Merge remote-tracking branch 'upstream/master'

test/test_utils.py

@@ -519,6 +519,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_age_limit('PG-13'), 13)
         self.assertEqual(parse_age_limit('TV-14'), 14)
         self.assertEqual(parse_age_limit('TV-MA'), 17)
+        self.assertEqual(parse_age_limit('TV14'), 14)
+        self.assertEqual(parse_age_limit('TV_G'), 0)
 
     def test_parse_duration(self):
         self.assertEqual(parse_duration(None), None)

youtube_dl/extractor/dvtv.py

@@ -91,17 +91,6 @@ 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, live_js=None):

youtube_dl/extractor/extractors.py

@@ -469,10 +469,7 @@ from .imgur import (
 )
 from .ina import InaIE
 from .inc import IncIE
-from .indavideo import (
+from .indavideo import IndavideoEmbedIE
-    IndavideoIE,
-    IndavideoEmbedIE,
-)
 from .infoq import InfoQIE
 from .instagram import InstagramIE, InstagramUserIE
 from .internazionale import InternazionaleIE
@@ -666,6 +663,7 @@ from .nbc import (
     NBCOlympicsIE,
     NBCOlympicsStreamIE,
     NBCSportsIE,
+    NBCSportsStreamIE,
     NBCSportsVPlayerIE,
 )
 from .ndr import (
@@ -810,6 +808,7 @@ from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE
 from .pearvideo import PearVideoIE
+from .peertube import PeerTubeIE
 from .people import PeopleIE
 from .performgroup import PerformGroupIE
 from .periscope import (

youtube_dl/extractor/generic.py

@@ -108,6 +108,8 @@ from .yapfiles import YapFilesIE
 from .vice import ViceIE
 from .xfileshare import XFileShareIE
 from .cloudflarestream import CloudflareStreamIE
+from .peertube import PeerTubeIE
+from .indavideo import IndavideoEmbedIE
 
 
 class GenericIE(InfoExtractor):
@@ -2012,6 +2014,33 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            # PeerTube embed
+            'url': 'https://joinpeertube.org/fr/home/',
+            'info_dict': {
+                'id': 'home',
+                'title': 'Reprenez le contrôle de vos vidéos ! #JoinPeertube',
+            },
+            'playlist_count': 2,
+        },
+        {
+            # Indavideo embed
+            'url': 'https://streetkitchen.hu/receptek/igy_kell_otthon_hamburgert_sutni/',
+            'info_dict': {
+                'id': '1693903',
+                'ext': 'mp4',
+                'title': 'Így kell otthon hamburgert sütni',
+                'description': 'md5:f5a730ecf900a5c852e1e00540bbb0f7',
+                'timestamp': 1426330212,
+                'upload_date': '20150314',
+                'uploader': 'StreetKitchen',
+                'uploader_id': '546363',
+            },
+            'add_ie': [IndavideoEmbedIE.ie_key()],
+            'params': {
+                'skip_download': True,
+            },
+        },
         {
             'url': 'http://share-videos.se/auto/video/83645793?uid=13',
             'md5': 'b68d276de422ab07ee1d49388103f457',
@@ -3029,6 +3058,16 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 cloudflarestream_urls, video_id, video_title, ie=CloudflareStreamIE.ie_key())
 
+        peertube_urls = PeerTubeIE._extract_urls(webpage)
+        if peertube_urls:
+            return self.playlist_from_matches(
+                peertube_urls, video_id, video_title, ie=PeerTubeIE.ie_key())
+
+        indavideo_urls = IndavideoEmbedIE._extract_urls(webpage)
+        if indavideo_urls:
+            return self.playlist_from_matches(
+                indavideo_urls, video_id, video_title, ie=IndavideoEmbedIE.ie_key())
+
         sharevideos_urls = [mobj.group('url') for mobj in re.finditer(
             r'<iframe[^>]+?\bsrc\s*=\s*(["\'])(?P<url>(?:https?:)?//embed\.share-videos\.se/auto/embed/\d+\?.*?\buid=\d+.*?)\1',
             webpage)]

youtube_dl/extractor/globo.py

@@ -126,16 +126,16 @@ class GloboIE(InfoExtractor):
                 continue
 
             hash_code = security_hash[:2]
-            received_time = int(security_hash[2:12])
+            received_time = security_hash[2:12]
             received_random = security_hash[12:22]
             received_md5 = security_hash[22:]
 
-            sign_time = received_time + 86400
+            sign_time = compat_str(int(received_time) + 86400)
             padding = '%010d' % random.randint(1, 10000000000)
 
