release 2018.02.11

they seem to be required now, else we get 403 errors.

.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 *2018.02.08*. 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 Issues with outdated version will be rejected.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 **2018.02.08**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2018.02.11*. 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 Issues with outdated version will be rejected.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 **2018.02.11**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through the [README](https://github.com/rg3/youtube-dl/blob/master/README.md), **most notably** the [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -36,7 +36,7 @@ Add the `-v` flag to **your command line** you run youtube-dl with (`youtube-dl
 [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 2018.02.08
+[debug] youtube-dl version 2018.02.11
 [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: {}

AUTHORS

@@ -233,3 +233,6 @@ Daniel Weber
 Kay Bouché
 Yang Hongbo
 Lei Wang
+Petr Novák
+Leonardo Taccari
+Martin Weinelt

ChangeLog

@@ -1,3 +1,25 @@
+version 2018.02.11
+
+Core
++ [YoutubeDL] Add support for filesize_approx in format selector (#15550)
+
+Extractors
++ [francetv] Add support for live streams (#13689)
++ [francetv] Add support for zouzous.fr and ludo.fr (#10454, #13087, #13103,
+  #15012)
+* [francetv] Separate main extractor and rework others to delegate to it
+* [francetv] Improve manifest URL signing (#15536)
++ [francetv] Sign m3u8 manifest URLs (#15565)
++ [veoh] Add support for embed URLs (#15561)
+* [afreecatv] Fix extraction (#15556)
+* [periscope] Use accessVideoPublic endpoint (#15554)
+* [discovery] Fix auth request (#15542)
++ [6play] Extract subtitles (#15541)
+* [newgrounds] Fix metadata extraction (#15531)
++ [nbc] Add support for stream.nbcolympics.com (#10295)
+* [dvtv] Fix live streams extraction (#15442)
+
+
 version 2018.02.08
 
 Extractors

docs/supportedsites.md

@@ -189,7 +189,7 @@
  - **CSpan**: C-SPAN
  - **CtsNews**: 華視新聞
  - **CTVNews**
- - **culturebox.francetvinfo.fr**
+ - **Culturebox**
  - **CultureUnplugged**
  - **curiositystream**
  - **curiositystream:collection**
@@ -291,6 +291,8 @@
  - **FranceTV**
  - **FranceTVEmbed**
  - **francetvinfo.fr**
+ - **FranceTVJeunesse**
+ - **FranceTVSite**
  - **Freesound**
  - **freespeech.org**
  - **FreshLive**
@@ -511,7 +513,8 @@
  - **NBA**
  - **NBC**
  - **NBCNews**
- - **NBCOlympics**
+ - **nbcolympics**
+ - **nbcolympics:stream**
  - **NBCSports**
  - **NBCSportsVPlayer**
  - **ndr**: NDR.de - Norddeutscher Rundfunk

youtube_dl/YoutubeDL.py

@@ -1033,7 +1033,7 @@ class YoutubeDL(object):
             '!=': operator.ne,
         }
         operator_rex = re.compile(r'''(?x)\s*
-            (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
+            (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
             \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
             (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
             $

youtube_dl/extractor/afreecatv.py

@@ -175,12 +175,23 @@ class AfreecaTVIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
+        webpage = self._download_webpage(url, video_id)
+
+        station_id = self._search_regex(
+            r'nStationNo\s*=\s*(\d+)', webpage, 'station')
+        bbs_id = self._search_regex(
+            r'nBbsNo\s*=\s*(\d+)', webpage, 'bbs')
+        video_id = self._search_regex(
+            r'nTitleNo\s*=\s*(\d+)', webpage, 'title', default=video_id)
+
         video_xml = self._download_xml(
             'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php',
             video_id, headers={
                 'Referer': 'http://vod.afreecatv.com/embed.php',
             }, query={
                 'nTitleNo': video_id,
+                'nStationNo': station_id,
+                'nBbsNo': bbs_id,
                 'partialView': 'SKIP_ADULT',
             })
 

youtube_dl/extractor/discovery.py

@@ -5,15 +5,16 @@ import re
 import string
 
 from .discoverygo import DiscoveryGoBaseIE
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
-    update_url_query,
+    try_get,
 )
 from ..compat import compat_HTTPError
 
