'precid': '0',
+            'playlicense': '0',
+            'screenx': '1920',
+            'screeny': '1080',
+            'playerversion': '6.0.00',
+            'gateway': 'html5',
+            'adGateway': '',
+            'explicitlanguage': 'en-US',
+            'addTextTemplates': '1',
+            'addDomainData': '1',
+            'addAdModel': '1',
+        }, headers={
+            'X-Request-Enable-Auth-Fallback': '1',
+        })
+
+        cid = result['general']['cid']
+
+        # As described in [1] X-Request-Token generation algorithm is
+        # as follows:
+        #   md5( operation + domain_id + domain_secret )
+        # where domain_secret is a static value that will be given by nexx.tv
+        # as per [1]. Here is how this "secret" is generated (reversed
+        # from _play.api.init function, search for clienttoken). So it's
+        # actually not static and not that much of a secret.
+        # 1. https://nexxtvstorage.blob.core.windows.net/files/201610/27.pdf
+        secret = result['device']['clienttoken'][int(device_id[0]):]
+        secret = secret[0:len(secret) - int(device_id[-1])]
+
+        op = 'byid'
+
+        # Reversed from JS code for _play.api.call function (search for
+        # X-Request-Token)
+        request_token = hashlib.md5(
+            ''.join((op, domain_id, secret)).encode('utf-8')).hexdigest()
+
+        video = self._call_api(
+            domain_id, 'videos/%s/%s' % (op, video_id), video_id, data={
+                'additionalfields': 'language,channel,actors,studio,licenseby,slug,subtitle,teaser,description',
+                'addInteractionOptions': '1',
+                'addStatusDetails': '1',
+                'addStreamDetails': '1',
+                'addCaptions': '1',
+                'addScenes': '1',
+                'addHotSpots': '1',
+                'addBumpers': '1',
+                'captionFormat': 'data',
+            }, headers={
+                'X-Request-CID': cid,
+                'X-Request-Token': request_token,
+            })
+
+        general = video['general']
+        title = general['title']
+
+        stream_data = video['streamdata']
+        language = general.get('language_raw') or ''
+
+        # TODO: reverse more cdns and formats
+
+        cdn = stream_data['cdnType']
+        assert cdn == 'azure'
+
+        azure_locator = stream_data['azureLocator']
+
+        AZURE_URL = 'http://nx-p%02d.akamaized.net/'
+
+        for secure in ('s', ''):
+            cdn_shield = stream_data.get('cdnShieldHTTP%s' % secure.upper())
+            if cdn_shield:
+                azure_base = 'http%s://%s' % (secure, cdn_shield)
+                break
+        else:
+            azure_base = AZURE_URL % int(stream_data['azureAccount'].replace('nexxplayplus', ''))
+
+        is_ml = ',' in language
+        azure_m3u8_url = '%s%s/%s_src%s.ism/Manifest(format=m3u8-aapl)' % (
+            azure_base, azure_locator, video_id, ('_manifest' if is_ml else ''))
+
+        protection_token = try_get(
+            video, lambda x: x['protectiondata']['token'], compat_str)
+        if protection_token:
+            azure_m3u8_url += '?hdnts=%s' % protection_token
+
+        formats = self._extract_m3u8_formats(
+            azure_m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
+            m3u8_id='%s-hls' % cdn)
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'alt_title': general.get('subtitle'),
+            'description': general.get('description'),
+            'release_year': int_or_none(general.get('year')),
+            'creator': general.get('studio') or general.get('studio_adref'),
+            'thumbnail': try_get(
+                video, lambda x: x['imagedata']['thumb'], compat_str),
+            'duration': parse_duration(general.get('runtime')),
+            'timestamp': int_or_none(general.get('uploaded')),
+            'episode_number': int_or_none(try_get(
+                video, lambda x: x['episodedata']['episode'])),
+            'season_number': int_or_none(try_get(
+                video, lambda x: x['episodedata']['season'])),
+            'formats': formats,
+        }
+
+
+class NexxEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'http://embed.nexx.cloud/748/KC1614647Z27Y7T?autoplay=1',
+        'md5': '16746bfc28c42049492385c989b26c4a',
+        'info_dict': {
+            'id': '161464',
+            'ext': 'mp4',
+            'title': 'Nervenkitzel Achterbahn',
+            'alt_title': 'Karussellbauer in Deutschland',
+            'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
+            'release_year': 2005,
+            'creator': 'SPIEGEL TV',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2761,
+            'timestamp': 1394021479,
+            'upload_date': '20140305',
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
+    }
+
+    @staticmethod
+    def _extract_urls(webpage):
+        # Reference:
+        # 1. https://nx-s.akamaized.net/files/201510/44.pdf
+
+        # iFrame Embed Integration
+        return [mobj.group('url') for mobj in re.finditer(
+            r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?:(?!\1).)+)\1',
+            webpage)]
+
+    def _real_extract(self, url):
+        embed_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, embed_id)
+
+        return self.url_result(NexxIE._extract_url(webpage), ie=NexxIE.ie_key())