-            md5_data = (received_md5 + str(sign_time) + padding + '0xFF01DD').encode()
+            md5_data = (received_md5 + sign_time + padding + '0xFF01DD').encode()
             signed_md5 = base64.urlsafe_b64encode(hashlib.md5(md5_data).digest()).decode().strip('=')
-            signed_hash = hash_code + compat_str(received_time) + received_random + compat_str(sign_time) + padding + signed_md5
+            signed_hash = hash_code + received_time + received_random + sign_time + padding + signed_md5
 
             signed_url = '%s?h=%s&k=%s' % (resource_url, signed_hash, 'flash')
             if resource_id.endswith('m3u8') or resource_url.endswith('.m3u8'):

youtube_dl/extractor/go90.py

@@ -6,7 +6,9 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     determine_ext,
+    ExtractorError,
     int_or_none,
+    parse_age_limit,
     parse_iso8601,
 )
 
@@ -23,6 +25,7 @@ class Go90IE(InfoExtractor):
             'description': 'VICE\'s Karley Sciortino meets with activists who discuss the state\'s strong anti-porn stance. Then, VICE Sports explains NFL contracts.',
             'timestamp': 1491868800,
             'upload_date': '20170411',
+            'age_limit': 14,
         }
     }
 
@@ -33,6 +36,8 @@ class Go90IE(InfoExtractor):
             video_id, headers={
                 'Content-Type': 'application/json; charset=utf-8',
             }, data=b'{"client":"web","device_type":"pc"}')
+        if video_data.get('requires_drm'):
+            raise ExtractorError('This video is DRM protected.', expected=True)
         main_video_asset = video_data['main_video_asset']
 
         episode_number = int_or_none(video_data.get('episode_number'))
@@ -123,4 +128,5 @@ class Go90IE(InfoExtractor):
             'season_number': season_number,
             'episode_number': episode_number,
             'subtitles': subtitles,
+            'age_limit': parse_age_limit(video_data.get('rating')),
         }

youtube_dl/extractor/hidive.py

@@ -17,6 +17,9 @@ class HiDiveIE(InfoExtractor):
     # Using X-Forwarded-For results in 403 HTTP error for HLS fragments,
     # so disabling geo bypass completely
     _GEO_BYPASS = False
+    _NETRC_MACHINE = 'hidive'
+    _LOGGED_IN = False
+    _LOGIN_URL = 'https://www.hidive.com/account/login'
 
     _TESTS = [{
         'url': 'https://www.hidive.com/stream/the-comic-artist-and-his-assistants/s01e001',
@@ -31,8 +34,30 @@ class HiDiveIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+        'skip': 'Requires Authentication',
     }]
 
+    def _real_initialize(self):
+        if self._LOGGED_IN:
+            return
+
+        (email, password) = self._get_login_info()
+        if email is None:
+            return
+
+        webpage = self._download_webpage(self._LOGIN_URL, None)
+        form = self._search_regex(
+            r'(?s)<form[^>]+action="/account/login"[^>]*>(.+?)</form>',
+            webpage, 'login form')
+        data = self._hidden_inputs(form)
+        data.update({
+            'Email': email,
+            'Password': password,
+        })
+        self._download_webpage(
+            self._LOGIN_URL, None, 'Logging in', data=urlencode_postdata(data))
+        self._LOGGED_IN = True
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         title, key = mobj.group('title', 'key')
@@ -43,6 +68,7 @@ class HiDiveIE(InfoExtractor):
             data=urlencode_postdata({
                 'Title': title,
                 'Key': key,
+                'PlayerId': 'f4f895ce1ca713ba263b91caeb1daa2d08904783',
             }))
 
         restriction = settings.get('restrictionReason')
@@ -79,6 +105,7 @@ class HiDiveIE(InfoExtractor):
                 subtitles.setdefault(cc_lang, []).append({
                     'url': cc_url,
                 })
+        self._sort_formats(formats)
 
         season_number = int_or_none(self._search_regex(
             r's(\d+)', key, 'season number', default=None))

youtube_dl/extractor/imgur.py

@@ -3,7 +3,6 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
 from ..utils import (
     int_or_none,
     js_to_json,
@@ -21,7 +20,7 @@ class ImgurIE(InfoExtractor):
             'id': 'A61SaA1',
             'ext': 'mp4',
             'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
-            'description': 'Imgur: The most awesome images on the Internet.',
+            'description': 'Imgur: The magic of the Internet',
         },
     }, {
         'url': 'https://imgur.com/A61SaA1',
@@ -29,7 +28,7 @@ class ImgurIE(InfoExtractor):
             'id': 'A61SaA1',
             'ext': 'mp4',
             'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
-            'description': 'Imgur: The most awesome images on the Internet.',
+            'description': 'Imgur: The magic of the Internet',
         },
     }, {
         'url': 'https://imgur.com/gallery/YcAQlkx',
@@ -37,8 +36,6 @@ class ImgurIE(InfoExtractor):
             'id': 'YcAQlkx',
             'ext': 'mp4',
             'title': 'Classic Steve Carell gif...cracks me up everytime....damn the repost downvotes....',
-            'description': 'Imgur: The most awesome images on the Internet.'
-
         }
     }, {
         'url': 'http://imgur.com/topic/Funny/N8rOudd',
@@ -50,8 +47,8 @@ class ImgurIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(
+        gifv_url = 'https://i.imgur.com/{id}.gifv'.format(id=video_id)
-            compat_urlparse.urljoin(url, video_id), video_id)
+        webpage = self._download_webpage(gifv_url, video_id)
 