 
 class DiscoveryIE(DiscoveryGoBaseIE):
-    _VALID_URL = r'''(?x)https?://(?:www\.)?(?:
+    _VALID_URL = r'''(?x)https?://(?:www\.)?(?P<site>
             discovery|
             investigationdiscovery|
             discoverylife|
@@ -44,7 +45,7 @@ class DiscoveryIE(DiscoveryGoBaseIE):
     _GEO_BYPASS = False
 
     def _real_extract(self, url):
-        path, display_id = re.match(self._VALID_URL, url).groups()
+        site, path, display_id = re.match(self._VALID_URL, url).groups()
         webpage = self._download_webpage(url, display_id)
 
         react_data = self._parse_json(self._search_regex(
@@ -55,14 +56,13 @@ class DiscoveryIE(DiscoveryGoBaseIE):
         video_id = video['id']
 
         access_token = self._download_json(
-            'https://www.discovery.com/anonymous', display_id, query={
-                'authLink': update_url_query(
-                    'https://login.discovery.com/v1/oauth2/authorize', {
-                        'client_id': react_data['application']['apiClientId'],
-                        'redirect_uri': 'https://fusion.ddmcdn.com/app/mercury-sdk/180/redirectHandler.html',
-                        'response_type': 'anonymous',
-                        'state': 'nonce,' + ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
-                    })
+            'https://www.%s.com/anonymous' % site, display_id, query={
+                'authRel': 'authorization',
+                'client_id': try_get(
+                    react_data, lambda x: x['application']['apiClientId'],
+                    compat_str) or '3020a40c2356a645b4b4',
+                'nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
+                'redirectUri': 'https://fusion.ddmcdn.com/app/mercury-sdk/180/redirectHandler.html?https://www.%s.com' % site,
             })['access_token']
 
         try:

youtube_dl/extractor/dvtv.py

@@ -32,7 +32,7 @@ class DVTVIE(InfoExtractor):
     }, {
         'url': 'http://video.aktualne.cz/dvtv/dvtv-16-12-2014-utok-talibanu-boj-o-kliniku-uprchlici/r~973eb3bc854e11e498be002590604f2e/',
         'info_dict': {
-            'title': 'DVTV 16. 12. 2014: útok Talibanu, boj o kliniku, uprchlíci',
+            'title': r're:^DVTV 16\. 12\. 2014: útok Talibanu, boj o kliniku, uprchlíci',
             'id': '973eb3bc854e11e498be002590604f2e',
         },
         'playlist': [{
@@ -91,10 +91,24 @@ class DVTVIE(InfoExtractor):
     }, {
         'url': 'http://video.aktualne.cz/v-cechach-poprve-zazni-zelenkova-zrestaurovana-mse/r~45b4b00483ec11e4883b002590604f2e/',
         'only_matching': True,
+    }, {
+        'url': 'https://video.aktualne.cz/dvtv/babis-a-zeman-nesou-vinu-za-to-ze-nemame-jasno-v-tom-kdo-bud/r~026afb54fad711e79704ac1f6b220ee8/',
+        'md5': '87defe16681b1429c91f7a74809823c6',
+        'info_dict': {
+            'id': 'f5ae72f6fad611e794dbac1f6b220ee8',
+            'ext': 'mp4',
+            'title': 'Babiš a Zeman nesou vinu za to, že nemáme jasno v tom, kdo bude vládnout, říká Pekarová Adamová',
+        },
+        'params': {
+            'skip_download': True,
+        },
     }]
 
-    def _parse_video_metadata(self, js, video_id):
+    def _parse_video_metadata(self, js, video_id, live_js=None):
         data = self._parse_json(js, video_id, transform_source=js_to_json)
+        if live_js:
+            data.update(self._parse_json(
+                live_js, video_id, transform_source=js_to_json))
 
         title = unescapeHTML(data['title'])
 
@@ -142,13 +156,18 @@ class DVTVIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
+        # live content
+        live_item = self._search_regex(
+            r'(?s)embedData[0-9a-f]{32}\.asset\.liveStarter\s*=\s*(\{.+?\});',
+            webpage, 'video', default=None)
+
         # single video
         item = self._search_regex(
             r'(?s)embedData[0-9a-f]{32}\[["\']asset["\']\]\s*=\s*(\{.+?\});',
-            webpage, 'video', default=None, fatal=False)
+            webpage, 'video', default=None)
 
         if item:
-            return self._parse_video_metadata(item, video_id)
+            return self._parse_video_metadata(item, video_id, live_item)
 