youtube_dl/extractor/pearvideo.py

@@ -0,0 +1,63 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    qualities,
+    unified_timestamp,
+)
+
+
+class PearVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?pearvideo\.com/video_(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.pearvideo.com/video_1076290',
+        'info_dict': {
+            'id': '1076290',
+            'ext': 'mp4',
+            'title': '小浣熊在主人家玻璃上滚石头：没砸',
+            'description': 'md5:01d576b747de71be0ee85eb7cac25f9d',
+            'timestamp': 1494275280,
+            'upload_date': '20170508',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        quality = qualities(
+            ('ldflv', 'ld', 'sdflv', 'sd', 'hdflv', 'hd', 'src'))
+
+        formats = [{
+            'url': mobj.group('url'),
+            'format_id': mobj.group('id'),
+            'quality': quality(mobj.group('id')),
+        } for mobj in re.finditer(
+            r'(?P<id>[a-zA-Z]+)Url\s*=\s*(["\'])(?P<url>(?:https?:)?//.+?)\2',
+            webpage)]
+        self._sort_formats(formats)
+
+        title = self._search_regex(
+            (r'<h1[^>]+\bclass=(["\'])video-tt\1[^>]*>(?P<value>[^<]+)',
+             r'<[^>]+\bdata-title=(["\'])(?P<value>(?:(?!\1).)+)\1'),
+            webpage, 'title', group='value')
+        description = self._search_regex(
+            (r'<div[^>]+\bclass=(["\'])summary\1[^>]*>(?P<value>[^<]+)',
+             r'<[^>]+\bdata-summary=(["\'])(?P<value>(?:(?!\1).)+)\1'),
+            webpage, 'description', default=None,
+            group='value') or self._html_search_meta('Description', webpage)
+        timestamp = unified_timestamp(self._search_regex(
+            r'<div[^>]+\bclass=["\']date["\'][^>]*>([^<]+)',
+            webpage, 'timestamp', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'timestamp': timestamp,
+            'formats': formats,
+        }

youtube_dl/extractor/spiegel.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from .nexx import NexxEmbedIE
 from .spiegeltv import SpiegeltvIE
 from ..compat import compat_urlparse
 from ..utils import (
@@ -143,6 +144,9 @@ class SpiegelArticleIE(InfoExtractor):
         entries = [
             self.url_result(compat_urlparse.urljoin(
                 self.http_scheme() + '//spiegel.de/', embed_path))
-            for embed_path in embeds
+            for embed_path in embeds]
-        ]
+        if embeds:
-        return self.playlist_result(entries)
+            return self.playlist_result(entries)
+
+        return self.playlist_from_matches(
+            NexxEmbedIE._extract_urls(webpage), ie=NexxEmbedIE.ie_key())

youtube_dl/extractor/spiegeltv.py

@@ -1,114 +1,17 @@
-# coding: utf-8
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_urlparse
+from .nexx import NexxIE
-from ..utils import (
-    determine_ext,
-    float_or_none,
-)
 