         width = int_or_none(self._og_search_property(
             'video:width', webpage, default=None))
@@ -107,7 +104,7 @@ class ImgurIE(InfoExtractor):
         return {
             'id': video_id,
             'formats': formats,
-            'description': self._og_search_description(webpage),
+            'description': self._og_search_description(webpage, default=None),
             'title': self._og_search_title(webpage),
         }
 

youtube_dl/extractor/indavideo.py

@@ -1,11 +1,15 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     int_or_none,
     parse_age_limit,
     parse_iso8601,
+    update_url_query,
 )
 
 
@@ -13,7 +17,7 @@ class IndavideoEmbedIE(InfoExtractor):
     _VALID_URL = r'https?://(?:(?:embed\.)?indavideo\.hu/player/video/|assets\.indavideo\.hu/swf/player\.swf\?.*\b(?:v(?:ID|id))=)(?P<id>[\da-f]+)'
     _TESTS = [{
         'url': 'http://indavideo.hu/player/video/1bdc3c6d80/',
-        'md5': 'f79b009c66194acacd40712a6778acfa',
+        'md5': 'c8a507a1c7410685f83a06eaeeaafeab',
         'info_dict': {
             'id': '1837039',
             'ext': 'mp4',
@@ -36,6 +40,20 @@ class IndavideoEmbedIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    # Some example URLs covered by generic extractor:
+    #   http://indavideo.hu/video/Vicces_cica_1
+    #   http://index.indavideo.hu/video/2015_0728_beregszasz
+    #   http://auto.indavideo.hu/video/Sajat_utanfutoban_a_kis_tacsko
+    #   http://erotika.indavideo.hu/video/Amator_tini_punci
+    #   http://film.indavideo.hu/video/f_hrom_nagymamm_volt
+    #   http://palyazat.indavideo.hu/video/Embertelen_dal_Dodgem_egyuttes
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+\bsrc=["\'](?P<url>(?:https?:)?//embed\.indavideo\.hu/player/video/[\da-f]+)',
+            webpage)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -45,7 +63,14 @@ class IndavideoEmbedIE(InfoExtractor):
 
         title = video['title']
 
-        video_urls = video.get('video_files', [])
+        video_urls = []
+
+        video_files = video.get('video_files')
+        if isinstance(video_files, list):
+            video_urls.extend(video_files)
+        elif isinstance(video_files, dict):
+            video_urls.extend(video_files.values())
+
         video_file = video.get('video_file')
         if video:
             video_urls.append(video_file)
@@ -58,11 +83,23 @@ class IndavideoEmbedIE(InfoExtractor):
             if flv_url not in video_urls:
                 video_urls.append(flv_url)
 
-        formats = [{
+        filesh = video.get('filesh')
+
+        formats = []
+        for video_url in video_urls:
+            height = int_or_none(self._search_regex(
+                r'\.(\d{3,4})\.mp4(?:\?|$)', video_url, 'height', default=None))
+            if filesh:
+                if not height:
+                    continue
+                token = filesh.get(compat_str(height))
+                if token is None:
+                    continue
+                video_url = update_url_query(video_url, {'token': token})
+            formats.append({
                 'url': video_url,
-            'height': int_or_none(self._search_regex(
+                'height': height,
-                r'\.(\d{3,4})\.mp4(?:\?|$)', video_url, 'height', default=None)),
+            })
-        } for video_url in video_urls]
         self._sort_formats(formats)
 
         timestamp = video.get('date')
@@ -89,55 +126,3 @@ class IndavideoEmbedIE(InfoExtractor):
             'tags': tags,
             'formats': formats,
         }
-
-
-class IndavideoIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:.+?\.)?indavideo\.hu/video/(?P<id>[^/#?]+)'
-    _TESTS = [{
-        'url': 'http://indavideo.hu/video/Vicces_cica_1',
-        'md5': '8c82244ba85d2a2310275b318eb51eac',
-        'info_dict': {
-            'id': '1335611',
-            'display_id': 'Vicces_cica_1',
-            'ext': 'mp4',
-            'title': 'Vicces cica',
-            'description': 'Játszik a tablettel. :D',
-            'thumbnail': r're:^https?://.*\.jpg$',
-            'uploader': 'Jet_Pack',
-            'uploader_id': '491217',
-            'timestamp': 1390821212,
-            'upload_date': '20140127',
-            'duration': 7,
-            'age_limit': 0,
-            'tags': ['vicces', 'macska', 'cica', 'ügyes', 'nevetés', 'játszik', 'Cukiság', 'Jet_Pack'],
-        },
-    }, {
-        'url': 'http://index.indavideo.hu/video/2015_0728_beregszasz',
-        'only_matching': True,
-    }, {
-        'url': 'http://auto.indavideo.hu/video/Sajat_utanfutoban_a_kis_tacsko',
-        'only_matching': True,
-    }, {
-        'url': 'http://erotika.indavideo.hu/video/Amator_tini_punci',
-        'only_matching': True,
-    }, {
-        'url': 'http://film.indavideo.hu/video/f_hrom_nagymamm_volt',
-        'only_matching': True,
-    }, {
-        'url': 'http://palyazat.indavideo.hu/video/Embertelen_dal_Dodgem_egyuttes',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        display_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, display_id)
-        embed_url = self._search_regex(
-            r'<link[^>]+rel="video_src"[^>]+href="(.+?)"', webpage, 'embed url')
-
-        return {
-            '_type': 'url_transparent',
-            'ie_key': 'IndavideoEmbed',
-            'url': embed_url,
-            'display_id': display_id,
-        }