         # playlist
         items = re.findall(

youtube_dl/extractor/extractors.py

@@ -373,8 +373,10 @@ from .franceculture import FranceCultureIE
 from .franceinter import FranceInterIE
 from .francetv import (
     FranceTVIE,
+    FranceTVSiteIE,
     FranceTVEmbedIE,
     FranceTVInfoIE,
+    FranceTVJeunesseIE,
     GenerationWhatIE,
     CultureboxIE,
 )
@@ -647,6 +649,7 @@ from .nbc import (
     NBCIE,
     NBCNewsIE,
     NBCOlympicsIE,
+    NBCOlympicsStreamIE,
     NBCSportsIE,
     NBCSportsVPlayerIE,
 )

youtube_dl/extractor/francetv.py

@@ -5,19 +5,93 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_str,
+    compat_urlparse,
+)
 from ..utils import (
     clean_html,
+    determine_ext,
     ExtractorError,
     int_or_none,
     parse_duration,
-    determine_ext,
+    try_get,
 )
 from .dailymotion import DailymotionIE
 
 
 class FranceTVBaseInfoExtractor(InfoExtractor):
+    def _make_url_result(self, video_or_full_id, catalog=None):
+        full_id = 'francetv:%s' % video_or_full_id
+        if '@' not in video_or_full_id and catalog:
+            full_id += '@%s' % catalog
+        return self.url_result(
+            full_id, ie=FranceTVIE.ie_key(),
+            video_id=video_or_full_id.split('@')[0])
+
+
+class FranceTVIE(InfoExtractor):
+    _VALID_URL = r'''(?x)
+                    (?:
+                        https?://
+                            sivideo\.webservices\.francetelevisions\.fr/tools/getInfosOeuvre/v2/\?
+                            .*?\bidDiffusion=[^&]+|
+                        (?:
+                            https?://videos\.francetv\.fr/video/|
+                            francetv:
+                        )
+                        (?P<id>[^@]+)(?:@(?P<catalog>.+))?
+                    )
+                    '''
+
+    _TESTS = [{
+        # without catalog
+        'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=162311093&callback=_jsonp_loader_callback_request_0',
+        'md5': 'c2248a8de38c4e65ea8fae7b5df2d84f',
+        'info_dict': {
+            'id': '162311093',
+            'ext': 'mp4',
+            'title': '13h15, le dimanche... - Les mystères de Jésus',
+            'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
+            'timestamp': 1502623500,
+            'upload_date': '20170813',
+        },
+    }, {
+        # with catalog
+        'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=NI_1004933&catalogue=Zouzous&callback=_jsonp_loader_callback_request_4',
+        'only_matching': True,
+    }, {
+        'url': 'http://videos.francetv.fr/video/NI_657393@Regions',
+        'only_matching': True,
+    }, {
+        'url': 'francetv:162311093',
+        'only_matching': True,
+    }, {
+        'url': 'francetv:NI_1004933@Zouzous',
+        'only_matching': True,
+    }, {
+        'url': 'francetv:NI_983319@Info-web',
+        'only_matching': True,
+    }, {
+        'url': 'francetv:NI_983319',
+        'only_matching': True,
+    }, {
+        'url': 'francetv:NI_657393@Regions',
+        'only_matching': True,
+    }, {
+        # france-3 live
+        'url': 'https://www.france.tv/france-3/direct.html',
+        'only_matching': True,
+    }, {
+        # france-3 live
+        'url': 'francetv:SIM_France3',
+        'only_matching': True,
+    }]
+
     def _extract_video(self, video_id, catalogue=None):
+        # Videos are identified by idDiffusion so catalogue part is optional.
+        # However when provided, some extra formats may be returned so we pass
+        # it if available.
         info = self._download_json(
             'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/',
             video_id, 'Downloading video JSON', query={
@@ -27,7 +101,8 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
 
         if info.get('status') == 'NOK':
             raise ExtractorError(
-                '%s returned error: %s' % (self.IE_NAME, info['message']), expected=True)
+                '%s returned error: %s' % (self.IE_NAME, info['message']),
+                expected=True)
         allowed_countries = info['videos'][0].get('geoblocage')
         if allowed_countries:
             georestricted = True
@@ -42,6 +117,21 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
         else:
             georestricted = False
 