 
 class SpiegeltvIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?spiegel\.tv/(?:#/)?filme/(?P<id>[\-a-z0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?spiegel\.tv/videos/(?P<id>\d+)'
-    _TESTS = [{
+    _TEST = {
-        'url': 'http://www.spiegel.tv/filme/flug-mh370/',
+        'url': 'http://www.spiegel.tv/videos/161681-flug-mh370/',
-        'info_dict': {
-            'id': 'flug-mh370',
-            'ext': 'm4v',
-            'title': 'Flug MH370',
-            'description': 'Das Rätsel um die Boeing 777 der Malaysia-Airlines',
-            'thumbnail': r're:http://.*\.jpg$',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        }
-    }, {
-        'url': 'http://www.spiegel.tv/#/filme/alleskino-die-wahrheit-ueber-maenner/',
         'only_matching': True,
-    }]
+    }
 
     def _real_extract(self, url):
-        if '/#/' in url:
+        return self.url_result(
-            url = url.replace('/#/', '/')
+            'https://api.nexx.cloud/v3/748/videos/byid/%s'
-        video_id = self._match_id(url)
+            % self._match_id(url), ie=NexxIE.ie_key())
-        webpage = self._download_webpage(url, video_id)
-        title = self._html_search_regex(r'<h1.*?>(.*?)</h1>', webpage, 'title')
-
-        apihost = 'http://spiegeltv-ivms2-restapi.s3.amazonaws.com'
-        version_json = self._download_json(
-            '%s/version.json' % apihost, video_id,
-            note='Downloading version information')
-        version_name = version_json['version_name']
-
-        slug_json = self._download_json(
-            '%s/%s/restapi/slugs/%s.json' % (apihost, version_name, video_id),
-            video_id,
-            note='Downloading object information')
-        oid = slug_json['object_id']
-
-        media_json = self._download_json(
-            '%s/%s/restapi/media/%s.json' % (apihost, version_name, oid),
-            video_id, note='Downloading media information')
-        uuid = media_json['uuid']
-        is_wide = media_json['is_wide']
-
-        server_json = self._download_json(
-            'http://spiegeltv-prod-static.s3.amazonaws.com/projectConfigs/projectConfig.json',
-            video_id, note='Downloading server information')
-
-        format = '16x9' if is_wide else '4x3'
-
-        formats = []
-        for streamingserver in server_json['streamingserver']:
-            endpoint = streamingserver.get('endpoint')
-            if not endpoint:
-                continue
-            play_path = 'mp4:%s_spiegeltv_0500_%s.m4v' % (uuid, format)
-            if endpoint.startswith('rtmp'):
-                formats.append({
-                    'url': endpoint,
-                    'format_id': 'rtmp',
-                    'app': compat_urllib_parse_urlparse(endpoint).path[1:],
-                    'play_path': play_path,
-                    'player_path': 'http://prod-static.spiegel.tv/frontend-076.swf',
-                    'ext': 'flv',
-                    'rtmp_live': True,
-                })
-            elif determine_ext(endpoint) == 'm3u8':
-                formats.append({
-                    'url': endpoint.replace('[video]', play_path),
-                    'ext': 'm4v',
-                    'format_id': 'hls',  # Prefer hls since it allows to workaround georestriction
-                    'protocol': 'm3u8',
-                    'preference': 1,
-                    'http_headers': {
-                        'Accept-Encoding': 'deflate',  # gzip causes trouble on the server side
-                    },
-                })
-            else:
-                formats.append({
-                    'url': endpoint,
-                })
-        self._check_formats(formats, video_id)
-
-        thumbnails = []
-        for image in media_json['images']:
-            thumbnails.append({
-                'url': image['url'],
-                'width': image['width'],
-                'height': image['height'],
-            })
-
-        description = media_json['subtitle']
-        duration = float_or_none(media_json.get('duration_in_ms'), scale=1000)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'duration': duration,
-            'thumbnails': thumbnails,
-            'formats': formats,
-        }

youtube_dl/extractor/youtube.py

@@ -1660,7 +1660,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     <a[^>]*
                         (?:
                             \bhref=["\']/red[^>]*>|             # drop possible
-                            >\s*Listen ad-free with YouTube Red # YouTube Red ad 
+                            >\s*Listen ad-free with YouTube Red # YouTube Red ad
                         )
                     .*?
                 )?</li

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2017.07.09'
+__version__ = '2017.07.15'