youtube_dl/extractor/izlesene.py

@@ -1,10 +1,11 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
+from ..compat import (
+    compat_str,
+    compat_urllib_parse_unquote,
+)
 from ..utils import (
     determine_ext,
     float_or_none,
@@ -57,12 +58,33 @@ class IzleseneIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        url = 'http://www.izlesene.com/video/%s' % video_id
+        webpage = self._download_webpage('http://www.izlesene.com/video/%s' % video_id, video_id)
-        webpage = self._download_webpage(url, video_id)
+
+        video = self._parse_json(
+            self._search_regex(
+                r'videoObj\s*=\s*({.+?})\s*;\s*\n', webpage, 'streams'),
+            video_id)
+
+        title = video.get('videoTitle') or self._og_search_title(webpage)
+
+        formats = []
+        for stream in video['media']['level']:
+            source_url = stream.get('source')
+            if not source_url or not isinstance(source_url, compat_str):
+                continue
+            ext = determine_ext(url, 'mp4')
+            quality = stream.get('value')
+            height = int_or_none(quality)
+            formats.append({
+                'format_id': '%sp' % quality if quality else 'sd',
+                'url': compat_urllib_parse_unquote(source_url),
+                'ext': ext,
+                'height': height,
+            })
+        self._sort_formats(formats)
 
-        title = self._og_search_title(webpage)
         description = self._og_search_description(webpage, default=None)
-        thumbnail = self._proto_relative_url(
+        thumbnail = video.get('posterURL') or self._proto_relative_url(
             self._og_search_thumbnail(webpage), scheme='http:')
 
         uploader = self._html_search_regex(
@@ -71,41 +93,15 @@ class IzleseneIE(InfoExtractor):
         timestamp = parse_iso8601(self._html_search_meta(
             'uploadDate', webpage, 'upload date'))
 
-        duration = float_or_none(self._html_search_regex(
+        duration = float_or_none(video.get('duration') or self._html_search_regex(
-            r'"videoduration"\s*:\s*"([^"]+)"',
+            r'videoduration["\']?\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
-            webpage, 'duration', fatal=False), scale=1000)
+            webpage, 'duration', fatal=False, group='value'), scale=1000)
 
         view_count = str_to_int(get_element_by_id('videoViewCount', webpage))
         comment_count = self._html_search_regex(
             r'comment_count\s*=\s*\'([^\']+)\';',
             webpage, 'comment_count', fatal=False)
 
-        content_url = self._html_search_meta(
-            'contentURL', webpage, 'content URL', fatal=False)
-        ext = determine_ext(content_url, 'mp4')
-
-        # Might be empty for some videos.
-        streams = self._html_search_regex(
-            r'"qualitylevel"\s*:\s*"([^"]+)"', webpage, 'streams', default='')
-
-        formats = []
-        if streams:
-            for stream in streams.split('|'):
-                quality, url = re.search(r'\[(\w+)\](.+)', stream).groups()
-                formats.append({
-                    'format_id': '%sp' % quality if quality else 'sd',
-                    'url': compat_urllib_parse_unquote(url),
-                    'ext': ext,
-                })
-        else:
-            stream_url = self._search_regex(
-                r'"streamurl"\s*:\s*"([^"]+)"', webpage, 'stream URL')
-            formats.append({
-                'format_id': 'sd',
-                'url': compat_urllib_parse_unquote(stream_url),
-                'ext': ext,
-            })
-
         return {
             'id': video_id,
             'title': title,

youtube_dl/extractor/nbc.py

@@ -1,7 +1,8 @@
 from __future__ import unicode_literals
 
-import re
 import base64
+import json
+import re
 
 from .common import InfoExtractor
 from .theplatform import ThePlatformIE
@@ -175,6 +176,65 @@ class NBCSportsIE(InfoExtractor):
             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 
 