+        def sign(manifest_url, manifest_id):
+            for host in ('hdfauthftv-a.akamaihd.net', 'hdfauth.francetv.fr'):
+                signed_url = self._download_webpage(
+                    'https://%s/esi/TA' % host, video_id,
+                    'Downloading signed %s manifest URL' % manifest_id,
+                    fatal=False, query={
+                        'url': manifest_url,
+                    })
+                if (signed_url and isinstance(signed_url, compat_str) and
+                        re.search(r'^(?:https?:)?//', signed_url)):
+                    return signed_url
+            return manifest_url
+
+        is_live = None
+
         formats = []
         for video in info['videos']:
             if video['statut'] != 'ONLINE':
@@ -49,6 +139,10 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
             video_url = video['url']
             if not video_url:
                 continue
+            if is_live is None:
+                is_live = (try_get(
+                    video, lambda x: x['plages_ouverture'][0]['direct'],
+                    bool) is True) or '/live.francetv.fr/' in video_url
             format_id = video['format']
             ext = determine_ext(video_url)
             if ext == 'f4m':
@@ -56,17 +150,14 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
                     # See https://github.com/rg3/youtube-dl/issues/3963
                     # m3u8 urls work fine
                     continue
-                f4m_url = self._download_webpage(
-                    'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url,
-                    video_id, 'Downloading f4m manifest token', fatal=False)
-                if f4m_url:
-                    formats.extend(self._extract_f4m_formats(
-                        f4m_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
-                        video_id, f4m_id=format_id, fatal=False))
+                formats.extend(self._extract_f4m_formats(
+                    sign(video_url, format_id) + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
+                    video_id, f4m_id=format_id, fatal=False))
             elif ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    video_url, video_id, 'mp4', entry_protocol='m3u8_native',
-                    m3u8_id=format_id, fatal=False))
+                    sign(video_url, format_id), video_id, 'mp4',
+                    entry_protocol='m3u8_native', m3u8_id=format_id,
+                    fatal=False))
             elif video_url.startswith('rtmp'):
                 formats.append({
                     'url': video_url,
@@ -97,33 +188,48 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
 
         return {
             'id': video_id,
-            'title': title,
+            'title': self._live_title(title) if is_live else title,
             'description': clean_html(info['synopsis']),
             'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
             'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']),
             'timestamp': int_or_none(info['diffusion']['timestamp']),
+            'is_live': is_live,
             'formats': formats,
             'subtitles': subtitles,
         }
 
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        catalog = mobj.group('catalog')
 
-class FranceTVIE(FranceTVBaseInfoExtractor):
+        if not video_id:
+            qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
+            video_id = qs.get('idDiffusion', [None])[0]
+            catalog = qs.get('catalogue', [None])[0]
+            if not video_id:
+                raise ExtractorError('Invalid URL', expected=True)
+
+        return self._extract_video(video_id, catalog)
+
+
+class FranceTVSiteIE(FranceTVBaseInfoExtractor):
     _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P<id>[^/]+)\.html'
 
     _TESTS = [{
         'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
         'info_dict': {
-            'id': '157550144',
+            'id': '162311093',
             'ext': 'mp4',
             'title': '13h15, le dimanche... - Les mystères de Jésus',
             'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
-            'timestamp': 1494156300,
-            'upload_date': '20170507',
+            'timestamp': 1502623500,
+            'upload_date': '20170813',
         },
         'params': {
-            # m3u8 downloads
             'skip_download': True,
         },
+        'add_ie': [FranceTVIE.ie_key()],
     }, {
         # france3
         'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
@@ -172,13 +278,14 @@ class FranceTVIE(FranceTVBaseInfoExtractor):
             video_id, catalogue = self._html_search_regex(
                 r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
                 webpage, 'video ID').split('@')
-        return self._extract_video(video_id, catalogue)
+
+        return self._make_url_result(video_id, catalogue)
 