+class NBCSportsStreamIE(AdobePassIE):
+    _VALID_URL = r'https?://stream\.nbcsports\.com/.+?\bpid=(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://stream.nbcsports.com/nbcsn/generic?pid=206559',
+        'info_dict': {
+            'id': '206559',
+            'ext': 'mp4',
+            'title': 'Amgen Tour of California Women\'s Recap',
+            'description': 'md5:66520066b3b5281ada7698d0ea2aa894',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'skip': 'Requires Adobe Pass Authentication',
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        live_source = self._download_json(
+            'http://stream.nbcsports.com/data/live_sources_%s.json' % video_id,
+            video_id)
+        video_source = live_source['videoSources'][0]
+        title = video_source['title']
+        source_url = None
+        for k in ('source', 'msl4source', 'iossource', 'hlsv4'):
+            sk = k + 'Url'
+            source_url = video_source.get(sk) or video_source.get(sk + 'Alt')
+            if source_url:
+                break
+        else:
+            source_url = video_source['ottStreamUrl']
+        is_live = video_source.get('type') == 'live' or video_source.get('status') == 'Live'
+        resource = self._get_mvpd_resource('nbcsports', title, video_id, '')
+        token = self._extract_mvpd_auth(url, video_id, 'nbcsports', resource)
+        tokenized_url = self._download_json(
+            'https://token.playmakerservices.com/cdn',
+            video_id, data=json.dumps({
+                'requestorId': 'nbcsports',
+                'pid': video_id,
+                'application': 'NBCSports',
+                'version': 'v1',
+                'platform': 'desktop',
+                'cdn': 'akamai',
+                'url': video_source['sourceUrl'],
+                'token': base64.b64encode(token.encode()).decode(),
+                'resourceId': base64.b64encode(resource.encode()).decode(),
+            }).encode())['tokenizedUrl']
+        formats = self._extract_m3u8_formats(tokenized_url, video_id, 'mp4')
+        self._sort_formats(formats)
+        return {
+            'id': video_id,
+            'title': self._live_title(title) if is_live else title,
+            'description': live_source.get('description'),
+            'formats': formats,
+            'is_live': is_live,
+        }
+
+
 class CSNNEIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?csnne\.com/video/(?P<id>[0-9a-z-]+)'
 