 
 class FranceTVEmbedIE(FranceTVBaseInfoExtractor):
     _VALID_URL = r'https?://embed\.francetv\.fr/*\?.*?\bue=(?P<id>[^&]+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://embed.francetv.fr/?ue=7fd581a2ccf59d2fc5719c5c13cf6961',
         'info_dict': {
             'id': 'NI_983319',
@@ -188,7 +295,11 @@ class FranceTVEmbedIE(FranceTVBaseInfoExtractor):
             'timestamp': 1493981780,
             'duration': 16,
         },
-    }
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': [FranceTVIE.ie_key()],
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -197,12 +308,12 @@ class FranceTVEmbedIE(FranceTVBaseInfoExtractor):
             'http://api-embed.webservices.francetelevisions.fr/key/%s' % video_id,
             video_id)
 
-        return self._extract_video(video['video_id'], video.get('catalog'))
+        return self._make_url_result(video['video_id'], video.get('catalog'))
 
 
 class FranceTVInfoIE(FranceTVBaseInfoExtractor):
     IE_NAME = 'francetvinfo.fr'
-    _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<title>[^/?#&.]+)'
+    _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&.]+)'
 
     _TESTS = [{
         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
@@ -217,51 +328,18 @@ class FranceTVInfoIE(FranceTVBaseInfoExtractor):
             },
         },
         'params': {
-            # m3u8 downloads
             'skip_download': True,
         },
+        'add_ie': [FranceTVIE.ie_key()],
     }, {
         'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
-        'info_dict': {
-            'id': 'EV_20019',
-            'ext': 'mp4',
-            'title': 'Débat des candidats à la Commission européenne',
-            'description': 'Débat des candidats à la Commission européenne',
-        },
-        'params': {
-            'skip_download': 'HLS (reqires ffmpeg)'
-        },
-        'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.',
+        'only_matching': True,
     }, {
         'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
-        'md5': 'f485bda6e185e7d15dbc69b72bae993e',
-        'info_dict': {
-            'id': 'NI_173343',
-            'ext': 'mp4',
-            'title': 'Les entreprises familiales : le secret de la réussite',
-            'thumbnail': r're:^https?://.*\.jpe?g$',
-            'timestamp': 1433273139,
-            'upload_date': '20150602',
-        },
-        'params': {
-            # m3u8 downloads
-            'skip_download': True,
-        },
+        'only_matching': True,
     }, {
         'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
-        'md5': 'f485bda6e185e7d15dbc69b72bae993e',
-        'info_dict': {
-            'id': 'NI_657393',
-            'ext': 'mp4',
-            'title': 'Olivier Monthus, réalisateur de "Bretagne, le choix de l’Armor"',
-            'description': 'md5:a3264114c9d29aeca11ced113c37b16c',
-            'thumbnail': r're:^https?://.*\.jpe?g$',
-            'timestamp': 1458300695,
-            'upload_date': '20160318',
-        },
-        'params': {
-            'skip_download': True,
-        },
+        'only_matching': True,
     }, {
         # Dailymotion embed
         'url': 'http://www.francetvinfo.fr/politique/notre-dame-des-landes/video-sur-france-inter-cecile-duflot-denonce-le-regard-meprisant-de-patrick-cohen_1520091.html',
@@ -283,9 +361,9 @@ class FranceTVInfoIE(FranceTVBaseInfoExtractor):
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        page_title = mobj.group('title')
-        webpage = self._download_webpage(url, page_title)
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
 
         dailymotion_urls = DailymotionIE._extract_urls(webpage)
         if dailymotion_urls:
@@ -297,12 +375,13 @@ class FranceTVInfoIE(FranceTVBaseInfoExtractor):
             (r'id-video=([^@]+@[^"]+)',
              r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"'),
             webpage, 'video id').split('@')
-        return self._extract_video(video_id, catalogue)
+
+        return self._make_url_result(video_id, catalogue)
 
 
 class GenerationWhatIE(InfoExtractor):
     IE_NAME = 'france2.fr:generation-what'
-    _VALID_URL = r'https?://generation-what\.francetv\.fr/[^/]+/video/(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://generation-what\.francetv\.fr/[^/]+/video/(?P<id>[^/?#&]+)'
 
     _TESTS = [{
         'url': 'http://generation-what.francetv.fr/portrait/video/present-arms',
@@ -314,6 +393,10 @@ class GenerationWhatIE(InfoExtractor):
             'uploader_id': 'UCHH9p1eetWCgt4kXBYCb3_w',
             'upload_date': '20160411',
         },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': ['Youtube'],
     }, {
         'url': 'http://generation-what.francetv.fr/europe/video/present-arms',
         'only_matching': True,
@@ -321,42 +404,87 @@ class GenerationWhatIE(InfoExtractor):
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
+
         webpage = self._download_webpage(url, display_id)
+
         youtube_id = self._search_regex(
             r"window\.videoURL\s*=\s*'([0-9A-Za-z_-]{11})';",
             webpage, 'youtube id')
-        return self.url_result(youtube_id, 'Youtube', youtube_id)
+
+        return self.url_result(youtube_id, ie='Youtube', video_id=youtube_id)
 
 
 class CultureboxIE(FranceTVBaseInfoExtractor):
-    IE_NAME = 'culturebox.francetvinfo.fr'
-    _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)'
+    _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
-    _TEST = {
-        'url': 'http://culturebox.francetvinfo.fr/live/musique/musique-classique/le-livre-vermeil-de-montserrat-a-la-cathedrale-delne-214511',
-        'md5': '9b88dc156781c4dbebd4c3e066e0b1d6',
+    _TESTS = [{
+        'url': 'https://culturebox.francetvinfo.fr/opera-classique/musique-classique/c-est-baroque/concerts/cantates-bwv-4-106-et-131-de-bach-par-raphael-pichon-57-268689',
         'info_dict': {
-            'id': 'EV_50111',
-            'ext': 'flv',
-            'title': "Le Livre Vermeil de Montserrat à la Cathédrale d'Elne",
-            'description': 'md5:f8a4ad202e8fe533e2c493cc12e739d9',
-            'upload_date': '20150320',
-            'timestamp': 1426892400,
-            'duration': 2760.9,
+            'id': 'EV_134885',
+            'ext': 'mp4',
+            'title': 'Cantates BWV 4, 106 et 131 de Bach par Raphaël Pichon 5/7',
+            'description': 'md5:19c44af004b88219f4daa50fa9a351d4',
+            'upload_date': '20180206',
+            'timestamp': 1517945220,
+            'duration': 5981,
         },
-    }
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': [FranceTVIE.ie_key()],
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        name = mobj.group('name')
+        display_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, name)
+        webpage = self._download_webpage(url, display_id)
 
         if ">Ce live n'est plus disponible en replay<" in webpage:
-            raise ExtractorError('Video %s is not available' % name, expected=True)
+            raise ExtractorError(
+                'Video %s is not available' % display_id, expected=True)
 
         video_id, catalogue = self._search_regex(
             r'["\'>]https?://videos\.francetv\.fr/video/([^@]+@.+?)["\'<]',
             webpage, 'video id').split('@')
 
-        return self._extract_video(video_id, catalogue)
+        return self._make_url_result(video_id, catalogue)
+
+
+class FranceTVJeunesseIE(FranceTVBaseInfoExtractor):
+    _VALID_URL = r'(?P<url>https?://(?:www\.)?(?:zouzous|ludo)\.fr/heros/(?P<id>[^/?#&]+))'
+
+    _TESTS = [{
+        'url': 'https://www.zouzous.fr/heros/simon',
+        'info_dict': {
+            'id': 'simon',
+        },
+        'playlist_count': 9,
+    }, {
+        'url': 'https://www.ludo.fr/heros/ninjago',
+        'info_dict': {
+            'id': 'ninjago',
+        },
+        'playlist_count': 10,
+    }, {
+        'url': 'https://www.zouzous.fr/heros/simon?abc',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+
+        playlist = self._download_json(
+            '%s/%s' % (mobj.group('url'), 'playlist'), playlist_id)
+
+        if not playlist.get('count'):
+            raise ExtractorError(
+                '%s is not available' % playlist_id, expected=True)
+
+        entries = []
+        for item in playlist['items']:
+            identity = item.get('identity')
+            if identity and isinstance(identity, compat_str):
+                entries.append(self._make_url_result(identity))
+
+        return self.playlist_result(entries, playlist_id)

youtube_dl/extractor/nbc.py

@@ -1,6 +1,7 @@
 from __future__ import unicode_literals
 
 import re
+import base64
 
 from .common import InfoExtractor
 from .theplatform import ThePlatformIE
@@ -358,6 +359,7 @@ class NBCNewsIE(ThePlatformIE):
 
 
 class NBCOlympicsIE(InfoExtractor):
+    IE_NAME = 'nbcolympics'
     _VALID_URL = r'https?://www\.nbcolympics\.com/video/(?P<id>[a-z-]+)'
 