youtube_dl/extractor/peertube.py

@@ -0,0 +1,228 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    parse_resolution,
+    try_get,
+    unified_timestamp,
+    urljoin,
+)
+
+
+class PeerTubeIE(InfoExtractor):
+    _INSTANCES_RE = r'''(?:
+                            # Taken from https://instances.joinpeertube.org/instances
+                            tube\.openalgeria\.org|
+                            peertube\.pointsecu\.fr|
+                            peertube\.nogafa\.org|
+                            peertube\.pl|
+                            megatube\.lilomoino\.fr|
+                            peertube\.tamanoir\.foucry\.net|
+                            peertube\.inapurna\.org|
+                            peertube\.netzspielplatz\.de|
+                            video\.deadsuperhero\.com|
+                            peertube\.devosi\.org|
+                            peertube\.1312\.media|
+                            tube\.worldofhauru\.xyz|
+                            tube\.bootlicker\.party|
+                            skeptikon\.fr|
+                            peertube\.geekshell\.fr|
+                            tube\.opportunis\.me|
+                            peertube\.peshane\.net|
+                            video\.blueline\.mg|
+                            tube\.homecomputing\.fr|
+                            videos\.cloudfrancois\.fr|
+                            peertube\.viviers-fibre\.net|
+                            tube\.ouahpiti\.info|
+                            video\.tedomum\.net|
+                            video\.g3l\.org|
+                            fontube\.fr|
+                            peertube\.gaialabs\.ch|
+                            peertube\.extremely\.online|
+                            peertube\.public-infrastructure\.eu|
+                            tube\.kher\.nl|
+                            peertube\.qtg\.fr|
+                            tube\.22decembre\.eu|
+                            facegirl\.me|
+                            video\.migennes\.net|
+                            janny\.moe|
+                            tube\.p2p\.legal|
+                            video\.atlanti\.se|
+                            troll\.tv|
+                            peertube\.geekael\.fr|
+                            vid\.leotindall\.com|
+                            video\.anormallostpod\.ovh|
+                            p-tube\.h3z\.jp|
+                            tube\.darfweb\.eu|
+                            videos\.iut-orsay\.fr|
+                            peertube\.solidev\.net|
+                            videos\.symphonie-of-code\.fr|
+                            testtube\.ortg\.de|
+                            videos\.cemea\.org|
+                            peertube\.gwendalavir\.eu|
+                            video\.passageenseine\.fr|
+                            videos\.festivalparminous\.org|
+                            peertube\.touhoppai\.moe|
+                            peertube\.duckdns\.org|
+                            sikke\.fi|
+                            peertube\.mastodon\.host|
+                            firedragonvideos\.com|
+                            vidz\.dou\.bet|
+                            peertube\.koehn\.com|
+                            peer\.hostux\.social|
+                            share\.tube|
+                            peertube\.walkingmountains\.fr|
+                            medias\.libox\.fr|
+                            peertube\.moe|
+                            peertube\.xyz|
+                            jp\.peertube\.network|
+                            videos\.benpro\.fr|
+                            tube\.otter\.sh|
+                            peertube\.angristan\.xyz|
+                            peertube\.parleur\.net|
+                            peer\.ecutsa\.fr|
+                            peertube\.heraut\.eu|
+                            peertube\.tifox\.fr|
+                            peertube\.maly\.io|
+                            vod\.mochi\.academy|
+                            exode\.me|
+                            coste\.video|
+                            tube\.aquilenet\.fr|
+                            peertube\.gegeweb\.eu|
+                            framatube\.org|
+                            thinkerview\.video|
+                            tube\.conferences-gesticulees\.net|
+                            peertube\.datagueule\.tv|
+                            video\.lqdn\.fr|
+                            meilleurtube\.delire\.party|
+                            tube\.mochi\.academy|
+                            peertube\.dav\.li|
+                            media\.zat\.im|
+                            pytu\.be|
+                            peertube\.valvin\.fr|
+                            peertube\.nsa\.ovh|
+                            video\.colibris-outilslibres\.org|
+                            video\.hispagatos\.org|
+                            tube\.svnet\.fr|
+                            peertube\.video|
+                            videos\.lecygnenoir\.info|
+                            peertube3\.cpy\.re|
+                            peertube2\.cpy\.re|
+                            videos\.tcit\.fr|
+                            peertube\.cpy\.re
+                        )'''
+    _VALID_URL = r'''(?x)
+                    https?://
+                        %s
+                        /(?:videos/(?:watch|embed)|api/v\d/videos)/
+                        (?P<id>[^/?\#&]+)
+                    ''' % _INSTANCES_RE
+    _TESTS = [{
+        'url': 'https://peertube.moe/videos/watch/2790feb0-8120-4e63-9af3-c943c69f5e6c',
+        'md5': '80f24ff364cc9d333529506a263e7feb',
+        'info_dict': {
+            'id': '2790feb0-8120-4e63-9af3-c943c69f5e6c',
+            'ext': 'mp4',
+            'title': 'wow',
+            'description': 'wow such video, so gif',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+            'timestamp': 1519297480,
+            'upload_date': '20180222',
+            'uploader': 'Luclu7',
+            'uploader_id': '7fc42640-efdb-4505-a45d-a15b1a5496f1',
+            'uploder_url': 'https://peertube.nsa.ovh/accounts/luclu7',
+            'license': 'Unknown',
+            'duration': 3,
+            'view_count': int,
+            'like_count': int,
+            'dislike_count': int,
+            'tags': list,
+            'categories': list,
+        }
+    }, {
+        'url': 'https://peertube.tamanoir.foucry.net/videos/watch/0b04f13d-1e18-4f1d-814e-4979aa7c9c44',
+        'only_matching': True,
+    }, {
+        # nsfw
+        'url': 'https://tube.22decembre.eu/videos/watch/9bb88cd3-9959-46d9-9ab9-33d2bb704c39',
+        'only_matching': True,
+    }, {
+        'url': 'https://tube.22decembre.eu/videos/embed/fed67262-6edb-4d1c-833b-daa9085c71d7',
+        'only_matching': True,
+    }, {
+        'url': 'https://tube.openalgeria.org/api/v1/videos/c1875674-97d0-4c94-a058-3f7e64c962e8',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return [
+            mobj.group('url')
+            for mobj in re.finditer(
+                r'''(?x)<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//%s/videos/embed/[^/?\#&]+)\1'''
+                % PeerTubeIE._INSTANCES_RE, webpage)]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        video = self._download_json(
+            urljoin(url, '/api/v1/videos/%s' % video_id), video_id)
+
+        title = video['name']
+
+        formats = []
+        for file_ in video['files']:
+            if not isinstance(file_, dict):
+                continue
+            file_url = file_.get('fileUrl')
+            if not file_url or not isinstance(file_url, compat_str):
+                continue
+            file_size = int_or_none(file_.get('size'))
+            format_id = try_get(
+                file_, lambda x: x['resolution']['label'], compat_str)
+            f = parse_resolution(format_id)
+            f.update({
+                'url': file_url,
+                'format_id': format_id,
+                'filesize': file_size,
+            })
+            formats.append(f)
+        self._sort_formats(formats)
+
+        def account_data(field):
+            return try_get(video, lambda x: x['account'][field], compat_str)
+
+        category = try_get(video, lambda x: x['category']['label'], compat_str)
+        categories = [category] if category else None
+
+        nsfw = video.get('nsfw')
+        if nsfw is bool:
+            age_limit = 18 if nsfw else 0
+        else:
+            age_limit = None
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': video.get('description'),
+            'thumbnail': urljoin(url, video.get('thumbnailPath')),
+            'timestamp': unified_timestamp(video.get('publishedAt')),
+            'uploader': account_data('displayName'),
+            'uploader_id': account_data('uuid'),
+            'uploder_url': account_data('url'),
+            'license': try_get(
+                video, lambda x: x['licence']['label'], compat_str),
+            'duration': int_or_none(video.get('duration')),
+            'view_count': int_or_none(video.get('views')),
+            'like_count': int_or_none(video.get('likes')),
+            'dislike_count': int_or_none(video.get('dislikes')),
+            'age_limit': age_limit,
+            'tags': try_get(video, lambda x: x['tags'], list),
+            'categories': categories,
+            'formats': formats,
+        }