     _TEST = {
@@ -395,3 +397,54 @@ class NBCOlympicsIE(InfoExtractor):
             'ie_key': ThePlatformIE.ie_key(),
             'display_id': display_id,
         }
+
+
+class NBCOlympicsStreamIE(AdobePassIE):
+    IE_NAME = 'nbcolympics:stream'
+    _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
+    _TEST = {
+        'url': 'http://stream.nbcolympics.com/2018-winter-olympics-nbcsn-evening-feb-8',
+        'info_dict': {
+            'id': '203493',
+            'ext': 'mp4',
+            'title': 're:Curling, Alpine, Luge [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+    _DATA_URL_TEMPLATE = 'http://stream.nbcolympics.com/data/%s_%s.json'
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
+        resource = self._search_regex(
+            r"resource\s*=\s*'(.+)';", webpage,
+            'resource').replace("' + pid + '", pid)
+        event_config = self._download_json(
+            self._DATA_URL_TEMPLATE % ('event_config', pid),
+            pid)['eventConfig']
+        title = self._live_title(event_config['eventTitle'])
+        source_url = self._download_json(
+            self._DATA_URL_TEMPLATE % ('live_sources', pid),
+            pid)['videoSources'][0]['sourceUrl']
+        media_token = self._extract_mvpd_auth(
+            url, pid, event_config.get('requestorId', 'NBCOlympics'), resource)
+        formats = self._extract_m3u8_formats(self._download_webpage(
+            'http://sp.auth.adobe.com/tvs/v1/sign', pid, query={
+                'cdn': 'akamai',
+                'mediaToken': base64.b64encode(media_token.encode()),
+                'resource': base64.b64encode(resource.encode()),
+                'url': source_url,
+            }), pid, 'mp4')
+        self._sort_formats(formats)
+
+        return {
+            'id': pid,
+            'display_id': display_id,
+            'title': title,
+            'formats': formats,
+            'is_live': True,
+        }

youtube_dl/extractor/newgrounds.py

@@ -87,19 +87,21 @@ class NewgroundsIE(InfoExtractor):
         self._check_formats(formats, media_id)
         self._sort_formats(formats)
 
-        uploader = self._search_regex(
-            r'(?:Author|Writer)\s*<a[^>]+>([^<]+)', webpage, 'uploader',
+        uploader = self._html_search_regex(
+            (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
+             r'(?:Author|Writer)\s*<a[^>]+>([^<]+)'), webpage, 'uploader',
             fatal=False)
 
-        timestamp = unified_timestamp(self._search_regex(
-            r'<dt>Uploaded</dt>\s*<dd>([^<]+)', webpage, 'timestamp',
+        timestamp = unified_timestamp(self._html_search_regex(
+            (r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+</dd>\s*<dd>[^<]+)',
+             r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+)'), webpage, 'timestamp',
             default=None))
         duration = parse_duration(self._search_regex(
-            r'<dd>Song\s*</dd><dd>.+?</dd><dd>([^<]+)', webpage, 'duration',
-            default=None))
+            r'(?s)<dd>\s*Song\s*</dd>\s*<dd>.+?</dd>\s*<dd>([^<]+)', webpage,
+            'duration', default=None))
 
         filesize_approx = parse_filesize(self._html_search_regex(
-            r'<dd>Song\s*</dd><dd>(.+?)</dd>', webpage, 'filesize',
+            r'(?s)<dd>\s*Song\s*</dd>\s*<dd>(.+?)</dd>', webpage, 'filesize',
             default=None))
         if len(formats) == 1:
             formats[0]['filesize_approx'] = filesize_approx

youtube_dl/extractor/periscope.py

@@ -56,18 +56,16 @@ class PeriscopeIE(PeriscopeBaseIE):
     def _real_extract(self, url):
         token = self._match_id(url)
 