youtube_dl/extractor/viewlift.py

@@ -1,24 +1,27 @@
 from __future__ import unicode_literals
 
+import base64
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     ExtractorError,
     clean_html,
     determine_ext,
     int_or_none,
     js_to_json,
+    parse_age_limit,
     parse_duration,
 )
 
 
 class ViewLiftBaseIE(InfoExtractor):
-    _DOMAINS_REGEX = r'(?:snagfilms|snagxtreme|funnyforfree|kiddovid|winnersview|monumentalsportsnetwork|vayafilm)\.com|kesari\.tv'
+    _DOMAINS_REGEX = r'(?:snagfilms|snagxtreme|funnyforfree|kiddovid|winnersview|(?:monumental|lax)sportsnetwork|vayafilm)\.com|hoichoi\.tv'
 
 
 class ViewLiftEmbedIE(ViewLiftBaseIE):
-    _VALID_URL = r'https?://(?:(?:www|embed)\.)?(?:%s)/embed/player\?.*\bfilmId=(?P<id>[\da-f-]{36})' % ViewLiftBaseIE._DOMAINS_REGEX
+    _VALID_URL = r'https?://(?:(?:www|embed)\.)?(?:%s)/embed/player\?.*\bfilmId=(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})' % ViewLiftBaseIE._DOMAINS_REGEX
     _TESTS = [{
         'url': 'http://embed.snagfilms.com/embed/player?filmId=74849a00-85a9-11e1-9660-123139220831&w=500',
         'md5': '2924e9215c6eff7a55ed35b72276bd93',
@@ -60,8 +63,10 @@ class ViewLiftEmbedIE(ViewLiftBaseIE):
 
         formats = []
         has_bitrate = False
-        for source in self._parse_json(js_to_json(self._search_regex(
+        sources = self._parse_json(self._search_regex(
-                r'(?s)sources:\s*(\[.+?\]),', webpage, 'json')), video_id):
+            r'(?s)sources:\s*(\[.+?\]),', webpage,
+            'sources', default='[]'), video_id, js_to_json)
+        for source in sources:
             file_ = source.get('file')
             if not file_:
                 continue
@@ -70,7 +75,8 @@ class ViewLiftEmbedIE(ViewLiftBaseIE):
             format_id = source.get('label') or ext
             if all(v in ('m3u8', 'hls') for v in (type_, ext)):
                 formats.extend(self._extract_m3u8_formats(
-                    file_, video_id, 'mp4', m3u8_id='hls'))
+                    file_, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
             else:
                 bitrate = int_or_none(self._search_regex(
                     [r'(\d+)kbps', r'_\d{1,2}x\d{1,2}_(\d{3,})\.%s' % ext],
@@ -85,6 +91,13 @@ class ViewLiftEmbedIE(ViewLiftBaseIE):
                     'tbr': bitrate,
                     'height': height,
                 })
+        if not formats:
+            hls_url = self._parse_json(self._search_regex(
+                r'filmInfo\.src\s*=\s*({.+?});',
+                webpage, 'src'), video_id, js_to_json)['src']
+            formats = self._extract_m3u8_formats(
+                hls_url, video_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False)
         field_preference = None if has_bitrate else ('height', 'tbr', 'format_id')
         self._sort_formats(formats, field_preference)
 