-        broadcast_data = self._call_api(
-            'getBroadcastPublic', {'broadcast_id': token}, token)
-        broadcast = broadcast_data['broadcast']
-        status = broadcast['status']
+        stream = self._call_api(
+            'accessVideoPublic', {'broadcast_id': token}, token)
 
-        user = broadcast_data.get('user', {})
+        broadcast = stream['broadcast']
+        title = broadcast['status']
 
-        uploader = broadcast.get('user_display_name') or user.get('display_name')
-        uploader_id = (broadcast.get('username') or user.get('username') or
-                       broadcast.get('user_id') or user.get('id'))
+        uploader = broadcast.get('user_display_name') or broadcast.get('username')
+        uploader_id = (broadcast.get('user_id') or broadcast.get('username'))
 
-        title = '%s - %s' % (uploader, status) if uploader else status
+        title = '%s - %s' % (uploader, title) if uploader else title
         state = broadcast.get('state').lower()
         if state == 'running':
             title = self._live_title(title)
@@ -77,9 +75,6 @@ class PeriscopeIE(PeriscopeBaseIE):
             'url': broadcast[image],
         } for image in ('image_url', 'image_url_small') if broadcast.get(image)]
 
-        stream = self._call_api(
-            'getAccessPublic', {'broadcast_id': token}, token)
-
         video_urls = set()
         formats = []
         for format_id in ('replay', 'rtmp', 'hls', 'https_hls', 'lhls', 'lhlsweb'):

youtube_dl/extractor/sixplay.py

@@ -52,6 +52,7 @@ class SixPlayIE(InfoExtractor):
         urls = []
         quality_key = qualities(['lq', 'sd', 'hq', 'hd'])
         formats = []
+        subtitles = {}
         for asset in clip_data['assets']:
             asset_url = asset.get('full_physical_path')
             protocol = asset.get('protocol')
@@ -60,6 +61,9 @@ class SixPlayIE(InfoExtractor):
             urls.append(asset_url)
             container = asset.get('video_container')
             ext = determine_ext(asset_url)
+            if protocol == 'http_subtitle' or ext == 'vtt':
+                subtitles.setdefault('fr', []).append({'url': asset_url})
+                continue
             if container == 'm3u8' or ext == 'm3u8':
                 if protocol == 'usp' and not compat_parse_qs(compat_urllib_parse_urlparse(asset_url).query).get('token', [None])[0]:
                     asset_url = re.sub(r'/([^/]+)\.ism/[^/]*\.m3u8', r'/\1.ism/\1.m3u8', asset_url)
@@ -102,4 +106,5 @@ class SixPlayIE(InfoExtractor):
             'duration': int_or_none(clip_data.get('duration')),
             'series': get(lambda x: x['program']['title']),
             'formats': formats,
+            'subtitles': subtitles,
         }

youtube_dl/extractor/veoh.py

@@ -12,7 +12,7 @@ from ..utils import (
 
 
 class VeohIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?veoh\.com/(?:watch|iphone/#_Watch)/(?P<id>(?:v|e|yapi-)[\da-zA-Z]+)'
+    _VALID_URL = r'https?://(?:www\.)?veoh\.com/(?:watch|embed|iphone/#_Watch)/(?P<id>(?:v|e|yapi-)[\da-zA-Z]+)'
 
     _TESTS = [{
         'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3',
@@ -24,6 +24,9 @@ class VeohIE(InfoExtractor):
             'uploader': 'LUMOback',
             'description': 'At LUMOback, we believe straight backs are stronger.  The LUMOback Posture & Movement Sensor:  It gently vibrates when you slouch, inspiring improved posture and mobility.  Use the app to track your data and improve your posture over time. ',
         },
+    }, {
+        'url': 'http://www.veoh.com/embed/v56314296nk7Zdmz3',
+        'only_matching': True,
     }, {
         'url': 'http://www.veoh.com/watch/v27701988pbTc4wzN?h1=Chile+workers+cover+up+to+avoid+skin+damage',
         'md5': '4a6ff84b87d536a6a71e6aa6c0ad07fa',

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2018.02.08'
+__version__ = '2018.02.11'