@@ -109,10 +122,13 @@ class ViewLiftIE(ViewLiftBaseIE):
             'display_id': 'lost_for_life',
             'ext': 'mp4',
             'title': 'Lost for Life',
-            'description': 'md5:fbdacc8bb6b455e464aaf98bc02e1c82',
+            'description': 'md5:ea10b5a50405ae1f7b5269a6ec594102',
             'thumbnail': r're:^https?://.*\.jpg',
             'duration': 4489,
-            'categories': ['Documentary', 'Crime', 'Award Winning', 'Festivals']
+            'categories': 'mincount:3',
+            'age_limit': 14,
+            'upload_date': '20150421',
+            'timestamp': 1429656819,
         }
     }, {
         'url': 'http://www.snagfilms.com/show/the_world_cut_project/india',
@@ -125,7 +141,9 @@ class ViewLiftIE(ViewLiftBaseIE):
             'description': 'md5:5c168c5a8f4719c146aad2e0dfac6f5f',
             'thumbnail': r're:^https?://.*\.jpg',
             'duration': 979,
-            'categories': ['Documentary', 'Sports', 'Politics']
+            'categories': 'mincount:2',
+            'timestamp': 1399478279,
+            'upload_date': '20140507',
         }
     }, {
         # Film is not playable in your area.
@@ -138,9 +156,6 @@ class ViewLiftIE(ViewLiftBaseIE):
     }, {
         'url': 'http://www.winnersview.com/videos/the-good-son',
         'only_matching': True,
-    }, {
-        'url': 'http://www.kesari.tv/news/video/1461919076414',
-        'only_matching': True,
     }, {
         # Was once Kaltura embed
         'url': 'https://www.monumentalsportsnetwork.com/videos/john-carlson-postgame-2-25-15',
@@ -156,11 +171,62 @@ class ViewLiftIE(ViewLiftBaseIE):
             raise ExtractorError(
                 'Film %s is not available.' % display_id, expected=True)
 
+        initial_store_state = self._search_regex(
+            r"window\.initialStoreState\s*=.*?JSON\.parse\(unescape\(atob\('([^']+)'\)\)\)",
+            webpage, 'Initial Store State', default=None)
+        if initial_store_state:
+            modules = self._parse_json(compat_urllib_parse_unquote(base64.b64decode(
+                initial_store_state).decode()), display_id)['page']['data']['modules']
+            content_data = next(m['contentData'][0] for m in modules if m.get('moduleType') == 'VideoDetailModule')
+            gist = content_data['gist']
+            film_id = gist['id']
+            title = gist['title']
+            video_assets = content_data['streamingInfo']['videoAssets']
+
+            formats = []
+            mpeg_video_assets = video_assets.get('mpeg') or []
+            for video_asset in mpeg_video_assets:
+                video_asset_url = video_asset.get('url')
+                if not video_asset:
+                    continue
+                bitrate = int_or_none(video_asset.get('bitrate'))
+                height = int_or_none(self._search_regex(
+                    r'^_?(\d+)[pP]$', video_asset.get('renditionValue'),
+                    'height', default=None))
+                formats.append({
+                    'url': video_asset_url,
+                    'format_id': 'http%s' % ('-%d' % bitrate if bitrate else ''),
+                    'tbr': bitrate,
+                    'height': height,
+                    'vcodec': video_asset.get('codec'),
+                })
+
+            hls_url = video_assets.get('hls')
+            if hls_url:
+                formats.extend(self._extract_m3u8_formats(
+                    hls_url, film_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+            self._sort_formats(formats, ('height', 'tbr', 'format_id'))
+
+            info = {
+                'id': film_id,
+                'display_id': display_id,
+                'title': title,
+                'description': gist.get('description'),
+                'thumbnail': gist.get('videoImageUrl'),
+                'duration': int_or_none(gist.get('runtime')),
+                'age_limit': parse_age_limit(content_data.get('parentalRating')),
+                'timestamp': int_or_none(gist.get('publishDate'), 1000),
+                'formats': formats,
+            }
+            for k in ('categories', 'tags'):
+                info[k] = [v['title'] for v in content_data.get(k, []) if v.get('title')]
+            return info
+        else:
             film_id = self._search_regex(r'filmId=([\da-f-]{36})"', webpage, 'film id')
 
             snag = self._parse_json(
                 self._search_regex(
-                r'Snag\.page\.data\s*=\s*(\[.+?\]);', webpage, 'snag'),
+                    r'Snag\.page\.data\s*=\s*(\[.+?\]);', webpage, 'snag', default='[]'),
                 display_id)
 
             for item in snag:

youtube_dl/utils.py

@@ -2253,12 +2253,12 @@ US_RATINGS = {
 
 
 TV_PARENTAL_GUIDELINES = {
-    'TV-Y': 0,
+    'Y': 0,
-    'TV-Y7': 7,
+    'Y7': 7,
-    'TV-G': 0,
+    'G': 0,
-    'TV-PG': 0,
+    'PG': 0,
-    'TV-14': 14,
+    '14': 14,
-    'TV-MA': 17,
+    'MA': 17,
 }
 
 
@@ -2272,7 +2272,10 @@ def parse_age_limit(s):
         return int(m.group('age'))
     if s in US_RATINGS:
         return US_RATINGS[s]
-    return TV_PARENTAL_GUIDELINES.get(s)
+    m = re.match(r'^TV[_-]?(%s)$' % '|'.join(TV_PARENTAL_GUIDELINES.keys()), s)
+    if m:
+        return TV_PARENTAL_GUIDELINES[m.group(1)]
+    return None
 
 
 def strip_jsonp(code):